Search Results

Genius AI Detection is a computer-aided detection and diagnosis (CADe/CADx) software device intended to be used with compatible digital breast tomosynthesis (DBT) systems to identify and mark regions of interest including soft tissue densities (masses, architectural distortions and asymmetries) and calcifications in DBT exams from compatible DBT systems and provide confidence scores that offer assessment for Certainty of Findings and a Case Score.

The device intends to aid in the interpretation of digital breast tomosynthesis exams in a concurrent fashion, where the interpreting physician confirms or dismisses the findings during the reading of the exam.

Genius AI Detection 2.0 is a software device intended to identify potential abnormalities in breast tomosynthesis images. Genius AI Detection 2.0 analyzes each standard mammographic view in a digital breast tomosynthesis examination using deep learning networks. For each detected lesion, Genius AI Detection 2.0 produces CAD results that include:

• the location of the lesion;
• an outline of the lesion;
• a confidence score for the lesion
• Genius AI Detection 2.0 also produces a case score for the entire breast tomosynthesis exam.

Genius AI Detection 2.0 packages all CAD findings derived from the corresponding analysis of a tomosynthesis exam into a DICOM Mammography CAD SR object and distributes it for display on DICOM compliant review workstations. The interpreting physician will have access to the CAD findings concurrently to the reading of the tomosynthesis exam. In addition, a combination of peripheral information such as number of marks and case scores may be used on the review workstation to enhance the interpreting physician's workflow by offering a better organization of the patient worklist.

Here's a breakdown of the acceptance criteria and study details for Genius AI Detection 2.0, based on the provided FDA 510(k) clearance letter:

Acceptance Criteria and Device Performance for Genius AI Detection 2.0

1. Table of Acceptance Criteria and Reported Device Performance

The provided document describes a non-inferiority study to demonstrate that the performance of Genius AI Detection 2.0 on Envision (ENV) images is equivalent to its performance on the predicate's Standard of Care (SOC) images (Hologic's Selenia Dimensions systems). The primary acceptance criterion was non-inferiority of the Area Under the Curve (AUC) of the ROC curve, with a 5% margin. Secondary metrics included sensitivity, specificity, and false marker rate per view.

(Note: The document focuses on demonstrating equivalence to the predicate's performance on a new platform rather than absolute performance against a fixed threshold for all metrics, except for the implant case where specific CIs are given and deemed acceptable.)

2. Sample Size Used for the Test Set and Data Provenance

• Sample Size (Main Comparison Study): 1475 subjects
  • 200 biopsy-proven cancer subjects
  • 275 biopsy-proven benign subjects
  • 78 BI-RADS 3 subjects (considered BI-RADS 1 or 2 upon diagnostic workup)
  • 922 BI-RADS 1 and 2 subjects (at screening)
  • Implant Case Test Set: 480 subjects
    • 132 biopsy-proven cancer subjects
    • 348 negative subjects (119 biopsy-proven benign, 229 screening negative)
• Data Provenance:
  • Country of Origin: Not explicitly stated, but collected from a "national multi-center breast imaging network" within the U.S., implying U.S. origin.
  • Retrospective or Prospective: The main comparison study data was collected for evaluating the safety and effectiveness of the Envision platform, with an IRB approved protocol. This suggests a retrospective study design, where existing images were gathered for evaluation. The implant cases were collected between 2015 and 2022, also indicating a retrospective approach.

3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications

• Number of Experts: Two
• Qualifications: Both were MQSA-certified radiologists with over 20 years of experience.

4. Adjudication Method for the Test Set

The document explicitly states that the "ground truthing to evaluate performance metrics including the locations of cancer lesions was done by two MQSA-certified radiologists with over 20 years of experience."

• Adjudication Method: It does not specify a particular adjudication method (e.g., 2+1, 3+1). It simply states that ground truthing was done by two experts. This implies either consensus was reached between the two, or potentially an unstated arbitration method if they disagreed, or that their individual findings were used for analysis. Given the phrasing, expert consensus is the most likely implied method, but not explicitly detailed.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study was done

• No, an MRMC comparative effectiveness study was NOT done. The study described is a standalone performance comparison of the AI algorithm on images from different modalities (Envision vs. Standard of Care), not a study involving human readers with and without AI assistance to measure effect size.

6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) was done

• Yes, a standalone study WAS done. The document explicitly states, "A standalone study was conducted to compare the detection performance of FDA cleared Genius AI Detection 2.0 (K221449) using Standard of Care (SOC) images acquired on the Dimensions systems against images acquired on the FDA approved Envision Mammography Platform (P080003/S009)." This study evaluated the algorithm's performance (fROC, ROC, sensitivity, specificity, false marker rate) directly against the ground truth without human intervention.

7. The Type of Ground Truth Used

• Ground Truth Type: A combination of biopsy-proven cancer and biopsy-proven benign cases, along with BI-RADS diagnostic outcomes (for negative cases). For the cancer cases, the "locations of cancer lesions" were part of the ground truth.

8. The Sample Size for the Training Set

• Not provided. The document states that the test dataset was "sequestered from any training datasets by isolating it on a secured server with controlled access permissions" and that the data for implant cases was "sequestered from the training datasets for Genius AI Detection." However, the actual sample size of the training set is not mentioned.

9. How the Ground Truth for the Training Set Was Established

• Not provided. Since the training set sample size and details are not disclosed, the method for establishing its ground truth is also not mentioned in this document. It is generally assumed that similar rigorous methods (e.g., biopsy-proven truth, expert review) would have been used for training data, but this specific filing does not detail it.

Ask a specific question about this device

The Aptima® SARS-CoV-2 Assay is a nucleic acid amplification in vitro diagnostic test intended for the qualitative detection of RNA from severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) isolated and purified from nasopharyngeal (NP) swab and anterior nasal (AN) swab specimens obtained from patients with signs and symptoms of COVID-19.

Positive results are indicative of the presence of SARS-CoV-2 RNA. The Aptima SARS-CoV-2 Assay is intended for use as an aid in the diagnosis of COVID-19 if used in conjunction with other clinical, epidemiological, and laboratory findings. Clinical correlation with patient history and other diagnostic information is necessary to determine patient infection status. Positive results do not rule out bacterial infection or co-infection with other viruses.

Negative results do not preclude SARS-CoV-2 infection and should not be used as the sole basis for patient management decisions. Negative results must be combined with clinical observations, patient history, and epidemiological information.

The Aptima SARS-CoV-2 Assay is a nucleic acid amplification in vitro diagnostic test developed for use on the fully automated Panther/Panther Fusion system to detect RNA from SARS-CoV-2 isolated and purified from nasopharyngeal and anterior nasal swab specimens collected into UTM/VTM or with the RespDirect Collection Kit.

The Aptima SARS-CoV-2 Assay combines the technologies of target capture, Transcription Mediated Amplification (TMA), and Dual Kinetic Assay (DKA).

Specimens are collected and transferred into their respective specimen transport tubes. The transport solutions in these tubes release the RNA target and protect them from degradation during storage. When the Aptima SARS-CoV-2 Assay is performed in the laboratory, the target RNA molecules are isolated from specimens by use of capture oligomers via target capture that utilizes magnetic microparticles. The capture oligomers contain sequences complementary to specific regions of the target molecules as well as a string of deoxyadenosine residues. A separate capture oligomer is used for each target. During the hybridization step, the sequence specific regions of the capture oligomers bind to specific regions of the target molecules. The capture oligomer:target complex is then captured out of solution by decreasing the temperature of the reaction to room temperature. This temperature reduction allows hybridization to occur between the deoxyadenosine region on the capture oligomer and the poly-deoxythymidine molecules that are covalently attached to the magnetic particles. The microparticles, including the captured target molecules bound to them, are pulled to the side of the reaction vessel using magnets and the supernatant is aspirated. The particles are washed to remove residual specimen matrix that may contain amplification reaction inhibitors. After the target capture steps are completed, the specimens are ready for amplification.

Target amplification assays are based on the ability of complementary oligonucleotide primers to specifically anneal and allow enzymatic amplification of the target nucleic acid strands. The Aptima SARS-CoV-2 Assay replicates specific regions of the RNA from SARS-CoV-2 virus. Detection of the RNA amplification product sequences (amplicon) is achieved using nucleic acid hybridization. Single-stranded chemiluminescent nucleic acid probes, which are unique and complementary to a region of each target amplicon and Internal Control (IC) amplicon, are labeled with different acridinium ester (AE) molecules. The AE-labeled probes combine with the amplicon to form stable hybrids. The Selection Reagent differentiates hybridized from unhybridized probe, eliminating the generation of signal from the unhybridized probe. During the detection step, light emitted from the labeled hybrids is measured as photon signals in a luminometer and are reported as Relative Light Units (RLU). In DKA, differences in the kinetic profiles of the labeled probes allow for the differentiation of signal; kinetic profiles are derived from measurements of photon output during the detection read time.

