Search Results

Ask a specific question about this device

Ask a specific question about this device

AIR Recon DL is a deep learning based reconstruction technique that is available for use on GE HealthCare 1.5T, 3.0T, and 7.0T MR systems. AIR Recon DL reduces noise and ringing (truncation artifacts) in MR images, which can be used to reduce scan time and improve image quality. AIR Recon DL is intended for use with all anatomies, and for patients of all ages. Depending on the anatomy of interest being imaged, contrast agents may be used.

AIR Recon DL is a software feature intended for use with GE HealthCare MR systems. It is a deep learning-based reconstruction technique that removes noise and ringing (truncation) artifacts from MR images. AIR Recon DL is an optional feature that is integrated into the MR system software and activated through purchasable software option keys. AIR Recon DL has been previously cleared for use with 2D Cartesian, 3D Cartesian, and PROPELLER imaging sequences.

The proposed device is a modified version of AIR Recon DL that includes a new deep-learning phase correction algorithm for applications that create multiple intermediate images and combine them, such as Diffusion Weighted Imaging where multiple NEX images are collected and combined. This enhancement is an optional feature that is integrated into the MR system software and activated through an additional purchasable software option key (separate from the software option keys of the predicate device).

This document describes the acceptance criteria and the studies conducted to prove the performance of the AIR Recon DL device, as presented in the FDA 510(k) clearance letter.

1. Table of Acceptance Criteria and Reported Device Performance

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

• Nonclinical Testing:
  • Phantom testing was conducted for the DLPC and PC-ARDL models. No specific sample size (number of phantom scans) is provided, but it implies a sufficient number for evaluation.
• In-Vivo Performance Testing:
  • ADC Accuracy: Diffusion-weighted brain images were acquired at 1.5T with b-values = 50, 400, 800, 1200 s/mm². The number of subjects is not explicitly stated, but it's referred to as "diffusion images" and "diffusion-weighted brain images."
  • Low-Contrast Detectability: Raw data from 4 diffusion-weighted brain scans were used.
• Quantitative Post Processing (Repeatability Study):
  • 6 volunteers were recruited. 2 volunteers scanned on a 1.5T scanner, 4 on a 3T scanner.
  • Scanned anatomical regions included brain, spine, abdomen, pelvis, and breast.
  • Each sequence was repeated 4 times.
  • Data Provenance: The document states "in-vivo data" and "volunteer scanning was performed simulating routine clinical workflows." This suggests prospective scanning of human subjects, likely in a controlled environment. The country of origin is not specified, but given the FDA submission, it's likely U.S. or international data meeting U.S. standards. The statement "previously acquired de-identified cases" for the Clinical Image Quality Study refers to retrospective data for that specific study, but the volunteer scanning for repeatability appears prospective.
• Clinical Image Quality Study:
  • 34 datasets of previously acquired de-identified cases.
  • Data Provenance: "previously acquired de-identified cases" indicates retrospective data. The country of origin is not specified.

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

• Nonclinical Testing: Ground truth established through phantom measurements and expected physical properties (e.g., signal bias, noise floor). No human experts involved in establishing ground truth here.
• In-Vivo Performance Testing:
  • ADC Accuracy: "Average ADC values were measured from regions of interest in the lateral ventricles." This implies expert selection of ROIs, but the number of experts is not specified. The ground truth for ADC is the expected isotropic Gaussian diffusion in these regions.
  • Low-Contrast Detectability: "The contrast ratio and contrast-to-noise ratio for each of the inserts were measured." This is a quantitative measure, not explicitly relying on expert consensus for ground truth on detectability, but rather on the known properties of the inserted synthetic objects.
• Quantitative Post Processing:
  • ADC Repeatability: Ground truth for repeatability is based on quantitative measurements and statistical analysis (coefficient of variability). ROI placement would typically be done by an expert, but the number is not specified.
  • Phase Correction Effectiveness: Ground truth is based on the theoretical expectation of signal distribution in real/imaginary channels after ideal phase correction.
• Clinical Image Quality Study:
  • One (1) U.S. Board Certified Radiologist was used.
  • Qualifications: "U.S. Board Certified Radiologist." No explicit number of years of experience is stated, but Board Certification indicates a high level of expertise.

