Search Results

JLK-NCCT is a radiological computer-aided triage and notification software designed for analyzing non-contrast head CT (NCCT) images. The software assists hospital networks and trained clinicians by flagging and communicating them of findings suggestive of (1) Intracranial Hemorrhage (ICH) and (2) large vessel occlusion (LVO) involving the internal carotid artery (ICA), middle cerebral artery M1 (MCA-M1) and middle cerebral artery M2 (MCA-M2) on NCCT images.

JLK-NCCT employs an artificial intelligence (AI) algorithm to analyze images and highlight cases with detected (1) ICH or (2) LVO on an on-premises or cloud-based JLK server. This occurs in parallel with the ongoing standard of care image interpretation. Users receive notifications for cases with suspected ICH and LVO findings via mobile devices. Notifications include compressed preview images for informational purposes only and not intended for diagnostic use beyond notification.

The device does not modify the original medical image, and is not intended to be used as a primary diagnostic device. The results of JLK-NCCT are intended to be used in conjunction with other patient information and professional judgment to assist with triage and prioritization of medical images. Clinicians who receive notifications are responsible for reviewing full images per the standard of care. JLK-NCCT is intended for adults use only.

JLK-NCCT is a radiological computer-assisted triage and notification (CADt) software that complies with the DICOM standard. It functions as a Non-Contrast Computed Tomography (NCCT) processing module, prioritizing triage, and notification for suspected intracranial hemorrhage (ICH) and large vessel occlusion (LVO). Operating as a notification-only tool, it assists hospital networks and clinicians by flagging critical cases and alerting specialists independent of standard workflows. JLK-NCCT's AI algorithm analyzes NCCT scans for ICH and LVO indicators and provides automated notifications to streamline clinical decision-making.

JLK-NCCT consists of an AI-based image analysis algorithm hosted on JLK servers either on-premises or cloud-based and a mobile software module for notification management. The AI processes NCCT head scans, detecting suspected ICH and LVO, and transmits mobile notifications with compressed preview images for triage. PACS integration is optional and supported when available. The system does not modify original medical images and is not intended for diagnostic use. JLK-NCCT integrates the JLK-ICH Model (ICH detection), LVO score Model (ischemic assessment), and HAS Model (Hyperdense Artery Sign detection), supporting real-time alerts and prioritization within hospital workflows (the ICH algorithm is the same as the device cleared under K243363).

JLK-NCCT was trained using clinical datasets from the U.S. and South Korea, totaling 3,067 cases, sourced from multiple institutions to ensure diversity and robustness. The dataset included NCCT scans acquired using imaging equipment from various manufacturers, such as GE, Siemens, Philips, and Toshiba, covering a range of scanning parameters to enhance model generalizability. Data were collected from institutions across different geographic locations, including hospitals in North Carolina and Texas in the U.S., as well as Seoul St. Mary's Hospital and other South Korean medical centers. All imaging studies were labeled by board-certified neuroradiologists. This diverse dataset strengthens the AI model's applicability across various clinical environments, supporting its role as a triage and notification tool for assisting clinicians in early detection and prioritization of suspected ICH and LVO cases.

The performance of the device's AI algorithms was validated in a standalone performance evaluation using an independent dataset different from the one used for algorithm training. Each case output from the JLK-NCCT device was compared with a ground truth standard determined by two ground truthers, with a third ground truther intervening in cases of disagreement (i.e., 2+1 truther scheme). All truthers were US board-certified neuroradiologists.

Here's a breakdown of the acceptance criteria and study details for the JLK-NCCT device, based on the provided FDA 510(k) clearance letter:

Acceptance Criteria and Reported Device Performance

Note: The document primarily focuses on LVO detection performance for the standalone study, as the ICH algorithm was previously cleared.

Study Details for JLK-NCCT Performance Evaluation

Here's a detailed breakdown of the study that proves the device meets the acceptance criteria:

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

   • Sample Size: 288 cases for the standalone evaluation (144 LVO Positive, 144 LVO Negative).
   • Data Provenance: Retrospective study. The data were "newly acquired" and "independent of the training dataset." No specific countries of origin for the test set are mentioned, but the training data was from the U.S. and South Korea.

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

   • Number of Experts: Two ground truthers, with a third intervening in cases of disagreement.
   • Qualifications of Experts: All truthers were U.S. board-certified neuroradiologists.

3. Adjudication Method for the Test Set:

   • Method: 2+1 truther scheme. Two ground truthers independently assessed each case, and a third ground truther intervened to resolve any disagreements.

4. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done, and the Effect Size:

   • Was it done?: Yes, a reader performance study was conducted.
   • Effect Size (Improvement with AI vs. without AI assistance):
     • Sensitivity: JLK-NCCT demonstrated a sensitivity of 0.792.
       • Improvement over all readers (Neuroradiologist and General Radiologist average): 0.224 (0.792 - 0.568), (95% CI: 0.144–0.306).
       • Improvement over general radiologists: 0.159 (0.792 - 0.633), (95% CI: 0.083–0.237).
       • Improvement over neuroradiologists: 0.267 (0.792 - 0.525), (95% CI: 0.174–0.356).
     • Specificity: JLK-NCCT demonstrated a specificity of 0.933.
       • Improvement over all readers (Neuroradiologist and General Radiologist average): 0.093 (0.933 - 0.840), (95% CI: 0.038–0.150).
       • Improvement over general radiologists: 0.137 (0.933 - 0.796), (95% CI: 0.070–0.205).
       • Improvement over neuroradiologists: 0.064 (0.933 - 0.869), (95% CI: 0.005–0.126).
     • The device "satisfied both criteria show General Radiologist superiority and Neuroradiologist non-inferiority." This indicates that the AI significantly improved performance for general radiologists and did not degrade (or potentially improved) performance for neuroradiologists.