The Aptima SARS-CoV-2 Assay amplifies and detects 2 conserved regions of the ORF1ab gene in the same reaction, using the "glower" kinetic type. The 2 regions are not differentiated and amplification of either or both regions lead to RLU signal. The assay results are determined by a cut-off based on the total RLU and the kinetic curve type.

Here's a breakdown of the acceptance criteria and the study that proves the device meets them, based on the provided text:

Acceptance Criteria and Reported Device Performance

The acceptance criteria are implicitly defined by the performance metrics presented as evidence of substantial equivalence to the predicate device. The key metrics are Positive Percent Agreement (PPA) and Negative Percent Agreement (NPA). While explicit numerical acceptance thresholds are not provided in the document, the reported performance demonstrates "comparable" results to the predicate device.

Study Details:

1. Sample Sizes Used for the Test Set and Data Provenance:

   • Clinical Study 1 (NP Swab Specimens):
     • Total Enrolled: 1646 specimens
     • Evaluable (Final Data Set): 1495 NP swab specimens (1195 fresh, 300 frozen)
     • Provenance: Prospective multicenter study from four participating US pediatric/adolescent, private and/or university hospitals. Specimens collected between June-July 2020 and January-April 2023. Remnant specimens.
   • Clinical Study 2 (Anterior Nasal Swab Specimens):
     • Total Enrolled Subjects: 2301
     • Evaluable Subjects: 2177 individuals (1159 with evaluable anterior nasal swab specimens in UTM/VTM, and 1018 with evaluable nasal swab specimens in eSTM).
     • Provenance: Prospective, multicenter clinical study at nine geographically and ethnically diverse US sites during the 2022-2023 respiratory season.

2. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications:

   • The document does not specify the number or qualifications of experts directly establishing the clinical ground truth. Instead, it relies on a Composite Comparator Algorithm (CCA). The CCA consists of "two highly sensitive US FDA EUA SARS-CoV-2 molecular tests" and a "validated PCR followed by bi-directional sequencing (PCR/BDS) assay." This implies that the 'expertise' comes from the validation and regulatory clearance of these comparator assays, rather than individual human experts adjudicating each case.

3. Adjudication Method for the Test Set:

   • Composite Comparator Algorithm (CCA): "A final CCA result was assigned when two of the three comparator assay results were in concordance." This serves as the adjudication method for determining the true positive/negative status of the clinical samples.

4. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done:

   • No. This type of study (MRMC) is typically performed for diagnostic imaging devices where human readers interpret images with and without AI assistance. The Aptima SARS-CoV-2 Assay is an in vitro diagnostic (IVD), a laboratory test that detects nucleic acids, not an imaging device requiring human interpretation of visual data. Therefore, an MRMC study is not applicable here.

5. If a Standalone (i.e., algorithm only without human-in-the-loop performance) Was Done:

   • Yes, this is effectively a standalone performance study. The Aptima SARS-CoV-2 Assay is an automated nucleic acid amplification test run on the Panther/Panther Fusion system. Its performance is directly compared to the CCA, without human intervention in the interpretation of the device's results.

6. The Type of Ground Truth Used:

   • Clinical Studies: Composite Comparator Algorithm (CCA) based on the concordance of results from two highly sensitive FDA EUA SARS-CoV-2 molecular tests and a validated PCR/BDS assay. This is a form of reference standard derived from established and highly sensitive laboratory methods.
   • Analytical Studies (e.g., LoD, Reactivity, Specificity): The ground truth was established by known concentrations of SARS-CoV-2 virus strains or other microorganisms/interfering substances, prepared in controlled laboratory settings (e.g., "spiked with inactivated cultured SARS-CoV-2 virus").

7. The Sample Size for the Training Set:

   • The document describes performance evaluation studies (analytical and clinical) for market clearance. It does not provide information on the specific training set size used for the development or training of the assay's internal algorithms (e.g., for the kinetic curve analysis or cut-off determination). This information would typically be part of the assay development and validation, not necessarily detailed in a 510(k) summary unless it significantly changed or impacted performance during the clearance process for the specific assay rather than the underlying platform. The focus here is on the performance of the final, already "trained" device.

8. How the Ground Truth for the Training Set Was Established:

   • As noted above, details regarding the training set's ground truth establishment are not provided in this 510(k) summary. For IVD devices, ground truth for training internal algorithms typically involves using characterized positive and negative clinical samples, spiked samples with known viral loads, and potentially synthetic data, all carefully confirmed by highly sensitive reference methods or gold standard assays during the R&D phase of the product.

Ask a specific question about this device

The Aptima® CMV Quant Assay is an in vitro nucleic acid amplification test for the quantitation of human cytomegalovirus (CMV) DNA in human EDTA plasma on the fully automated Panther® system.

The Aptima CMV Quant Assay is intended for use to aid in the management of solid-organ transplant patients and hematopoietic stem cell transplant patients receiving anti-CMV therapy, serial DNA measurements can be used to assess viral response to treatment. The results from Aptima CMV Quant assay must be interpreted within the context of all relevant clinical and laboratory findings.

The Aptima CMV Quant Assay is not intended for use as a screening assay for the presence of CMV in blood or blood products.

The Aptima CMV Quant Assay is an in vitro nucleic acid amplification test that uses real-time transcription mediated amplification (TMA) technology on the Panther Fusion system to quantify CMV DNA, genotypes 1, 2, 3, and 4. The primer design targets the highly conserved UL56 gene to ensure accurate quantitation of the CMV DNA. The assay is standardized to the WHO International Standard for human cytomegalovirus. The Aptima CMV Quant Assay is intended for use to aid in the management of solid-organ transplant patients and hematopoietic stem cell transplant patients.

The addition of a protocol to treat specific plasma samples with proteinase K in the assay package insert will allow the retesting of those samples to obtain a valid result.

The provided text describes a 510(k) premarket notification for the "Aptima CMV Quant Assay" and specifically details an update to the assay's package insert to include a protocol for retesting plasma samples that initially yield an "ML2 flag" invalid result. This retesting involves a manual pretreatment with proteinase K. The information provided is for a diagnostic assay, not an AI/ML-based device. Therefore, many of the requested elements pertaining to AI/ML development, such as training sets, ground truth establishment for AI/ML, expert consensus for imaging, MRMC studies, and effect sizes of AI assistance, are not applicable.

However, I can extract information related to the acceptance criteria and the study that proves the device (specifically, the proteinase K pretreatment protocol) meets the acceptance criteria within the context of a diagnostic assay's performance and validation.

Acceptance Criteria and Study for the Proteinase K Pretreatment Protocol

The primary acceptance criterion for this change is to demonstrate that the proteinase K pretreatment protocol effectively resolves the "ML2 flag" invalid results and does not negatively impact the accuracy of the CMV quantification.

1. Table of Acceptance Criteria and the Reported Device Performance:

2. Sample Size Used for the Test Set and the Data Provenance:

• Test Set Sample Sizes:
  • Without Proteinase K: 1039 plasma specimens
  • With Proteinase K: 4098 plasma specimens
• Data Provenance: The plasma specimens were collected from the transplant patient population by the National Institute of Health (NIH). This suggests a retrospective analysis of archived samples or a prospective study conducted at the NIH. The location is the USA.

3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications:

• Not applicable as this is a quantitative diagnostic assay for DNA detection, not an image-based AI/ML device requiring expert interpretation of images for ground truth. The "ground truth" here is the actual CMV DNA concentration, which the assay is designed to measure.

4. Adjudication Method for the Test Set:

• Not applicable for a quantitative diagnostic assay. The performance is assessed against the ability to produce a valid quantitative result for CMV DNA and maintain accuracy, not against human consensus.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study was done:

• No, this is not an MRMC study. MRMC studies are typically performed for visual diagnostic aids (e.g., medical imaging AI) to compare human reader performance with and without AI assistance. This device is a direct quantitative assay.

6. If a Standalone (i.e. algorithm only without human-in-the loop performance) was done:

• Yes, in a sense. The study validates the protocol (which includes a manual pretreatment step) and its effect on the automated assay system's ability to produce a valid result. The "performance" refers to the assay's output after the sample preparation, not an AI algorithm's independent decision. The core assay itself operates in a "standalone" fashion once the sample is loaded onto the Panther system.