4. Adjudication Method for the Test Set

• Nonclinical/Phantom Testing: No explicit adjudication method described beyond passing defined acceptance criteria for quantitative metrics.
• In-Vivo Performance Testing: Quantitative measurements (ADC values, contrast ratios, CNR) were used. Paired t-tests were conducted, which is a statistical comparison method, not an adjudication process as typically defined for expert readings.
• Quantitative Post Processing: Quantitative measurements and statistical analysis (coefficient of variability, comparison of real/imaginary channels).
• Clinical Image Quality Study: A single U.S. Board Certified Radiologist made the assessment. There is no stated adjudication method described, implying a single-reader assessment for clinical image quality.

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

• An MRMC comparative effectiveness study was not explicitly described as a formal study design in the provided text.
• The "Clinical Image Quality Study" involved only one radiologist, so it does not qualify as an MRMC study.
• There is no reported effect size of how much human readers improve with AI vs. without AI assistance. The study rather focused on the AI-reconstructed images' standalone diagnostic quality.

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

• Yes, performance was evaluated in a standalone manner.
  • Nonclinical Testing: Phantom studies directly evaluate the algorithm's output against known physical properties and defined metrics.
  • In-Vivo Performance Testing: ADC accuracy and low-contrast detectability were measured directly from the reconstructed images, which is a standalone evaluation of the algorithm's quantitative output.
  • Quantitative Post Processing: Repeatability and effectiveness of phase correction in real/imaginary channels are algorithm-centric evaluations.
  • Even the clinical image quality study, while involving a human reader, assessed the standalone output of the algorithm (AIR Recon DL with Phase Correction) for diagnostic quality.

7. Type of Ground Truth Used

• Expert Consensus: Not explicitly stated as the primary ground truth for quantitative metrics, but one radiologist's assessment served as the primary clinical ground truth.
• Pathology: Not used as ground truth in the provided study descriptions. While some datasets "included pathological features such as prostate cancer... hepatocellular carcinoma," the assessment by the radiologist was on "diagnostic quality" of the images, not a comparison against pathology reports for definitive disease identification.
• Outcomes Data: Not used as ground truth.
• Other:
  • Physical Properties/Known Standards: For phantom testing (e.g., signal bias, noise floor, SNR, sharpness), and for theoretical expectations of ADC values in specific regions (lateral ventricles).
  • Known Synthetic Inserts: For low-contrast detectability.
  • Theoretical Expectations: For phase correction effectiveness (signal in real, noise in imaginary).

8. Sample Size for the Training Set

• The document does not provide any specific sample size for the training set used for the deep learning models (DLPC and PC-ARDL). It only states that the models are "deep learning-based."

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

• The document does not provide any information on how the ground truth for the training set was established. It only describes the testing of the final, trained models.

Ask a specific question about this device

The angiographic X-ray systems are indicated for use for patients from newborn to geriatric in generating fluoroscopic and rotational images of human anatomy for cardiovascular, vascular and non-vascular, diagnostic and interventional procedures.

Additionally, with the OR table, the angiographic X-ray systems are indicated for use in generating fluoroscopic and rotational images of human anatomy for image-guided surgical procedures. The OR table is suitable for interventional and surgical procedures.

GE HealthCare interventional x-ray systems are designed to perform monoplane fluoroscopic X-ray examinations to provide the imaging information needed to perform minimally invasive interventional X-Ray imaging procedures. Additionally, with an OR table, these systems allow to perform surgery and X-Ray image guided surgical procedures in a hybrid Operating Room.

Allia™ Moveo is a GE HealthCare interventional X-Ray system product model. It consists of a C-arm positioner, an X-ray table, an X-ray tube assembly, an X-ray power unit with its exposure control unit, an X-ray imaging chain (including a digital detector and an image processing unit).

Allia™ Moveo is a monoplane system (C-arm with mobile AGV gantry), with a square 41cm digital detector and the InnovaIQ table (with an option to make it an OR table).

Allia™ Moveo is an image acquisition system requiring connection to the GE HealthCare Advantage Workstation (AW) for 3D reconstruction. When a 3D acquisition is performed on the Allia™ Moveo system, the acquired 2D images are transferred to the Advantage Workstation (AW) to be processed by 3DXR (reference device K243446) for 3D reconstruction.