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

   • Was it done?: Yes. The document states, "A standalone performance evaluation using an independent dataset different from the one used for algorithm training." Performance metrics (Sensitivity, Specificity, AUC for LVO detection, and Time-to-Notification) are reported for this standalone performance.

6. The Type of Ground Truth Used:

   • Type: Expert consensus, established by U.S. board-certified neuroradiologists using a 2+1 truther scheme.

7. The Sample Size for the Training Set:

   • Sample Size: 3,067 cases.

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

   • Method: "All imaging studies were labeled by board-certified neuroradiologists." The exact number of neuroradiologists per case or the adjudication method for the training set is not specified, but it was expert-labeled.

Ask a specific question about this device

Intended context:
Annalise Enterprise is a device designed to be used in the medical care environment to aid in triage and prioritization of studies with features suggestive of the following findings:

• acute infarct*

See Additional Information, next page.

The device analyzes studies using an artificial intelligence algorithm to identify findings. It makes study-level output available to an order and imaging management system for worklist prioritization or triage.

The device is not intended to direct attention to specific portions of an image and only provides notification for suspected findings.

Its results are not intended:

• to be used on a standalone basis for clinical decision making
• to rule out specific findings, or otherwise preclude clinical assessment of non-contrast computed tomography brain

Intended modality:
Annalise Enterprise identifies suspected findings in non-contrast brain CT studies.

Intended user:
The device is intended to be used by trained clinicians who are qualified to interpret CTB studies as part of their scope of practice.

Intended patient population:
The intended population is patients who are 22 years or older.

Additional information:
The device includes acute infarct of the cerebral hemispheres or cerebellum, also including early signs of acute middle cerebral artery (MCA) infarct such as insular ribbon sign and disappearing basal ganglia sign.

The infarct must be a completed infarct (i.e. include an ischemic core of ≥5mL). The device also includes hyperdense artery in the anterior circulation but does not include lacunar infarcts, brainstem infarcts or venous infarcts.

The radiological device definition of acute infarct includes the following territories and regions:

• anterior cerebral artery (ACA)
• middle cerebral artery (MCA)
• posterior cerebral artery (PCA)
• cerebellum
• basilar artery occlusions
• watershed regions

Specificity may be reduced in the presence of infarcts of <5mL.

Annalise Enterprise is a software workflow tool which uses an artificial intelligence (AI) algorithm to identify suspected findings on non-contrast brain CT studies in the medical care environment. The finding identified by the device is acute infarct.

Radiological findings are identified by the device using an AI algorithm – a convolutional neural network trained using deep-learning techniques. Images used to train the algorithm were sourced from datasets that included a range of patient demographics and technical characteristics, including different equipment manufacturers and machines. This dataset, which contained over 200,000 CT brain imaging studies, was labelled by trained radiologists regarding the presence of the findings of interest.

The performance of the device's AI algorithm was validated in a standalone performance evaluation, in which the case-level output from the device was compared with a reference standard ('ground truth'). This was determined by two ground truthers, with a third truther used in the event of disagreement. All truthers were US board-certified neuroradiologists or neurologists.

The device interfaces with image and order management systems (such as PACS/RIS) to obtain non-contrast brain CT studies for processing by the AI algorithm. Following processing, if any of the clinical findings of interest are identified in the study, the device provides a notification to the image and order management system for prioritization of that study in the worklist. This enables users to review the studies containing features suggestive of these clinical findings earlier than in the standard clinical workflow. It is important to note that the device will never decrease a study's existing priority in the worklist. This ensures that worklist items will never have their priorities downgraded based on AI results.

The device workflow is performed parallel to and in conjunction with the standard clinical workflow for interpretation of non-contrast brain CT studies. The device is intended to aid in prioritization and triage of radiological medical images only.

Here's a summary of the acceptance criteria and the study proving the device meets those criteria, based on the provided FDA 510(k) Clearance Letter for Annalise Enterprise:

Acceptance Criteria and Device Performance

The acceptance criteria for Annalise Enterprise are established through its performance in detecting acute infarct on non-contrast brain CT studies. The device's performance is demonstrated through its Area Under the Curve (AUC), sensitivity, and specificity at various operating points for different slice thicknesses.

Table of Acceptance Criteria and Reported Device Performance (Acute Infarct Detection)

Study Proving Device Meets Acceptance Criteria

The device performance was assessed in four performance studies, including standalone performance and triage effectiveness evaluations. The primary study described in detail for acute infarct detection is a standalone performance evaluation.

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

• Acute Infarct Detection (Standalone Performance Evaluation):

  • Slice Thickness ≤ 1.5mm cohort: 977 cases.
  • Slice Thickness > 1.5mm & ≤ 5.0mm cohort: 1050 cases.
  • Total: 2027 cases.
  • Data Provenance: Retrospective, anonymized cases collected consecutively from five US hospital network sites. The test dataset was newly acquired and independent from the training dataset. The datasets included a range of patient demographics (gender, age, ethnicity, race) and technical parameters (imaging equipment make and model: GE Healthcare, NeuroLogica, Siemens, and Toshiba CT scanners).

• Triage Effectiveness (Turn-around Time):

  • Sample Size: n=277 cases.
  • Data Provenance: Cases positive for any of the findings eligible for prioritization, collected from multiple data sources spanning a variety of geographical locations, patient demographics and technical characteristics. Internal bench study.

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

• Ground Truthers: At least two experts for each case, with a third adjudicator in case of disagreement.
• Qualifications:
  • For cases with advanced imaging: ABR-certified and protocol-trained neuroradiologists.
  • For negative cases with chart-based interpretations: ABR-certified, protocol-trained neuroradiologists and/or neurologists.