7. The type of ground truth used:

• The implicit ground truth is the actual CMV DNA concentration in patient samples (or spiked controls used in accuracy assessments), as determined by a reference method or known concentrations. The study's focus, however, is on resolving an assay technical issue (ML2 flag) and demonstrating that the resolution method maintains the accuracy of CMV quantification. The reference to the NIH demonstrating that "using proteinase K for pretreatment of specimens did not impact the accuracy of CMV quantification" confirms that accurate quantitation (the ground truth outcome) was assessed.

8. The Sample Size for the Training Set:

• Not applicable. This is not an AI/ML device that requires a training set in the conventional sense. The "training" for such an assay would be its initial development and optimization, which isn't described in terms of a specific "training set" size for this modification. The data provided refers to validation/test sets for the new protocol.

9. How the ground truth for the training set was established:

• Not applicable, as it's not an AI/ML device with a training set for model development.

Ask a specific question about this device

Aptima BV Assay: The Aptima BV Assay is an in vitro nucleic acid amplification test that utilizes real time Transcription-Mediated Amplification (TMA) technology for detection and quantitation of ribosomal RNA from bacterial vaginosis (BV), including Lactobacillus (L. gasseri, L. crispatus, and L. jenseni), Gardnerella vaginalis), and Atopobium vaginae (A. vaginae). The assay reports a qualitative result for BV and does not report results for individual organisms. The assay is intended to aid in the diagnosis of BV on the automated Panther System using clinician-collected and patient-collected vaginal swab specimens from females with a clinical presentation consistent with vaginitis and/or vaginosis.

Aptima CV/TV Assay: The Aptima CV/TV Assay is an in vitro nucleic acid amplification of RNA from microorganisms associated with vulvovaginal candidiasis and trichomoniasis. The assay utilizes real time Transcription-Mediated Amplification (TMA) technology to detect and qualitatively report results for the following organisms: Candida species group (C. albicans, C. tropicalis, C. parapsilosis, C. dubliniensis), Candida glabrata (C. glabrata), Trichomonas vaginalis (TV). The assay differentiates between C. glabrata and the Candida species group (C spp) by targeting the RNA component of RNAse P ribonucleoprotein; the assay does not differentiate among C spp. For TV, the assay targets ribosomal RNA (rRNA) and differentiates the results for C. glabrata and C spp. The assy is intended to aid in the diagnosis of vulvovaginal candidiasis and trichomoniasis on the automated Panther System using clinician-collected and patientcollected vaginal swab specimens from females with a clinical presentation consistent with vagintis.

Aptima BV Assay: Like the Aptima BV assay 100 test kit, the Aptima BV assay 250 test kit is an in vitro nucleic acid amplification test for the detection and quantitation of rRNA from bacteria associated with bacterial vaginosis in women with a clinical presentation consistent with vaginitis/vaginosis. The Aptima BV assay utilizes the automated Panther system to provide qualitative results to aid in the diagnosis of bacterial vaginosis.

Aptima CV/TV Assay: Like the Aptima CV/TV assay 100 test kit, the Aptima CV/TV assay 250 test kit is an in vitro nucleic acid amplification test for the detection and quantitation of RNA from microorganisms associated with vaginitis, trichomoniasis, and vulvovaginitis, in women with a clinical presentation consistent with vaginitis, trichomoniasis, and vulvovaginitis. The Aptima CV/TV assay utilizes the automated Panther system to provide qualitative results to aid in the diagnosis of vulvovaginal candidiasis and trichomoniasis.

The provided document, a 510(k) summary for the Hologic Aptima BV Assay and Aptima CV/TV Assay, describes the device's technical specifications and a general statement regarding performance data for substantial equivalence. However, it does not contain the detailed acceptance criteria and a comprehensive study report with specific performance metrics (like sensitivity, specificity, PPV, NPV against a clinical gold standard) that would typically be expected for demonstrating a device meets acceptance criteria in a clinical validation context.

Specifically, the document focuses on demonstrating substantial equivalence of a new kit configuration (250 tests) to an already cleared kit configuration (100 tests) by showing similar analytical performance, particularly Limit of Detection (LoD). It does not present a de novo clinical study with pre-defined acceptance criteria for diagnostic accuracy against a true clinical ground truth.

Therefore, many of the requested items (e.g., number of experts, adjudication method, MRMC studies, training set details) are not applicable or not provided in this specific type of submission, which is for a modification to an already cleared device, not a novel device requiring full clinical validation from scratch.

Here's an attempt to answer the questions based only on the provided text:

1. A table of acceptance criteria and the reported device performance

The document does not explicitly present a table of acceptance criteria and reported device performance in terms of diagnostic accuracy (e.g., sensitivity, specificity) against a clinical reference for the 250-test kit. Instead, it asserts equivalence by confirming L.O.D. for the new kit configuration.

The primary acceptance criteria for the new 250-test kit configuration, as implied by the performance data section, is that it must meet the established Limit of Detection (LoD) of the previously cleared 100-test kit configuration.

Aptima BV Assay (250 Test Kit):

Aptima CV/TV Assay (250 Test Kit):

The conclusion states: "The performance study results demonstrate that the Aptima BV assay 250 Test Kit on the Panther system performs comparably to the predicate device that is currently marketed for the same intended use." and "The Aptima BV 100 Test Kit LoD was confirmed in the Aptima BV 250 Test Kit configuration." (Similar statements are made for the CV/TV assay).

2. Sample sized used for the test set and the data provenance (e.g. country of origin of the data, retrospective or prospective)

• Sample Size: For the LoD confirmation study for both assays, at least 20 replicates per concentration, per reagent lot, using three lots were tested. This totals to at least 60 replicates per strain (for BV) or per organism (for CV/TV).
• Data Provenance: The document does not specify the country of origin of the data or whether the study was retrospective or prospective. It describes analytical sensitivity studies using prepared dilutions of cell lysates/suspensions. For the CV/TV assay, the use of "Natural Vaginal Swab Matrix (NVSM)" and "Simulated Vaginal Swab Matrix (SVSM)" is mentioned for dilutions.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts (e.g. radiologist with 10 years of experience)

This information is not provided and is not applicable to the type of analytical study performed. The LoD confirmation uses known concentrations of target organisms, not clinical samples requiring expert interpretation for ground truth.

4. Adjudication method (e.g. 2+1, 3+1, none) for the test set

This information is not provided and is not applicable to an analytical LoD confirmation study.

5. If a multi reader multi case (MRMC) comparative effectiveness study was done, If so, what was the effect size of how much human readers improve with AI vs without AI assistance

This information is not provided and is not applicable. The device is a diagnostic assay (in vitro nucleic acid amplification test), not an AI-assisted imaging device that would involve human readers.

6. If a standalone (i.e. algorithm only without human-in-the loop performance) was done

The device itself is a standalone diagnostic assay (an in vitro nucleic acid amplification test) run on an automated system (Panther system). Its performance (LoD) was confirmed without human interpretation of raw signals, as the Panther system software interprets the amplification signal emergence times to generate results.

7. The type of ground truth used (expert consensus, pathology, outcomes data, etc)

The ground truth used for this specific study (LoD confirmation) was known concentrations of purified target organisms/cell lysates. This is an analytical ground truth, not a clinical ground truth derived from expert consensus, pathology, or outcomes data.

8. The sample size for the training set

This information is not provided and is not applicable. This is not an AI/ML device that requires a training set in the conventional sense. The "training" for the assay involves internal optimization and validation during development, but the document does not detail these earlier stages.

9. How the ground truth for the training set was established

This information is not provided and is not applicable, as it's not an AI/ML device with a distinct training set and ground truth establishment methodology in the context of this submission.

Ask a specific question about this device

The Panther Fusion® SARS-CoV-2/Flu A/B/RSV assay is a fully automated multiplexed real-time polymerase chain reaction (RT-PCR) in vitro diagnostic test intended for the qualitative detection and differentiation of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), influenza A virus (Flu A), influenza B virus (Flu B), and respiratory syncytial virus (RSV). Nucleic acids are isolated and purified from nasopharyngeal (NP) swab specimens and anterior nasal (AN) swab specimens obtained from individuals exhibiting signs and symptoms of a respiratory tract infection. Clinical signs and symptoms of respiratory viral infection due to SARS-CoV-2, influenza, and RSV can be similar. This assay is intended to aid in the differential diagnosis of SARS-CoV-2, Flu A, Flu B, and RSV infections in humans and is not intended to detect influenza C virus infections.

Nucleic acids from the viral organisms identified by this test are generally detectable in NP and AN swab specimens during the acute phase of infection and identification of specific viral nucleic acids from individuals exhibiting signs and symptoms of respiratory tract infective of the presence of the identified virus and aids in diagnosis if used in conjunction with other clinical and epidemiological information, and laboratory findings. The results of this test should not be used as the sole basis for diagnosis, treatment, or other patient management decisions.