The purpose of this Premarket Notification is the introduction of a new C-arm with a modified detector mount.

N/A

Ask a specific question about this device

The device is a general purpose ultrasound system intended for use by qualified and trained healthcare professionals. Specific clinical applications remain the same as previously cleared: Fetal/OB; Abdominal (including GYN, pelvic and infertility monitoring/follicle development); Pediatric; Small Organ (breast, testes, thyroid etc.); Neonatal and Adult Cephalic; Cardiac (adult and pediatric); Musculo-skeletal Conventional and Superficial; Vascular; Transvaginal (including GYN); Transrectal

Modes of operation include: B, M, PW Doppler, CW Doppler, Color Doppler, Color M Doppler, Power Doppler, Harmonic Imaging, Coded Pulse, 3D/4D Imaging mode, Elastography, Shear Wave Elastography and Combined modes: B/M, B/Color, B/PWD, B/Color/PWD, B/Power/ PWD, B/Elastography. The Voluson™ Expert 18, Voluson™ Expert 20, Voluson™ Expert 22 is intended to be used in a hospital or medical clinic.

The systems are full-featured Track 3 ultrasound systems, primarily for general radiology use and specialized for OB/GYN with particular features for real-time 3D/4D acquisition. They consist of a mobile console with keyboard control panel; color LCD/TFT touch panel, color video display and optional image storage and printing devices. They provide high performance ultrasound imaging and analysis and have comprehensive networking and DICOM capability. They utilize a variety of linear, curved linear, matrix phased array transducers including mechanical and electronic scanning transducers, which provide highly accurate real-time three-dimensional imaging supporting all standard acquisition modes.

The following probes are the same as the predicate: RIC5-9-D, IC5-9-D, RIC6-12-D, 9L-D, 11L-D, ML6-15-D, RAB6-D, C1-6-D, C2-9-D, M5Sc-D, RM7C, eM6CG3, RSP6-16-D , RIC10-D, 6S-D and L18-18iD, RIC12-D.

The existing cleared Probe C1-6-D is being added to previously cleared SW- AI Feature Sonolyst 1st Trimester.

The provided text describes the FDA 510(k) clearance for the Voluson Expert Series ultrasound systems, specifically focusing on the AI feature "Sonolyst 1st Trimester" and the addition of the C1-6-D transducer to this feature.

Here's an analysis of the acceptance criteria and the study that proves the device meets them:

1. Table of Acceptance Criteria and Reported Device Performance

Additional Performance Data (Mean values across transabdominal and transvaginal scans):

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

• SonoLyst 1st Trim IR: 7970 images

• SonoLyst 1st Trim X: 4931 images

• SonoLyst 1st Trim Live: 9111 images

• SonoBiometry CRL: 243 images

• Specific to Probegroup CL2 (which includes C1-6-D Probe): Data was collected from 396 patients.

• Data Provenance: Data was collected from multiple geographical sites including the UK, Austria, India, and USA. The data was collected using different systems (GE Voluson V730, P8, S6/S8, E6, E8, E10, Expert 22, Philips Epiq 7G).

• Retrospective/Prospective: The document does not explicitly state whether the test data was retrospective or prospective. However, the mention of "data acquired with transabdominal vs transvaginal probes" and "patients within the dataset includes pregnancies between 11 and 14 weeks of gestation, with no known fetal abnormalities at the time of imaging" suggests that the images were pre-existing or collected specifically for this evaluation, implying a retrospective or a pre-defined prospective collection for the study.

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

• Initial Curation: A single sonographer curated (sorted and graded) the images initially.
• Review Panel for Graded Images: Where the system's grading differed from the initial ground truth, the images were reviewed by a 5-sonographer review panel to determine the grading accuracy of the system.
• Qualifications: The document identifies them as "sonographers." Specific years of experience or expertise in fetal ultrasound are not provided, other than their role in image curation and review.

4. Adjudication Method for the Test Set

• Initial Sorting and Grading: Images were initially curated (sorted and graded) by a single sonographer.
• Reclassification during Sorting: The SonoLyst IR/X First Trimester process resulted in some images being reclassified during sorting based on the majority view of the panel (after the step where the system had sorted them).
• Grading Accuracy Review: For graded images where the initial single sonographer's ground truth differed from the system, a 5-sonographer review panel was used to determine the accuracy. This suggests an adjudication process where the panel formed a consensus or majority opinion to establish the final ground truth when discrepancies arose. The exact method (e.g., simple majority, weighted vote) is not specified beyond "majority view of the panel."