3. Adjudication Method for the Test Set

The adjudication method used was 2+1, meaning:

• Each deidentified case was annotated in a blinded fashion by at least two ground truthers.
• Consensus was determined by the two ground truthers.
• In the event of disagreement between the first two ground truthers, a third ground truther adjudicated to establish the final ground truth.

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

No, an MRMC comparative effectiveness study was not explicitly described as being done to measure how much human readers improve with AI vs without AI assistance. The performance evaluation focuses on the standalone performance of the AI algorithm and its triage effectiveness (turnaround time effectiveness) compared to the predicate device and standard of care.

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

Yes, a standalone performance evaluation was done for acute infarct detection. The case-level output from the device's AI algorithm was compared directly with the established ground truth.

6. The Type of Ground Truth Used

• For positive cases (acute infarct): Ground truth was established by expert consensus of ABR-certified neuroradiologists, utilizing advanced imaging data for deidentified cases.
• For negative cases: Ground truth was established by expert consensus of ABR-certified neuroradiologists and/or neurologists, potentially referencing chart-based interpretations.

7. The Sample Size for the Training Set

The training set contained over 200,000 CT brain imaging studies.

8. How the Ground Truth for the Training Set was Established

The images used to train the algorithm were sourced from datasets where the presence of the findings of interest (acute infarct) was labelled by trained radiologists. The document does not specify the number of radiologists per case or an adjudication method for the training data ground truth, nor does it detail their specific qualifications beyond "trained radiologists."

Ask a specific question about this device

BriefCase-Triage: CARE (Clinical AI Reasoning Engine) Multi-Triage CT for Pneumothorax; Pericardial effusion; Large aortic aneurysm; Shoulder fracture or dislocation device is a radiological computer aided triage and notification software indicated for use in the analysis of contrast and non-contrast CT images of the chest, abdomen, or chest/abdomen, in adults or transitional adolescents aged 18 and older. The device is intended to assist hospital networks and appropriately trained medical specialists in workflow triage by flagging and communicating suspected positive findings, per study, of:

• Pneumothorax;
• Pericardial effusion;
• Large aortic aneurysm
• Shoulder Fracture or Dislocation

The device flags cases with at least one suspected finding to assist with triage/prioritization of medical images. The device will provide a flag for each suspected finding within this study. A preview image will be provided for each distinct suspected finding.

BriefCase-Triage uses a foundation model-based artificial intelligence (AI) system to analyze images and highlight cases with detected findings in parallel to the ongoing standard of care image interpretation. The user is presented with notifications for cases with suspected findings. Notifications include compressed preview images for each suspected finding that are meant for informational purposes only and not intended for diagnostic use beyond notification. The device does not alter the original medical images and is not intended to be used as a diagnostic device.

The results of BriefCase-Triage are intended to be used in conjunction with other patient information and based on their professional judgment to assist with triage/prioritization of medical images. Notified clinicians are responsible for viewing full images per the standard of care.

BriefCase-Triage: CARE Multi-Triage CT for Pneumothorax; Pericardial effusion; Large aortic aneurysm; Shoulder fracture or dislocation device is a radiological computer-assisted triage and notification software device. The software is based on an algorithm programmed component and is intended to run on a linux-based server in a cloud environment.

The BriefCase-Triage device receives images that match meta-data criteria according to the BriefCase-Triage predefined set of parameters. Then, the BriefCase-Triage processes the series chronologically, identifying cases with suspected positive finding(s) and selecting key slice(s) for preview. BriefCase-Triage output consists of suspected positive flag/notification regarding the existence of each finding in the analyzed study. Each finding includes a Representative Key Slice. The Key Slice(s) may be presented to the users as compressed, low-quality, grayscale, preview images with the date and time imprinted. The previews are not annotated and are captioned with the disclaimer "Not for diagnostic use, for prioritization only" according to the device requirement from the Image Communication Platform (ICP).

Presenting the users with worklist prioritization facilitates efficient triage by prompting the user to assess the relevant original images in the PACS. Thus, the suspect case receives attention earlier than would have been the case in the standard of care practice alone.

Here's a breakdown of the acceptance criteria and the study proving device performance, based on the provided FDA 510(k) clearance letter:

1. Acceptance Criteria and Reported Device Performance

The core acceptance criteria are based on standalone performance metrics for each of the four clinical indications.

Study Proving Device Meets Acceptance Criteria

The study conducted was a retrospective, blinded, multicenter standalone performance analysis.

2. Sample size used for the test set and the data provenance:
* Sample Size: N = 280 for each of the 4 clinical indications, totaling 772 unique scans across all indications.
* Data Provenance: The cases were collected from 6 US-based clinical sites, representing diverse geographic locations and site types. The data was "distinct in time or center from the cases used to train the algorithm," and "sequestered from algorithm development activities." This indicates a high level of independence for the test set. The data is retrospective.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
* Number of Experts: Three (3)
* Qualifications: Senior board-certified radiologists.

4. Adjudication method (e.g., 2+1, 3+1, none) for the test set:
* The document states ground truth was "determined by three senior board-certified radiologists." It does not explicitly mention an adjudication method like 2+1 or 3+1, but the plural "radiologists" and the method of "determined by" suggests a consensus or majority opinion among these three, rather than individual opinions without interaction.

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:
* No MRMC comparative effectiveness study was explicitly described. The study was a "standalone performance analysis" of the software itself. The comparison of "time-to-notification" with the predicate device implies a comparison of software performance characteristics related to triage, not a study of human readers with and without AI assistance.

6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:
* Yes, a standalone performance study was done. The document explicitly refers to it as a "standalone performance analysis" to "evaluate the software's performance."

7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.):
* Expert Consensus: The ground truth was established by "three senior board-certified radiologists."

8. The sample size for the training set:
* The document does not specify the exact sample size for the training set. It only mentions that the "algorithm was trained during software development on images of the pathology."