Positive results do not rule out coinfection with other organism(s) detected by the Panther Fusion SARS-CoV-2/Flu A/B/RSV assay may not be the definite cause of disease. Negative results do not preclude SARS-CoV-2, influenza A virus, influenza B virus, or RSV infections. This assay is designed for use on the Panther Fusion system.

The Panther Fusion SARS-CoV-2/Flu A/B/RSV assay is classified as a Class II in vitro diagnostic device per 21 CFR 866.3981 and has product code QOF. The Panther Fusion SARS-CoV-2/Flu A/B/RSV assay is designed for use on the fully automated Panther Fusion System.

The Panther Fusion system is a class II exempt device under 21CFR 862.2570 that has product code OOI.

The Panther Fusion SARS-CoV-2/Flu A/B/RSV assay is a multiplex real-time reverse transcriptase PCR (RT-PCR) in vitro diagnostic test developed for use on the fully automated Panther Fusion system to detect and differentiate SARS-CoV-2, influenza A, influenza B, and respiratory syncytial virus (RSV) directly from nasopharyngeal and anterior nasal swab specimens, from individuals exhibiting signs and symptoms of a respiratory tract infection.

The Panther Fusion SARS-CoV-2/Flu A/B/RSV assay involves the following steps: sample lysis, nucleic acid capture and elution transfer, and multiplex RT-PCR where analytes (when present) are simultaneously amplified, detected and differentiated. Nucleic acid capture and elution takes place in a single tube on the Panther Fusion system. The eluate is transferred to the Panther Fusion system reaction tube containing the assay reagents. Multiplex RT-PCR is then performed for the eluted nucleic acid on the Panther Fusion system.

Sample lysis, nucleic acid capture, and elution: Prior to processing and testing on the Panther Fusion system, specimens are transferred to a Specimen Lysis Tube containing specimen transport media (STM). Alternatively, samples can be collected with the RespDirect Collection kit which contains enhanced specimen transport media (eSTM). STM and eSTM lyse the cells, release target nucleic acid and protect them from degradation during storage. The Internal Control-S (IC-S) is added to each test specimen and controls via the working Panther Fusion Capture Reagent-S (wFCR-S). The IC-S is the reagent used to monitor specimen processing, amplification and detection. Magnetic particles with covalently bound oligonucleotides mediate the nucleic acid capture. Capture oligonucleotides hybridize to nucleic acid in the test specimen. Hybridized nucleic acid is then separated from the lysed specimen in a magnetic field. Wash steps remove extraneous components from the reaction tube. The elution step elutes purified nucleic acid. During the nucleic acid capture and elution step, total nucleic acid is isolated from specimens.

Elution transfer and RT-PCR: During the elution transfer step, eluted nucleic acid is transferred to a Panther Fusion tube already containing oil and reconstituted mastermix. Target amplification occurs via RT-PCR. A reverse transcriptase generates a DNA copy of the target sequence. Target specific forward and reverse primers and probes then amplify targets while simultaneously detecting and discriminating multiple target types via multiplex RT-PCR. The Panther Fusion system compares the fluorescence signal to a predetermined cut-off to produce a qualitative result for the presence or absence of the analyte. The positive result for each analyte will be accompanied by the cycle threshold (Ct value).

Here's a breakdown of the acceptance criteria and the study details for the Hologic Panther Fusion SARS-CoV-2/Flu A/B/RSV assay, based on the provided document:

Acceptance Criteria and Device Performance

The document describes the performance of the Panther Fusion SARS-CoV-2/Flu A/B/RSV assay in detecting SARS-CoV-2, Flu A, Flu B, and RSV in prospectively collected Anterior Nasal (AN) swab specimens in RespDirect eSTM.

Table of Acceptance Criteria (Implicit) and Reported Device Performance

While explicit acceptance criteria (e.g., minimum PPA/NPA values) are not stated, the document presents the observed Positive Percent Agreement (PPA) and Negative Percent Agreement (NPA) with their 95% Confidence Intervals. These values are implicitly the performance targets for the assay.

Study Information

2. Sample Size Used for the Test Set and Data Provenance

• Prospective Study:
  • Sample Size: 1021 evaluable anterior nasal (AN) swab specimens.
  • Data Provenance: Prospective, multicenter study conducted in the US between January 2023 and May 2023 at nine participating medical facilities.
• Enrichment Study (Supplemental Clinical Data for Low Prevalence Analytes):
  • Sample Size: 205 evaluable anterior nasal (AN) swab specimens.
  • Data Provenance: Enrichment phase of the study, conducted in the US between October 2023 and February 2024 at six participating medical facilities. The specimens were from individuals with a positive standard of care (SOC) test result for Flu A, Flu B, and/or RSV. All were frozen prior to testing.

3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications of Those Experts

The document does not specify the number or qualifications of experts used to establish the ground truth. Instead, it refers to:

• For SARS-CoV-2: A composite comparator algorithm (CCA) consisting of up to three highly sensitive US FDA EUA SARS-CoV-2 molecular tests.
• For Flu A, Flu B, and RSV: A US FDA-cleared molecular Flu A/B/RSV assay (in the prospective study) and a U.S. FDA-cleared molecular Flu A/B/RSV assay (in the enrichment study).
• For the enrichment study, initial selection was based on a positive standard of care (SOC) test result for Flu A, Flu B, and/or RSV.

4. Adjudication Method for the Test Set

• For SARS-CoV-2 in the prospective study: A Composite Comparator Algorithm (CCA) was used. A final CCA result was assigned when two of the three composite comparator assays were in concordance.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done

No, an MRMC comparative effectiveness study involving human readers and AI assistance was not done. This study is for an in vitro diagnostic (IVD) assay (a molecular test), not an AI-powered diagnostic imaging device.

6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) Was Done

Yes, this study represents a standalone performance evaluation of the Panther Fusion SARS-CoV-2/Flu A/B/RSV assay. The assay itself is an automated molecular diagnostic test on the Panther Fusion system, and its performance was evaluated against comparator methods (other molecular tests or algorithms), not against human interpretation of its outputs in a clinical setting for diagnostic aid.

7. The Type of Ground Truth Used

• SARS-CoV-2: Composite comparator algorithm (CCA) based on multiple FDA EUA molecular tests.
• Flu A/B/RSV: US FDA-cleared molecular Flu A/B/RSV assay.
• Enrichment Study (Flu A/B/RSV): US FDA-cleared molecular Flu A/B/RSV assay, and initial selection based on positive Standard of Care (SOC) test results.

This represents a reference standard method ground truth, where another established diagnostic test is used as the benchmark.

8. The Sample Size for the Training Set

The document does not provide information on a specific "training set" sample size for algorithm development. The Panther Fusion SARS-CoV-2/Flu A/B/RSV assay is a real-time RT-PCR in vitro diagnostic test, not a machine learning or AI algorithm in the context of typical training/validation splits. Its development would involve analytical studies and design verification, but not typically a "training set" in the same sense as an AI model. The provided clinical data are for performance evaluation.

9. How the Ground Truth for the Training Set Was Established

As noted in point 8, the concept of a "training set" with established ground truth in the context of AI models is not directly applicable here. The assay's performance characteristics (e.g., limit of detection, analytical reactivity) are established through various analytical studies (bench studies), not through a "training set" with a clinical ground truth. The clinical studies described are for validation and performance assessment, not for training an algorithm.

Ask a specific question about this device

The Fluent Pro Fluid Management System is intended to provide liquid distension of the uterus during diagnostic and operative hysteroscopy, and to monitor the volume differential between the irrigation liquid flowing into and out of the uterus while providing drive, control and suction for hysteroscopic morcellators.

The Fluent Pro Fluid Management System is a hysteroscopic fluid management system and drive console for hysteroscopic morcellators. The Fluent Pro Fluid Management System consists of a console and single-use procedure kit. The single-use procedure kit consists of sterile inflow (In-FloPak) and outflow (Out-FloPak) tube sets, and a non-sterile waste bag. The FloPaks and waste bag connect for use in hysteroscopic procedures. The console includes motors that control fluid inflow and outflow for hysteroscopic insufflation, as well as a pressure sensor monitoring system and graphical user interface. In addition, the console includes a connection for MyoSure Tissue Removal Devices and a pneumatic foot pedal to perform tissue removal procedures. The Fluent Pro Fluid Management System is compatible with currently marketed MyoSure Tissue Removal Devices.

This FDA 510(k) summary describes a medical device, the Fluent Pro Fluid Management System, for which substantial equivalence is claimed against a predicate device. The document focuses on regulatory compliance and device characteristics rather than a clinical study evaluating diagnostic or treatment efficacy.

Therefore, the requested information regarding acceptance criteria, reported device performance, sample sizes for test/training sets, data provenance, expert ground truth, adjudication methods, multi-reader multi-case studies, and standalone algorithm performance cannot be fully extracted or accurately described from the provided text. These types of details are typically found in clinical study reports, which are not included here.

However, based on the provided text, I can infer and extract some information related to performance testing and acceptance criteria as described in the context of device safety and functional equivalence:

1. Table of Acceptance Criteria and Reported Device Performance

The acceptance criteria here are functional and safety performance, demonstrated through various engineering and bench tests, rather than clinical efficacy metrics.

2. Sample size used for the test set and the data provenance

The document does not specify exact sample sizes for the various performance tests (e.g., number of units tested for electrical safety, number of sterility tests etc.). The provenance of the "data" is from internal company testing and validation activities as part of the 510(k) submission process. These are all prospective tests conducted on the device.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts

Not applicable in the context of this 510(k) submission. The "ground truth" for these engineering and performance tests is defined by adherence to international standards and design specifications, not expert consensus on medical images or patient outcomes.

4. Adjudication method

Not applicable. Testing involves engineering verification and validation against objective standards and specifications, not subjective expert adjudication.

5. Multi-reader multi-case (MRMC) comparative effectiveness study

No, an MRMC comparative effectiveness study was not done or reported in this 510(k) summary. This type of study is relevant for diagnostic devices where human readers interpret medical data (e.g., images) with and without AI assistance to assess improved accuracy or efficiency. The Fluent Pro Fluid Management System is a hysteroscopic insufflator and morcellator control system; its functions are mechanical and fluidic control, not diagnostic interpretation.

6. Standalone (i.e. algorithm only without human-in-the loop performance) was done

This refers to an AI algorithm's performance without human interaction. This is not applicable to the Fluent Pro Fluid Management System, as it is a physical medical device with software components that control mechanical actions, not an AI-driven image analysis or diagnostic algorithm. Its "performance" is its functional operation in conjunction with a human operator during a hysteroscopic procedure.

7. The type of ground truth used

For the performance testing mentioned, the "ground truth" is defined by:

• International Standards: e.g., ISO 10993 for biocompatibility, ISO 11135 for sterilization, IEC 60601-1/1-2 for electrical safety/EMC, IEC 62366-1 for usability.
• Device Design Specifications: For metrics like Intrauterine Pressure Control, Fluid Deficit Accuracy (± 50 mL), Maximum Inflow Rate, etc. These are objective engineering parameters.

8. The sample size for the training set

Not applicable. The Fluent Pro Fluid Management System is not an AI/ML device that requires a "training set" in the conventional sense for learning and model development. The software within the device undergoes testing (unit, integration, system testing) to ensure it performs according to its predefined logical and control functions, not to train a learning algorithm.

9. How the ground truth for the training set was established

Not applicable, as there is no "training set" for an AI/ML algorithm.

Ask a specific question about this device

The Aptima HCV Quant Dx assay is a real-time transcription-mediated amplification (TMA) test used for both detection and quantitation of hepatitis C virus (HCV) RNA in fresh and frozen human serum and plasma from HCV-infected individuals.

Plasma may be prepared in ethylenediaminetetraacetic acid (EDTA), anticoagulant citrate dextrose (ACD) solution, and plasma preparation tubes (PPT). Serum may be prepared in serum separator tubes (SST). Specimens are tested using the Panther system for automated specimen processing, and quantitation. Specimens containing HCV genotypes 1 to 6 are validated for detection and quantitation in the assay.

The Aptima HCV Quant Dx assay is indicated for use as an aid in the diagnosis of active HCV infection in the following populations: individuals with antibody evidence of HCV infection with evidence of liver disease of to be actively infected with HCV antibody evidence, and individuals at risk for HCV infection with antibodies to HCV. Detection of HCV RNA indicates that the virus is replicating and, therefore, is evidence of active infection of HCV RNA does not discriminate between acute and chronic states of infection.

The Aptima HCV Quant Dx assay is also indicated for use as an aid in the management of HCV infected patients undergoing HCV antiviral drug therapy. The assay can be used to measure HCV RNA levels periodically prior to, during, and after treatment to determine sustained virological response (SVR) or nonsustained virological response (NSVR), Assay performance characteristics have been established for individuals infected with HCV and treated with certain direct-acting antiviral agents (DAA) regimens. No information is available on the assay's predictive value when other therapies are used. The results from the Aptima HCV Quant Dx assay must be interpreted within the context of all relevant clinical and laboratory findings.

The Aptima HCV Quant Dx assay is not approved for use as a screening test for the presence of HCV RNA in blood or blood products.

Clearance of this pre-market application will 1) enable additional reagent on-board / off-board cycling, allowing operators to load reagents onto the Panther System 8 times instead of 5 times; and 2) update the Assay Definition Module (ADM) to detect, flag and invalidate results impacted by a faulty or flickering LED, using estimated background minimum limits. These changes do not introduce any changes to the original design, method of manufacture, assay procedure, principle of operation, mechanism of action, conditions of use or hardware of the Panther instrument, or to the results interpretation for the cleared assay.

The Aptima HCV Quant Dx assay is a nucleic acid amplification test that uses real-time TMA technology to detect and quantitate HCV RNA for aiding diagnosis or to establish baseline viral load, as well as to measure on-treatment and post-treatment responses. The assay targets a conserved region of the HCV genome, detecting and quantitating genotypes 1, 2, 3, 4, 5, and 6. The assay is standardized against the 2nd WHO International Standard for Hepatitis C Virus (NIBSC Code 96/798).

The Aptima HCV Quant Dx assay involves three main steps, which all take place in a single tube on the Panther system: target capture, target amplification by TMA, and detection of the amplification products (amplicon) by the fluorescent labeled probes (torches).

During target capture, viral RNA is isolated from specimens. The specimen is treated with a detergent to solubilize the viral envelope, denature proteins, and release viral genomic RNA. Capture oligonucleotides hybridize to highly conserved regions of HCV RNA, if present, in the test specimen. The hybridized target is then captured onto magnetic microparticles that are separated from the specimen in a magnetic field. Wash steps remove extraneous components from the reaction tube.

Target amplification occurs via TMA, which is a transcription-mediated nucleic acid amplification method that utilizes two enzymes, Moloney murine leukemia virus (MMLV) reverse transcriptase and T7 RNA polymerase. The reverse transcriptase is used to generate a DNA copy (containing a promoter sequence for T7 RNA polymerase) of the target sequence. T7 RNA polymerase produces multiple copies of RNA amplicon from the DNA copy template. The Aptima HCV Quant Dx assay utilizes the TMA method to amplify a portion of the 5' UTR of the HCV genome. Amplification of this region is achieved using specific primers which are designed to amplify HCV genotypes 1, 2, 3, 4, 5, and 6.

Detection is achieved using single-stranded nucleic acid torches that are present during the amplification of the target and that hybridize specifically to the amplicon in real time. Each torch has a fluorophore and a quencher. When the torch is not hybridized to the amplicon, the quencher is in close proximity of the fluorophore and suppresses the fluorescence. When the torch binds to the amplicon, the quencher is moved farther away from the fluorophore and it will emit a signal at a specific wavelength when excited by a light source. As more torches hybridize to amplicon a higher fluorescent signal is generated. The time taken for the fluorescent signal to reach a specified threshold is proportional to the starting HCV concentration. Each reaction has an internal calibrator/internal control (IC) that controls for variations in specimen processing, amplification, and detection. The concentration of a sample is determined by the Panther system software using the HCV and IC signals for each reaction and comparing them to calibration information.

The Aptima HCV Quant Dx assay has been designed for and validated on the Panther system. The Panther system is an integrated hardware and software system that together with the Aptima HCV Quant Dx assay fully automates all the steps necessary to perform the assay from sample preparation through amplification of nucleic acid, detection, data reduction and amplicon inactivation.

The provided document describes the acceptance criteria and a study for the Aptima HCV Quant Dx Assay.

Here's a breakdown of the requested information:

1. Table of Acceptance Criteria and Reported Device Performance

(*) Note: The document states "may be removed from calculation of Standard Deviatio logs from Target Concentration if Value Assignm" and "±0.50 logs from Target Concentration if Value", which appear to be incomplete sentences or truncated notes in the original document.