9. How the ground truth for the training set was established:
* The ground truth for the training set was established by "labeled ('tagged') images. In that process, each image in the training dataset was tagged based on the presence of the critical finding." The method or type of tagging (e.g., by experts, automated, etc.) is not detailed, but it's implied that there was a process of assigning labels/tags to the images to indicate the presence or absence of the target pathologies.

Ask a specific question about this device

BriefCase-Triage is a radiological computer aided triage and notification software indicated for use in the analysis of contrast-enhanced CT images that include the brain, in adults or transitional adolescents aged 18 and older. The device is intended to assist hospital networks and appropriately trained medical specialists in workflow triage by flagging and communication of suspected positive cases of Brain Aneurysm (BA) findings that are 3.0 mm or larger.

BriefCase-Triage uses an artificial intelligence algorithm to analyze images and flag suspect cases in parallel to the ongoing standard of care image interpretation. The user is presented with notifications for suspect cases. Notifications include compressed preview images that are meant for informational purposes only and not intended for diagnostic use beyond notification. The device does not alter the original medical image and is not intended to be used as a diagnostic device.

The results of BriefCase-Triage are intended to be used in conjunction with other patient information and based on professional judgment, to assist with triage/prioritization of medical images. Notified clinicians are responsible for viewing full images per the standard of care.

BriefCase-Triage is a radiological computer-assisted triage and notification software device.

The software is based on an algorithm programmed component and is intended to run on a linux-based server in a cloud environment.

The BriefCase-Triage receives filtered DICOM Images, and processes them chronologically by running the algorithms on each series to detect suspected cases. Following the AI processing, the output of the algorithm analysis is transferred to an image review software (desktop application). When a suspected case is detected, the user receives a pop-up notification and is presented with a compressed, low-quality, grayscale image that is captioned "not for diagnostic use, for prioritization only" which is displayed as a preview function. This preview is meant for informational purposes only, does not contain any marking of the findings, and is not intended for primary diagnosis beyond notification.

Here's a breakdown of the acceptance criteria and study details for the BriefCase-Triage device, based on the provided FDA 510(k) clearance letter:

Acceptance Criteria and Reported Device Performance

Note on Additional Operating Points (AOPs): The device also met performance goals (80% sensitivity and specificity) for three additional operating points (AOP1, AOP2, AOP3) with slightly varying sensitivity/specificity trade-offs (e.g., AOP3: Sensitivity 86.2%, Specificity 93.6%).

Study Details

1. Sample size used for the test set and the data provenance:

• Sample Size: 544 cases
• Data Provenance: Retrospective, blinded, multicenter study from 6 US-based clinical sites. The cases were distinct in time or center from those used for algorithm training.

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

• Number of Experts: Three (3) senior board-certified radiologists.
• Qualifications: "Senior board-certified radiologists." (Specific number of years of experience not detailed in the provided text).

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

• The text states the ground truth was "determined by three senior board-certified radiologists." It doesn't explicitly describe an adjudication method like "2+1" or "3+1." This implies a consensus approach where all three radiologists agreed, or a majority rule, but the exact mechanism for resolving discrepancies (if any) is not specified.

4. 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:

• No, an MRMC comparative effectiveness study was NOT done. The study's primary objective was to evaluate the standalone performance of the BriefCase-Triage software. The secondary endpoint compared the device's time-to-notification to that of the predicate device, but not its impact on human reader performance.

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

• Yes, a standalone performance study was done. The primary endpoints (sensitivity and specificity) measure the algorithm's performance in identifying Brain Aneurysm (BA) findings.

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

• Expert Consensus: The ground truth was "determined by three senior board-certified radiologists."

7. The sample size for the training set:

• Not explicitly stated. The document mentions the algorithm was "trained during software development on images of the pathology" and that "critical findings were tagged in all CTs in the training data set." However, the specific sample size for this training data is not provided.

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

• Manually labeled ("tagged") images: The text states, "As is customary in the field of machine learning, deep learning algorithm development consisted of training on manually labeled ('tagged') images. In that process, critical findings were tagged in all CTs in the training data set." It does not specify who performed the tagging or their qualifications, nor the method of consensus if multiple taggers were involved.

Ask a specific question about this device

BriefCase-Triage: CARE (Clinical AI Reasoning Engine) Multi-Triage CT Body is a radiological computer aided triage and notification software indicated for use in the analysis of contrast and non-contrast CT images of the chest, abdomen, and/or pelvis, in adults or transitional adolescents aged 18 and older. The device is intended to assist hospital networks and appropriately trained medical specialists in workflow triage by flagging and communicating suspected positive findings, per study, of:

1. Diverticulitis;
2. Abdominal-pelvic abscess;
3. Appendicitis;
4. Intestinal ischemia and/or pneumatosis;
5. Obstructive renal stone;
6. Small bowel obstruction;
7. Large bowel obstruction;
8. Spleen injury;
9. Liver injury;
10. Kidney injury;
11. Pelvic fracture.

The device flags cases with at least one suspected finding to assist with triage/prioritization of medical images. The device will provide a flag for each suspected finding within this study. A preview image will be provided for each distinct suspected finding.

BriefCase-Triage uses a foundation model-based artificial intelligence (AI) system to analyze images and highlight cases with detected findings in parallel to the ongoing standard of care image interpretation. The user is presented with notifications for cases with suspected findings. Notifications include compressed preview images for each suspected finding that are meant for informational purposes only and not intended for diagnostic use beyond notification. The device does not alter the original medical images and is not intended to be used as a diagnostic device.

The results of BriefCase-Triage are intended to be used in conjunction with other patient information and based on their professional judgment to assist with triage/prioritization of medical images. Notified clinicians are responsible for viewing full images per the standard of care.

Briefcase-Triage is a radiological computer-assisted triage and notification software device. The software is based on an algorithm programmed component and is intended to run on a linux-based server in a cloud environment.