2. Sample Sizes Used for the Test Set and Data Provenance

• Intended Use Testing (Software 7.2.9 vs 7.2.7):
  • QC Control A (Negative): 120 samples.
  • QC Control B (2.16 log IU/mL): 40 samples.
  • QC Control D (6.52 log IU/mL): 40 samples.
  • QC Panel S (1.13 log IU/mL): 80 samples.
  • Data Provenance: Not explicitly stated beyond "previous software version, 7.2.7 (reported to the FDA in the 2022 PMA annual report for the Aptima HCV Quant Dx Assay, P160023/R004)".
• Assay Verification (5 to 8 reagent reload change):
  • HCV 3.3 log IU/mL: Not explicitly stated, but "5-10 cycles of reagent storage" and performance evaluated.
  • HCV 7.0 log IU/mL: Not explicitly stated, but "5-10 cycles of reagent storage" and performance evaluated.
  • Data Provenance: "prepared virus and armored RNA (aRNA) spiked into negative human serum matrix (BI0052)".
• Assay Verification (Flickering LED process control change):
  • 381 field instruments and 445,910 unique test orders were analyzed from March 2018 to March 2022.
  • Data Provenance: "field instruments" indicates real-world data, likely retrospective. Country of origin not specified.
• Additional Intended Use Testing (HCV Assay Software 5.2.5.1 vs 5.3.5.1):
  • Negative clinical specimen: 1 (tested with 10 replicates per run).
  • HCV RNA concentration around 3xLLoQ (30 IU/ml): 3 specimens (tested with 10 replicates of each per run).
  • Total replicates per software version: (1 negative specimen * 10 replicates) + (3 positive specimens * 10 replicates) = 40 replicates. This was done for both software versions.
  • Data Provenance: "clinical specimens", "negative clinical matrix panel", "low positive clinical positive panel". Samples prepared by diluting high positive sample in negative clinical matrix.
• Additional On-Board Stability (OBS) and Reagent Cycling:
  • Minimum of one negative sample, a sample around 3xLLoQ (30 IU/ml), and a moderate positive or high positive sample.
  • Minimum of 5 replicates per sample for each timepoint, with more replicates at baseline.
  • Data Provenance: Not explicitly stated, but implies laboratory-controlled testing.

3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications of Those Experts

Not applicable. This device is a diagnostic assay for HCV RNA detection and quantitation, not an imaging device that requires interpretation by human experts. The ground truth would be based on the quantitative/qualitative results of the assay itself and reference standards/controls.

4. Adjudication Method for the Test Set

Not applicable. As noted above, this refers to a diagnostic assay. The "results" are numerical or qualitative (detected/not detected) from an automated system compared against established performance characteristics and controls.

5. If a Multi Reader Multi Case (MRMC) Comparative Effectiveness Study was done, If so, what was the effect size of how much human readers improve with AI vs without AI assistance

Not applicable. This information is for an in-vitro diagnostic (IVD) assay, not an AI-powered medical imaging or decision support system that typically involves human readers.

6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done

Yes, the studies described are essentially standalone performance evaluations of the assay system (Aptima HCV Quant Dx Assay on the Panther system software). The device operates automatically from sample preparation through amplification, detection, and quantitation. The performance checks compare the output of the system against expected values, referent methods, or established criteria.

7. The Type of Ground Truth Used

The ground truth for the performance evaluations appears to be based on:

• Reference standards and controls: For quantitative and qualitative measurements (e.g., QC Control A, B, D, S, prepared virus and armored RNA, 3xLLoQ and negative samples).
• Previous established performance: Comparison of new software/hardware changes against the predicate device's established performance (PMA P160023).
• Root-cause analysis: For the flickering LED study, specific root-cause analysis was used to identify true faults.

8. The Sample Size for the Training Set

No training set information is provided as this is a diagnostic assay product, not a machine learning algorithm that requires a distinct training phase. The "development" or "calibration" of the assay would typically involve internal studies not detailed here, but these are not referred to as "training sets" in the context of conventional IVD development.

9. How the Ground Truth for the Training Set Was Established

Not applicable, as no training set is described.

Ask a specific question about this device

The Panther Fusion SARS-CoV-2/Flu A/B/RSV assay is a fully automated multiplexed real-time polymerase chain reaction (RT-PCR) in vitro diagnostic test intended for the qualitative detection and differentiation of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), influenza A virus (Flu A), influenza B virus (Flu B), and respiratory syncytial virus (RSV). Nucleic acids are isolated and purified from nasopharyngeal (NP) swab specimens and anterior nasal (AN) swab specimens obtained from individuals exhibiting signs and symptoms of a respiratory tract infection. Clinical signs and symptoms of respiratory viral infection due to SARS-CoV-2, influenza, and RSV can be similar. This assay is intended to aid in the differential diagnosis of SARS-CoV-2, Flu A, Flu B, and RSV infections in humans and is not intended to detect influenza C virus infections.

Nucleic acids from the viral organisms identified by this test are generally detectable in NP and AN swab specimens during the acute phase of infection and identification of specific viral nucleic acids from individuals exhibiting signs and symptoms of respiratory tract infective of the presence of the identified virus and aids in diagnosis if used in conjunction with other clinical and epidemiological information, and laboratory findings. The results of this test should not be used as the sole basis for diagnosis, treatment, or other patient management decisions. Positive results do not rule out coinfection with other organism(s) detected by the Panther Fusion SARS-CoV-2/Flu A/B/RSV assay may not be the definite cause of disease. Negative results do not preclude SARS-CoV-2, influenza A virus, influenza B virus, or RSV infections. This assay is designed for use on the Panther Fusion system.

The Hologic RespDirect Collection Kit is cleared for NP swab specimens only for testing with the Panther Fusion SARS-CoV-2/Flu A/B/RSV assay.

The Panther Fusion SARS-CoV-2/Flu A/B/RSV assay is classified as a Class II in vitro diagnostic device per 21 CFR 866.3981 and has product code QOF. The Panther Fusion SARS-CoV-2/Flu A/B/RSV assay is designed for use on the fully automated Panther Fusion System.

The Panther Fusion system is a class II exempt device under 21CFR 862.2570 that has product code OOI.

The Panther Fusion SARS-CoV-2/Flu A/B/RSV assay is a multiplex real-time reverse transcriptase PCR (RT-PCR) in vitro diagnostic test developed for use on the fully automated Panther Fusion system to detect and differentiate SARS-CoV-2, influenza A, influenza B, and respiratory syncytial virus (RSV) directly from nasopharyngeal and anterior nasal swab specimens, from individuals exhibiting signs and symptoms of a respiratory tract infection.

The Panther Fusion SARS-CoV-2/Flu A/B/RSV assay involves the following steps: sample lysis, nucleic acid capture and elution transfer, and multiplex RT-PCR where analytes (when present) are simultaneously amplified, detected and differentiated. Nucleic acid capture and elution takes place in a single tube on the Panther Fusion system. The eluate is transferred to the Panther Fusion system reaction tube containing the assay reagents. Multiplex RT-PCR is then performed for the eluted nucleic acid on the Panther Fusion system.

Here's a breakdown of the acceptance criteria and the study details for the Panther Fusion SARS-CoV-2/Flu A/B/RSV assay, based on the provided document:

1. Table of Acceptance Criteria and Reported Device Performance

Note on Acceptance Criteria: The document explicitly states the goal is to demonstrate "comparable" performance to the predicate device (which uses an NP swab) for the new anterior nasal (AN) swab specimen type. While specific numerical acceptance criteria (e.g., "PPA must be >X%") are not provided in this summary, the reported performance metrics (PPA, NPA, and their confidence intervals) are presented to demonstrate this comparability and overall effectiveness for diagnostic use, as typical for molecular assays expanding their indications. The high NPA values for all targets indicate good specificity, and generally high PPA values indicate good sensitivity, especially when considering the confidence intervals. The lower PPA for Flu B in the prospective study was addressed and supported by the retrospective study, which showed strong performance.

2. Sample Sizes Used for the Test Set and Data Provenance

• Prospective Clinical Study (Anterior Nasal Swab):

  • Sample Size: 1,268 individuals were enrolled. 1,230 AN swab specimens were tested, resulting in 1,189 evaluable specimens for analysis (not all evaluable for all analytes).
  • Data Provenance: Prospective, multicenter study conducted in the US. Specimens were collected from individuals attending nine participating medical facilities in the US during the 2022-2023 respiratory infection season.

• Retrospective Clinical Study (Anterior Nasal Swab - Supplement for Flu B and RSV):

  • Sample Size: 175 preselected retrospective specimens.
  • Data Provenance: Retrospective. The origin country is not explicitly stated but is implied to be related to the US context from the prospective study.

• Analytical Studies (Reprocessing from K222736):

  • Sample Size: Not explicitly stated as "sample size" for each analytical study detail, but refers to the number of replicates/conditions tested for LOD, inclusivity, exclusivity, etc. For example, Flu B LoD confirmation involved 28 replicates per panel.
  • Data Provenance: Already existing data from the previous 510(k) submission (K222736) using Nasopharyngeal (NP) swabs, reprocessed with the updated software.