The BriefCase-Triage receives images that match meta-data criteria according to BriefCase-Triage: CARE Multi-Triage CT Body's predefined set of parameters. Then, the BriefCase-Triage processes the series chronologically, identifying cases with suspected positive finding(s) and selecting key slice(s) for preview. BriefCase-Triage output consists of suspected positive flag/notification regarding the existence of each finding in the analyzed study. Each finding includes a Representative Key Slice. The Key Slice(s) may be presented to the users as compressed, low-quality, grayscale, preview images with the date and time imprinted. The previews are not annotated and are captioned with the disclaimer "Not for diagnostic use, for prioritization only" according to the device requirement from the Image Communication Platform (ICP).

Acceptance Criteria and Study Details for BriefCase-Triage: CARE Multi-triage CT Body

1. Table of Acceptance Criteria and Reported Device Performance

The acceptance criteria for the BriefCase-Triage: CARE Multi-triage CT Body device were primarily defined by performance goals for Area Under the Curve (AUC), Sensitivity (Se), and Specificity (Sp). The study demonstrated that the device met and exceeded these criteria for all 11 indications.

Note: The reported device performance for AUC, Sensitivity, and Specificity are ranges covering the minimum and maximum values observed across the 11 indications in the pivotal study. Detailed values for each indication are provided in the source text.

2. Sample Size and Data Provenance for the Test Set

• Sample Size: N = 280 for each of the 11 clinical indications, resulting in 1769 unique scans included across all device indications.
• Data Provenance: The data was collected from 6 US-based clinical sites. It was retrospective and the cases were distinct in time or center from the cases used to train the algorithm.

3. Number of Experts and Qualifications for Ground Truth

• Number of Experts: Three senior board-certified radiologists.
• Qualifications: The document specifically states "senior board-certified radiologists." No further details on years of experience were provided.

4. Adjudication Method for the Test Set

The adjudication method used to establish ground truth was based on the "consensus" of the three senior board-certified radiologists ("as determined by three senior board-certified radiologists"). This implies a consensus-based adjudication, but the specific mechanics (e.g., majority vote like 2+1, or requiring all three to agree) are not explicitly detailed.

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

No MRMC study comparing human readers with AI assistance versus without AI assistance was reported in this document. The study described is a standalone performance analysis of the algorithm.

6. Standalone Performance Study

Yes, a standalone performance study was done. The document states: "Aidoc conducted a retrospective, blinded, multicenter study with the Briefcase-Triage software to evaluate the standalone performance analysis individually for each of the 11 clinical indications supported by BriefCase-Triage: CARE Multi-triage CT Body device."

7. Type of Ground Truth Used

The ground truth was established by expert consensus of three senior board-certified radiologists.

8. Sample Size for the Training Set

The sample size for the training set is not explicitly provided in the given text. It is only mentioned that "the algorithm was trained during software development on images of the pathology."

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

The ground truth for the training set was established through labeled ("tagged") images. The document states: "As is customary in the field of machine learning, deep learning algorithm development consisted of training on labeled ("tagged") images. In that process, each image in the training dataset was tagged based on the presence of the critical finding." The specific method or expert involvement in this tagging process is not detailed, but it implies human expert labeling.

Ask a specific question about this device

The MedCognetics CogNet AI-MT+ software is a passive notification for prioritization-only, parallel-workflow software tool used by MQSA qualified interpreting physicians to prioritize viewing patients with suspicious findings in the medical care environment. CogNet AI-MT+ utilizes an artificial intelligence algorithm to analyze DBT screening mammograms and flags those that are suggestive of the presence of at least one suspicious finding at the exam level.

CogNet AI-MT+ produces an exam level output to a PACS/Workstation for flagging the suspicious study and allows for worklist prioritization. MQSA qualified interpreting physicians are responsible for reviewing each exam on a display approved for use in mammography, according to the current standard of care. The CogNet AI-MT+ device is limited to the categorization of exams, does not provide any diagnostic information beyond triage and prioritization, does not remove images from the interpreting physician's worklist, and should not be used in lieu of full patient evaluation, or relied upon to make or confirm diagnosis.

The CogNet AI-MT+ device is intended for use with DBT mammography exams acquired using validated DBT equipment only.

The MedCognetics CogNet AI-MT+ is a non-invasive computer-assisted triage and notification software as a medical device (SaMD) that analyzes DBT screening mammograms using a machine learning algorithm and notifies a PACS/workstation of the presence of findings suspicious of cancer in a study. The passive-notification enables interpreting physicians to prioritize their worklist and assists them in viewing prioritized studies using the standard PACS or workstation viewing software. The device aim is to aid in the prioritization and triage of radiological medical images only. It is a software tool for MQSA interpreting physicians reading mammograms and does not replace complete evaluation according to the standard of care.

The software modules that compose the CogNet AI-MT+ Deep Learning software are:

The Qualification Module - The requirement for acceptance into the CogNet AI-MT+ analysis is a completed Mammogram DICOM image. In the Qualification Module, the image arrives from the Mammogram modality and is "read" to determine if this qualification applies.

The Mammogram Pre-Processing module – The DBT pixel brightness, image size, and shape is adjusted for consistency in this module. After the DICOM image has been qualified, the Pre-Processing module assures that the images are from a mammogram device and then validates that the DICOM is properly formed and consists of "For Presentation" image pixel data.

Mammogram Learning Module – This module accepts the normalized image data from the pre-processing module and uses Deep Learning techniques to extract features to determine if any lesions suspicious for cancer exist in the mammogram study

Failures in any of the above modules will generate error messages that are recorded in an accessible log file and, if user specific issues are encountered, sent to the user in a secondary capture report.

CogNet AI-MT+ has no viewing capability, but the results data are sent via a secure network function to the PACS/workstation, and the PACS/workstation "reads" the necessary DICOM tags and matches it with the original mammogram study images as a normal function of a PACS or Workstation.

When the study data is fed into the configured reading worklist, the results are merged as part of the mammogram study. This process allows an AI Result to be ready for prioritization of the study prior to the interpreting physician's review.

A reading worklist is a listing of available studies for reading and diagnosis. The worklist is populated by the parsing of a DICOM file of a completed mammogram study, using the demographic and study fields to fill in the designated columns of the worklist. The columns are sortable by study, based on the column headings. CogNet AI-MT+ provides an API for adding an AI Results column with 0 to 1 response per study. If an analysis was not performed on that study, the AI Results indication is 0. If an analysis was performed on that study, then the AI Results column indicates either Suspicious (red diamond icon) or Processed (blue circle icon). The AI Results column may be sorted by the interpreting physicians by clicking an up or down arrow next to the column heading. This sort would allow the studies that contain suspicious findings to be brought to the top of the viewing list.

Here's a detailed description of the acceptance criteria and the study proving the device meets them, based on the provided FDA 510(k) clearance letter:

Acceptance Criteria and Device Performance

1. Table of Acceptance Criteria and Reported Device Performance

The reported device performance for CogNet AI-MT+ met or exceeded both the primary AUROC objective and the secondary sensitivity and specificity objectives.

Study Details Proving Device Meets Acceptance Criteria

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

• Sample Size for Test Set: 806 women (patients).
  • This consisted of 403 cases labeled "benign" (negative diagnosis with 2-year follow-up) and 403 cases labeled "malignant" (positive biopsy result).
  • The breakdown of these 806 samples by biopsy outcome is:
    • Biopsy Proven Benign: 21
    • Malignant: 403
    • Screening Benign: 382
• Data Provenance: The test set data was obtained from a site or facility that was not used to source the training or development data, to ensure generalizability. The specific country of origin is not explicitly stated, but it is implied to be distinct from the diverse geographical regions listed for training data (Europe, South Asia, South America, Africa, United States). It is a retrospective study.

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

The document does not explicitly state the number of experts used or their qualifications for establishing the ground truth for the test set. However, the ground truth was based on:

• "negative diagnosis (BI-RADS 1 or 2 assessment) throughout 2-years of follow-up" for benign cases.
• "positive biopsy result" for malignant cases.
  This implies clinical follow-up and pathology reports, which are inherently established by qualified medical professionals, but not specifically "experts" in the context of reader studies.

4. Adjudication method for the test set

The document does not explicitly describe an adjudication method (e.g., 2+1, 3+1) for the test set. The ground truth appears to be based on definitive clinical outcomes such as biopsy results and 2-year follow-up on BI-RADS classifications, which typically do not require an adjudication process among readers for establishing the final ground truth label itself.

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

• A Multi-Reader Multi-Case (MRMC) comparative effectiveness study was not done.
• The study conducted was a standalone retrospective study of the device performance (algorithm only).

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

• Yes, a standalone retrospective study of device performance was conducted. The results (AUROC, Sensitivity, Specificity) listed in the table above reflect the algorithm's performance without human-in-the-loop assistance.

7. The type of ground truth used

The ground truth used was:

• Outcomes data / Clinical Follow-up: "negative diagnosis (BI-RADS 1 or 2 assessment) throughout 2-years of follow-up" for benign cases.
• Pathology: "positive biopsy result" for malignant cases.

8. The sample size for the training set

• Total Patients for Training Set: 32,292 patients.
  • This corresponds to approximately 129,168 images (assuming 4 images per patient for bilateral studies with 4 standard views, as mentioned in inclusion criteria).
  • The breakdown was 10,496 positive cases and 21,796 negative cases.

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

The document states that the algorithm was "trained with samples both suspicious of cancer and not suspicious of cancer." While the precise method for establishing ground truth for each individual training sample is not detailed, it can be inferred that it followed similar clinical standards as the test set:

• "Biopsy proven cancer studies (soft tissues and microcalcifications)" for positive cases.
• "BIRADS 1 and 2 normal/benign cases with 2-year follow-up of a negative diagnosis" for negative cases.
  These methods would involve pathology reports and clinical follow-up, similar to the test set ground truth.

Ask a specific question about this device

Ask a specific question about this device

a2z-Unified-Triage is a radiological computer-aided triage and notification software indicated for use in the analysis of abdominal/pelvic CT images in adults aged 22 and older. The device is intended to assist hospital networks and appropriately trained medical specialists in workflow triage by flagging and communicating suspected positive cases of the 7 specified abdominopelvic findings: Acute Cholecystitis, Acute Pancreatitis, Unruptured Abdominal Aortic Aneurysm, Acute Diverticulitis, Free Air, Hydronephrosis, and Small Bowel Obstruction. These findings are intended to be used together as one device. The device supports both cloud-based and on-premises deployment, with integration either directly with healthcare facility systems or through third-party healthcare technology platforms.

a2z-Unified-Triage uses an artificial intelligence algorithm to analyze images and flag cases with detected findings in parallel to the ongoing standard of care image interpretation. The device provides analysis results that enable client systems to generate notifications for cases with suspected findings. These results can include DICOM instance UIDs for key images, which are meant for informational purposes only and not intended for primary diagnosis beyond notification. The device does not alter the original medical image and is not intended to be used as a diagnostic device.

The results of a2z-Unified-Triage are intended to be used in conjunction with other patient information and based on clinicians' professional judgment, to assist with triage/prioritization of medical images. Notified clinicians are responsible for viewing full images per the standard of care.

a2z-Unified-Triage is a radiological computer-assisted triage and notification software device. The software consists of an algorithmic component that supports both cloud-based and on-premises deployment on standard server hardware. The device processes abdomen/pelvis CT images from clinical imaging systems, analyzing them using artificial intelligence algorithms to detect suspected cases of 7 abdominopelvic conditions: Acute Cholecystitis, Acute Pancreatitis, Unruptured Abdominal Aortic Aneurysm, Acute Diverticulitis, Free Air, Hydronephrosis, and Small Bowel Obstruction.