3. Number of Experts Used to Establish Ground Truth for the Test Set and Qualifications

The document does not mention the use of experts to establish ground truth for the clinical test sets. Instead, the ground truth was established by comparing the candidate device's results to reference molecular assays:

• SARS-CoV-2: A composite comparator algorithm (CCA) consisting of up to three highly sensitive US FDA EUA SARS-CoV-2 molecular tests. A final CCA result was assigned when two of the three comparator assays were in concordance.
• Flu A, Flu B, and RSV: A US FDA-cleared molecular Flu A/B/RSV assay.

4. Adjudication Method for the Test Set

• SARS-CoV-2 Ground Truth: For SARS-CoV-2, a composite comparator algorithm (CCA) was used. A final CCA result was assigned when "two of the three composite comparator assays were in concordance." This effectively acts as an adjudication method where agreement among a majority of reference tests determines the ground truth.
• Flu A, Flu B, RSV Ground Truth: For Flu A, Flu B, and RSV, a single "US FDA-cleared molecular Flu A/B/RSV assay" was used as the comparator method. There's no mention of an adjudication process among multiple comparators for these targets.

5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study

• No, a Multi-Reader Multi-Case (MRMC) comparative effectiveness study was not explicitly mentioned or conducted. This type of study is more common for imaging devices or diagnostics that involve subjective interpretation by human readers. This device is an automated molecular diagnostic assay, where the output is objective (positive/negative, Ct value).

6. Standalone (Algorithm Only) Performance Study

• Yes, the performance data presented (PPA, NPA) directly reflects the standalone performance of the Panther Fusion SARS-CoV-2/Flu A/B/RSV assay. It is compared to a ground truth established by other molecular tests, without human interpretation of the assay's output influencing the performance metrics. The assay itself is automated, so its performance is inherently "algorithm only" in the context of generating results.

7. Type of Ground Truth Used

• Molecular Comparator Assays: The ground truth for the clinical test sets (both prospective and retrospective) was established using other legally marketed and highly sensitive molecular diagnostic tests.
  • For SARS-CoV-2: A composite of up to three US FDA EUA SARS-CoV-2 molecular tests.
  • For Flu A, Flu B, RSV: A US FDA-cleared molecular Flu A/B/RSV assay.

8. Sample Size for the Training Set

• The document does not explicitly state the sample size used for the training set. This submission focuses on the validation of an existing assay with a new specimen type (anterior nasal swab) and a software update (Adaptive Crosstalk Correction). The original development and training data for the core assay (K222736) would have been described in that prior submission, but are not detailed here for this specific submission's scope. The mention of "reprocessed with validated Results Processor Tool" for previously submitted analytical studies also suggests reuse of data rather than new training.

9. How the Ground Truth for the Training Set Was Established

• Since the document doesn't explicitly detail a new "training set" for this specific submission, it also doesn't describe how ground truth for such a set was established. It's likely that the original assay development followed similar methods of using well-characterized samples or comparator assays for internal development and optimization.
• Regarding the Adaptive Crosstalk Correction (ACC) factor, the validation involved "testing of SARS-CoV-2 positive clinical specimens that are representative of those that yielded false positive Flu B results in the field." This indicates that the problem the ACC addresses arose from real-world performance, and the resolution was validated against specific clinical scenarios rather than a formal "training set" in the machine learning sense.

Ask a specific question about this device

The Genius TM Digital Diagnostics System with the Genius™ Cervical AI algorithm includes the Genius TM Digital Imager, Genius™ Image Management Server (IMS), the Genius™ Review Station, and the Genius™ Cervical AI algorithm. The Genius™ Digital Diagnostics System with the Genius™ Cervical AI algorithm is intended for the creation and viewing of digital images of scanned ThinPrep® Pap Test glass slides. Objects of interest selected by the Genius™ Cervical AI algorithm from the scanned digital image are presented in a gallery format next to the image of the whole cell spot on the Genius™ Review Station for review and interpretation. The Genius Digital Diagnostics System with the Genius™ Cervical AI algorithm is intended to aid in cervical cancer screening for the presence of atypical cells, cervical neoplasia, including its precursor lesions (Low Grade Squamous Intraepithelial Lesions, High Grade Squamous Intraepithelial Lesions) and carcinoma, as well as all other cytological categories as defined by The Bethesda System for Reporting Cervical Cytology1.

After digital review with the Genius Cervical AI algorithm, if there is uncertainty in the diagnosis, then direct examination of the glass slide by light microscopy should be performed. Digital images from the Genius Digital Diagnostics System with the Genius™ Cervical AI algorithm should be interpreted by qualified cytologists in conjunction with the patient's screening history, other risk factors, and professional guidelines which guide patient management.

The Genius Digital Diagnostics System includes, Genius Digital Image. Management Server, Genius Review Station(s), and Genius Cervical AI algorithm, as shown in Figure 1 below.

Here's a breakdown of the acceptance criteria and the study that proves the Genius™ Digital Diagnostics System with the Genius™ Cervical AI algorithm meets them, based on the provided text:

Acceptance Criteria and Device Performance for Genius™ Digital Diagnostics System with Genius™ Cervical AI algorithm

1. Table of Acceptance Criteria and Reported Device Performance

The document doesn't explicitly lay out bulleted acceptance criteria for the entire system in a single section. However, performance benchmarks derived from various studies are detailed. The primary clinical acceptance criteria revolve around the AI-assisted review being comparable or superior to the standard of care (manual review) and other automated systems (TIS review) in terms of sensitivity and specificity at various diagnostic thresholds for cervical cytology. Key technical performance criteria are also highlighted.

Here is a summary, inferred from the "Performance Characteristics" section:

2. Sample Size Used for the Test Set and Data Provenance

• Sample Size for Clinical Test Set: 1994 slides.
• Data Provenance:
  • Country of Origin: United States ("A multi-center Genius Cervical AI Algorithm Clinical Study was performed within the United States.").
  • Retrospective/Prospective: The study used "residual material after the clinical sites signed out the case," implying these were historical samples (retrospective) collected from women screened for cervical cancer using the ThinPrep Pap test.

3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications

• Number of Experts: An adjudication panel composed of 3 adjudication CT/PT teams, each consisting of 1 Cytotechnologist (CT) and 1 Pathologist (PT). So, a minimum of 3 CTs and 3 PTs were involved in establishing the ground truth.
• Qualifications of Experts: They were "qualified cytologists" and "qualified CTs and PTs" (implied as experts in cervical cytology based on their role in the study and laboratory context). The participating laboratories had "extensive experience in the processing and evaluation of gynecologic ThinPrep Pap test slides" and the experts were "trained in the use of the Genius Digital Diagnostics System with Genius Cervical AI algorithm."

4. Adjudication Method for the Test Set

• Method: Consensus-based adjudication followed by multi-headed microscope review for discordant cases.
  1. Each of the three adjudication CT/PT teams independently reviewed the slides.
  2. A consensus result was obtained if there was majority agreement (at least two of the three adjudication CT/PT teams).
  3. If a consensus was not initially obtained, these cases underwent review by the three adjudication PTs simultaneously using a multi-headed microscope (multi-head review).
  4. The final "reference" or "ground truth" diagnosis was based on either the initial consensus or the multi-head review result.

5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study

• Was it done? Yes, a multi-center, multi-reader, multi-case study was performed, comparing Genius AI review, Manual review, and TIS review. This is evident from the "Clinical Studies" section, where 4 clinical sites participated, and at each site, 3 independent CT/PT teams reviewed cases.

• Effect Size of Human Readers Improvement with AI vs. without AI Assistance:
  The study directly compares human performance with AI assistance (Genius AI review) to human performance without AI assistance (Manual review). The improvement is shown in sensitivity at various diagnostic thresholds:

  • LSIL+: Sensitivity increased by 4.4% with Genius AI assistance compared to Manual review (89.1% vs 84.7%). This was statistically significant.
  • ASC-H+: Sensitivity increased by 8.2% with Genius AI assistance compared to Manual review (87.8% vs 79.6%). This was statistically significant.
  • HSIL+: Sensitivity increased by 7.5% with Genius AI assistance compared to Manual review (81.5% vs 74.0%). This was statistically significant.
  • ASCUS+: Sensitivity increased by 1.6% (91.7% vs 90.1%), but this was not statistically significant.
  • UNSAT (for correct identification): Sensitivity increased by 18.6% with Genius AI assistance (80.4% vs 61.8%).