Following the AI processing, the analysis results are returned to the client system for worklist prioritization. When a suspected case is detected, the software provides analysis results that enable the client system to generate appropriate notifications. These results can include DICOM instance UIDs for key images, which are for informational purposes only, do not contain any marking of the findings, and are not intended for primary diagnosis beyond notification.

Integration with clinical imaging systems facilitates efficient triage by enabling prioritization of suspect cases for review of the relevant original images in the PACS. Thus, the suspect case receives attention earlier than would have been the case in the standard of care practice alone.

Here's a detailed summary of the acceptance criteria and the study proving the device meets them, based on the provided FDA clearance letter:

Acceptance Criteria and Device Performance

1. Table of Acceptance Criteria and Reported Device Performance

a2z-Unified-Triage differentiates between two types of findings for regulatory purposes: QAS (Qualitative, Automated, and Subjective) and QFM (Quantitative, Functional, and Measurable).

Turnaround Time Acceptance Criteria and Performance:

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

• Test Set Sample Size: 675 cases from 643 unique patients (after excluding 3 cases due to quality control failures from an initial 678 cases).
• Data Provenance: The data was sourced from multiple clinical sites within the United States. Specific states mentioned are New York (45.2%), Kansas (21.2%), Missouri (18.4%), Texas (15.0%), and Nebraska (0.3%). The study evaluated against clinical standards consistent with U.S. practice patterns. The data appears to be retrospective, as it was used for development and testing after collection.

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

• Number of Experts: A minimum of two U.S. board-certified radiologists, with a third U.S. board-certified expert adjudicator for discordant cases.
• Qualifications: All experts were U.S. board-certified radiologists. The third adjudicator was specifically fellowship-trained in body imaging.

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

• Adjudication Method: 2+1 methodology. Each case was independently reviewed by two U.S. board-certified radiologists. If the two initial readers disagreed, a third U.S. board-certified expert adjudicator (fellowship-trained in body imaging) provided the tie-breaking determination.

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

• The provided document does not indicate that an MRMC comparative effectiveness study was performed or submitted for this clearance. The study described is a standalone performance assessment of the algorithm itself against ground truth.

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

• Yes, a standalone performance assessment was done. The document explicitly states: "A standalone performance assessment was performed for a2z-Unified-Triage to validate the accuracy of detecting the 7 findings against a reference standard established by U.S. board-certified radiologists."

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

• Type of Ground Truth: Expert consensus, specifically a 2+1 consensus of U.S. board-certified radiologists, with the third adjudicator being fellowship-trained in body imaging.

8. The sample size for the training set

• The document states, "The algorithms were developed on an extensive dataset of abdomen/pelvis CT studies from multiple clinical sites." However, a specific numerical sample size for the training set is not provided. It only mentions that strict protocols ensured complete independence between development and testing datasets (mutually exclusive patients).

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

• The document does not explicitly detail how the ground truth for the training set was established. It only describes the ground truth establishment for the test set (2+1 radiologist consensus). It states that the algorithms were developed on an "extensive dataset" and implies internal processes for data collection and annotation during development.

Ask a specific question about this device

BriefCase-Triage is a radiological computer aided triage and notification software indicated for use in the analysis of abdominal CT images in adults or transitional adolescents aged 18 and older. The device is intended to assist hospital networks and appropriately trained medical specialists in workflow triage by flagging and communication of suspected positive findings of Intra-abdominal free gas (IFG) pathologies.

BriefCase-Triage uses an artificial intelligence algorithm to analyze images and highlight cases with the detected findings in parallel to the ongoing standard of care image interpretation. The user is presented with notifications for cases with suspected findings. Notifications include compressed preview images that are meant for informational purposes only and not intended for diagnostic use beyond notification. The device does not alter the original medical image and is not intended to be used as a diagnostic device.

The results of BriefCase-Triage are intended to be used in conjunction with other patient information and based on their professional judgment, to assist with triage/prioritization of medical images. Notified clinicians are responsible for viewing full images per the standard of care.

Briefcase-Triage is a radiological computer-assisted triage and notification software device.

The software is based on an algorithm programmed component and is intended to run on a linux-based server in a cloud environment.

The Briefcase-Triage receives filtered DICOM Images, and processes them chronologically by running the algorithms on each series to detect suspected cases. Following the AI processing, the output of the algorithm analysis is transferred to an image review software (desktop application). When a suspected case is detected, the user receives a pop-up notification and is presented with a compressed, low-quality, grayscale image that is captioned "not for diagnostic use, for prioritization only" which is displayed as a preview function. This preview is meant for informational purposes only, does not contain any marking of the findings, and is not intended for primary diagnosis beyond notification.

Presenting the users with worklist prioritization facilitates efficient triage by prompting the user to assess the relevant original images in the PACS. Thus, the suspect case receives attention earlier than would have been the case in the standard of care practice alone.

The algorithm was trained during software development on images of the pathology. As is customary in the field of machine learning, deep learning algorithm development consisted of training on labeled ("tagged") images. In that process, each image in the training dataset was tagged based on the presence of the critical finding.

Here's a detailed breakdown of the acceptance criteria and the study that proves the device meets them, based on the provided FDA 510(k) clearance letter for BriefCase-Triage:

Acceptance Criteria and Reported Device Performance

Note: The document explicitly states that the primary endpoints were "sensitivity and specificity with an 80% performance goal." The reported performance for both sensitivity and specificity (94.2% and 94.6% respectively) significantly exceeds this 80% goal. The time-to-notification for the subject device is significantly faster than the predicate, demonstrating improved "time savings to the standard of care review."