  While specificity generally saw small, statistically significant decreases at some thresholds with AI assistance, the overall benefit-risk assessment concluded that the probable benefits outweigh the risks due to the increased sensitivity, particularly for higher-grade lesions. The document also highlights a reduction in false negative rates (false NILM) with the Genius AI: "The 7.5% increase in HSIL + sensitivity means a decrease in Manual false negative rate of 26% to 18.5% false negative rate by the Genius Digital Diagnostics System with the Genius Cervical AI algorithm resulted in 28.8% reduction in the number false negative reviews (28.8%=(26%-18.5%) / 26%)."

6. Standalone (Algorithm Only Without Human-in-the-Loop Performance) Study

• Was it done? Not explicitly for diagnostic interpretation. The "Objects of Interest (OOI) Reproducibility" study evaluated the AI algorithm's ability to select OOIs against the adjudicated reference diagnosis, but the primary clinical performance assessment was human-in-the-loop (AI-assisted review). The AI's function is to "selects OOIs to be displayed for review by a CT or PT," not to render a final diagnosis on its own.

7. Type of Ground Truth Used

• Type of Ground Truth: Expert Consensus. The "adjudicated diagnosis" served as the "gold standard" or "reference" (ground truth). This was established by a panel of 3 CT/PT teams reaching consensus, with multi-head microscope review for discordant cases.

8. Sample Size for the Training Set

• The document mentions the use of "Convolutional Neural Network (CNN) technology" for the AI algorithm but does not explicitly state the sample size for the training set. It only mentions that the "Genius Cervical AI algorithm v1.0.16.0 will remain locked for use with the authorized device and will not be continually trained and improved with each cohort analyzed in clinical practice, after marketing authorization," implying it was trained on a fixed dataset prior to these performance studies.

9. How Ground Truth for the Training Set Was Established

• The document does not explicitly describe how the ground truth for the training set was established. It focuses only on the ground truth for the test set used for performance validation.

Ask a specific question about this device

The Aptima Neisseria gonorrhoeae (GC) Assav is an in vitro qualitative nucleic acid amplification (NAAT) the detection of ribosomal RNA (rRNA) from Neisseria gonorrhoeae (GC) to aid in the diagnosis of gonocccal urogenital disease using the Panther System. The assay may be used to test male urine specimens from symptomatic and asymptomatic individuals.

The Aptima GC assay is a target amplification nucleic acid probe test for in vitro qualitative detection of ribosomal RNA (rRNA) from Neisseria gonorrhoeae (GC). The Aptima Neisseria gonorrhoeae Assay combines the technologies of target capture, transcription-mediated amplification (TMA), and hybridization protection assay (HPA). Specimens are collected and transferred into their respective specimen transport tubes. The transport solution in these tubes releases the rRNA target and protects it from degradation during storage. When the Aptima Neisseria gonorrhoeae Assay is performed in the laboratory, the target rRNA molecule is isolated from the specimens by use of a capture oligomer via target capture that utilizes magnetic microparticles. The capture oligomer contains a sequence complementary to a specific region of the target molecule as well as a string of deoxyadenosine residues. During the hybridization step, the sequence specific region of the capture oligomer binds to a specific region of the target molecule. The capture oligomer:target complex is then captured out of solution by decreasing the temperature of the reaction to room temperature. This temperature reduction allows hybridization to occur between the deoxyadenosine region on the capture oligomer and the poly-deoxythymidine molecules that are covalently attached to the magnetic particles. The micro particles, including the captured target molecule bound to them, are pulled to the side of the reaction vessel using magnets and the supernatant is aspirated. The particles are washed to remove residual specimen matrix that may contain amplification reaction inhibitors. After the target capture steps are completed, the specimens are ready for amplification. Target amplification assays are based on the ability of complementary oligonucleotide primers to specifically anneal and allow enzymatic amplification of the target nucleic acid strands. The Hologic TMA reaction replicates a specific region of the 16S rRNA from GC via DNA intermediates. A unique set of primers is used for the target molecule. Detection of the rRNA amplification product sequences (amplicon) is achieved using nucleic acid hybridization. A single-stranded chemiluminescent DNA probe, which is complementary to a region of the target amplicon, is labeled with an acridinium ester molecule. The labeled DNA probe combines with amplicon to form stable RNA:DNA hybrids. The Selection Reagent differentiates hybridized from unhybridized probe, eliminating the generation of signal from unhybridized probe. During the detection step, light emitted from the labeled RNA:DNA hybrids is measured as photon signals in a luminometer, and are reported as Relative Light Units (RLU). The device reagents are identical to the Aptima Neisseria gonorrhoeae Assay reagents for use on the Tigris DTS system but are intended for use on the Panther system with different specimen type indications. The Panther and Tigris DTS systems use the same principles of operation.

Here's a breakdown of the acceptance criteria and study details for the Aptima Neisseria gonorrhoeae Assay, based on the provided FDA 510(k) summary:

This device is an in vitro qualitative nucleic acid amplification test (NAAT) for the detection of Neisseria gonorrhoeae (GC) rRNA to aid in the diagnosis of gonococcal urogenital disease in male urine specimens using the Panther System. It does not appear to involve AI assistance or human reader studies.

1. Acceptance Criteria and Reported Device Performance

The provided document does not explicitly state pre-defined acceptance criteria for clinical performance (sensitivity and specificity) as a set of numerical thresholds. Instead, it states that the study data demonstrate that performance of the Aptima Neisseria gonorrhoeae Assay on the Panther system is substantially equivalent to that of currently FDA-cleared assays for male urine specimens.

However, analytical studies do present performance metrics that implicitly act as acceptance criteria, for example, for Limit of Detection (LoD), precision, and specificity (no interference).

Here's a summary of reported device performance based on the provided text:

Table of Reported Device Performance

2. Sample Size Used for the Test Set and Data Provenance

• Clinical Test Set:
  • Total Enrolled Subjects: 2085 male subjects.
  • Evaluable Subjects: 1959 male subjects (126 subjects not evaluable).
  • Specimens Included in Performance Analysis: 1958 male urine specimens (one specimen with final GC equivocal result was excluded).
  • Data Provenance: Prospective, multi-center clinical study conducted at 11 geographically and ethnically diverse US clinical sites.
    • Sites included obstetrics and gynecology, family planning, and STI clinics.
    • Specimens were collected from symptomatic and asymptomatic men.

3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications

• The document does not explicitly state the number of experts used to establish ground truth for the clinical test set.
• The ground truth (Patient Infected Status - PIS) was established using up to 3 FDA-cleared NAATs (Nucleic Acid Amplification Tests). This implies that a consensus or reference standard approach using multiple existing, cleared diagnostic methods was used, rather than individual expert review or pathological examination for individual cases.

4. Adjudication Method for the Test Set

• The document states that the Patient Infected Status (PIS) was established using "up to 3 FDA-cleared NAATs". This indicates a composite reference standard approach.
• The specific adjudication rule (e.g., 2/3 positive, 3/3 positive) is not detailed, but the use of multiple FDA-cleared methods implies a robust, multi-test consensus for the ground truth.

5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study

• No, an MRMC comparative effectiveness study was not done.
• This device is an in vitro diagnostic (IVD) assay (a qualitative nucleic acid amplification test), not an imaging AI device that assists human readers. Therefore, there is no "human-in-the-loop" component or an effect size for human readers improving with AI assistance.

6. Standalone Performance

• Yes, standalone performance was done.
• The described clinical study evaluates the performance of the "Aptima Neisseria gonorrhoeae Assay on the Panther System" directly against a composite reference standard (PIS derived from FDA-cleared NAATs). This is a standalone performance assessment of the algorithm/device itself, without human intervention in the result interpretation.

7. Type of Ground Truth Used

• The ground truth used for the clinical test set was a composite reference standard, referred to as "Patient Infected Status (PIS)".
• PIS was established by testing male urine specimens with "up to 3 FDA-cleared NAATs". This is a highly robust method, relying on multiple well-validated diagnostic tests to determine the true infection status.

8. Sample Size for the Training Set

• The document does not specify the sample size for the training set.
• This document is a 510(k) summary for premarket notification, focusing on the validation of the device for its intended use. Information regarding the development and training of the assay (e.g., specific molecular sequences, probe design) is generally proprietary and not included in this type of submission. The focus is on the performance of the final, locked version of the device.

9. How the Ground Truth for the Training Set Was Established

• The document does not specify how the ground truth for any potential training set was established, nor explicitly mention a distinct training set.
• For an IVD such as this, the "training" (or development and optimization) typically involves extensive analytical studies (e.g., primer design, probe specificity, optimization of reaction conditions, LoD determination, cross-reactivity testing) rather than machine learning-style "training data" with a "ground truth" in the same sense as an AI imaging algorithm. The core of the assay relies on well-established molecular biology principles and target identification.

Ask a specific question about this device