Study Information

1. Sample Size Used for the Test Set and Data Provenance:
* Sample Size: 394 cases
* Data Provenance:
* Country of Origin: US (6 clinical sites)
* Retrospective/Prospective: Retrospective
* Additional Detail: Cases were distinct in time or center from the training data.

2. Number of Experts Used to Establish Ground Truth for the Test Set and Qualifications:
* Number of Experts: 3
* Qualifications: Senior board-certified radiologists

3. Adjudication Method for the Test Set:
* The document states "as determined by three senior board-certified radiologists." While it doesn't explicitly state "2+1" or "3+1," this implies a consensus-based approach among the three experts. Without further detail, it's reasonable to infer a consensus was reached, or a specific rule for disagreement (e.g., majority) was applied.

4. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study:
* No, a multi-reader multi-case (MRMC) comparative effectiveness study was not conducted to assess how much human readers improve with AI vs. without AI assistance. The study focuses purely on the standalone performance of the AI algorithm.

5. Standalone Performance Study (Algorithm Only):
* Yes, a standalone study was performed. The "Pivotal Study Summary" describes evaluating "the software's performance to the ground truth," indicating a standalone performance assessment of the algorithm without human-in-the-loop performance measurement.

6. Type of Ground Truth Used:
* Expert consensus (as determined by three senior board-certified radiologists).

7. Sample Size for the Training Set:
* The document states, "The algorithm was trained during software development on images of the pathology." However, it does not provide a specific sample size for the training set.

8. How the Ground Truth for the Training Set Was Established:
* "each image in the training dataset was tagged based on the presence of the critical finding." This indicates that human experts (or a similar method to the test set ground truth) labeled the images in the training set for the presence of the pathology. However, the specific number and qualifications of these experts are not explicitly stated for the training set.

Ask a specific question about this device

qER-CTA is a notification-only, parallel workflow tool for use by hospital networks and trained clinicians to identify and communicate images of specific patients to a specialist, independent of the standard of care workflow. qER-CTA uses a deep learning algorithm to analyze images for findings suggestive of a pre-specified clinical condition and to notify an appropriate medical specialist in parallel to standard of care image interpretation. Identification of suspected findings is not for diagnostic use beyond notification. Specifically, the device analyses CT angiogram images of the brain acquired in the acute setting and sends notifications to a neurovascular specialist that a suspected large vessel occlusion has been identified, recommending review of those images. Images can be previewed through a mobile application. qER-CTA is intended to analyze the internal carotid artery (ICA) and M1 segment of the middle cerebral artery (MCA) for LVOs on CTA scans of adults (≥ 22 years of age). Images previewed through the mobile application are compressed and for informational purposes only, not intended for diagnostic use beyond notification. Notified clinicians are responsible for viewing non-compressed images on a diagnostic viewer, conducting appropriate patient evaluation, and engaging in relevant discussions with the treating physician before making care-related decisions or requests. qER-CTA is limited to the analysis of imaging data and should not be used as a substitute for full patient evaluation or relied upon to make or confirm a diagnosis.

qER-CTA is a radiological computer-aided triage and notification (CADt) software designed to assist trained clinicians and radiologists in analyzing and triaging head CTA scans for suspected LVO (Large Vessel Occlusion) in the anterior circulation.

The software uses a deep learning algorithm to analyze CTA images and provide a case-level output available in the PACS or workstation for worklist prioritization or triage. It does not alter the original image, change the worklist order, or send proactive alerts directly to the end user. Instead, the end user can sort the worklist based on the passive notification flag. Images can be previewed through a mobile application also. There are two alternatives' users can choose from engaging with qER-CTA.

1. For de-identified CTA scans, they are sent to qER-CTA via transmission functions built within the user's PACS or workstation. Results are pushed back to the user's PACS or other user-specified radiology software database once the processing is complete.

2. For the client system that does not have de-identification and re-identification capabilities, qER-CTA interacts with on-premises gateway rather than directly with the PACS.

qER-CTA is not intended to direct attention to specific portions of the image, rule out target conditions, or be used as a standalone tool for clinical decision-making. It operates as a parallel workflow tool, independent of the standard of care, to assist in identifying and communicating suspected LVO cases to appropriate medical specialists for further review. Images previewed through the mobile application are compressed and are for informational purposes only.

1. Acceptance Criteria and Reported Device Performance

2. Sample Size for Test Set and Data Provenance

• Sample Size: 584 head CTA scans (289 LVO, 295 non-LVO).
• Data Provenance: Not explicitly stated in the provided text (e.g., country of origin, retrospective or prospective).

3. Number of Experts and Qualifications for Ground Truth Establishment (Test Set)

• Number of Experts: Three.
• Qualifications: U.S. board-certified neuroradiologists with at least 10 years of experience.

4. Adjudication Method for the Test Set

• The adjudication method is not explicitly mentioned. It states that "Three U.S. board certified neuroradiologists with at least 10 years of experience did the ground truthing," implying a consensus or majority vote might have been used, but specific details (e.g., 2+1, 3+1) are not provided.

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

• No, a MRMC comparative effectiveness study was not done. The document explicitly states that the performance was assessed using a "standalone study."

6. Standalone Performance (Algorithm Only) Study

• Yes, a standalone study was done. The performance of the qER-CTA device was assessed using a standalone study, evaluating its classification of large vessel occlusion.

7. Type of Ground Truth Used

• Expert Consensus: The ground truth was established by three U.S. board-certified neuroradiologists with at least 10 years of experience.

8. Sample Size for the Training Set

• The sample size for the training set is not provided in the given text.

9. How Ground Truth for the Training Set Was Established

• How the ground truth for the training set was established is not provided in the given text. The document only mentions the ground truthing for the clinical performance testing (test set).

Ask a specific question about this device