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

• The document does not indicate that a Multi-Reader Multi-Case (MRMC) comparative effectiveness study was done to evaluate how much human readers improve with AI vs. without AI assistance. The testing focused on the standalone performance of the AI algorithm against a ground truth established by sonographers.
• The verification of SonoLystLive 1st Trim Trimester features was based on the "average agreement between a sonographer panel and the output of the algorithm regarding Traffic light quality," which involves human readers assessing traffic light quality in relation to the algorithm's output, but it's not a study designed to measure human improvement with AI assistance.

6. If a Standalone (Algorithm Only Without Human-in-the-Loop Performance) Was Done

• Yes, a standalone performance evaluation was conducted. The performance metrics (SonoLystIR, SonoLystX, SonoLystLive, SonoBiometry CRL success rate) are reported as the accuracy of the algorithm comparing its output directly against the established ground truth. This is a measure of the algorithm's ability to perform its specified functions independently.

7. The Type of Ground Truth Used

• The ground truth was established through expert consensus/review by sonographers.
  • Initial curation by a single sonographer.
  • Review and reclassification during sorting based on the "majority view of the panel."
  • A 5-sonographer review panel was used to determine grading accuracy for discrepancies.
• The ground truth also adhered to standardized imaging protocols based on internationally recognized guidelines (AIUM Practice Parameter, AIUM Detailed Protocol, ISUOG Practice Guidelines, ISUOG Detailed Protocol, and the study by Yimei Liao et al.) which informed the quality and consistency of the expert review.

8. The Sample Size for the Training Set

• 122,711 labelled source images from 35,861 patients were used for training.

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

• The document states that "Data used for both training and validation has been collected across multiple geographical sites using different systems to represent the variations in target population."
• While the specific method for establishing ground truth for the training set is not explicitly detailed in the same way as the test set, it can be inferred that similar expert labeling and curation processes would have been applied given the emphasis on "labelled source images." The document focuses on the test set truthing process as part of verification, implying that the training data would have undergone a robust labeling process to ensure quality for machine learning.

Ask a specific question about this device

LOGIQ E10 is intended for use by a qualified physician for ultrasound evaluation of Fetal/Obstetrics; Abdominal (including Renal, Gynecology/Pelvic); Pediatric; Small Organ (Breast, Testes, Thyroid); Neonatal Cephalic, Adult Cephalic; Cardiac (Adult and Pediatric); Peripheral Vascular; Musculo-skeletal Conventional and Superficial; Urology (including Prostate); Transrectal; Transvaginal; Tranesophageal and Intraoperative (Abdominal and Vascular).

Modes of operation include: B, M, PW Doppler, CW Doppler, Color Doppler, Color M Doppler, Power Doppler, Harmonic Imaging, Coded Pulse, 3D/4D Imaging mode, Elastography, Shear Wave Elastography, Attenuation Imaging and combined modes: B/M, B/Color, B/PWD, B/Color/PWD, B/Power/PWD.

The LOGIQ E10 is intended to be used in a hospital or medical clinic.

The LOGIQ E10 is a full featured, Track 3, general purpose diagnostic ultrasound system which consists of a mobile console approximately 585 mm wide (keyboard), 991 mm deep and 1300 mm high that provides digital acquisition, processing and display capability. The user interface includes a computer keyboard, specialized controls, 12-inch high-resolution color touch screen and 23.8-inch High Contrast LED LCD monitor.

Here's an analysis of the acceptance criteria and supporting studies for the LOGIQ E10 ultrasound system, derived from the provided FDA 510(k) Clearance Letter:

1. Table of Acceptance Criteria and Reported Device Performance

2. Sample Size for Test Set and Data Provenance

• Auto Abdominal Color Assistant 2.0:
  • Test Set Sample Size: 49 individual subjects, 1186 annotation images.
  • Data Provenance: Retrospective, all data from the USA.
• Auto Aorta Measure Assistant:
  • Test Set Sample Size:
    • Long View Aorta: 36 subjects (11 Male, 25 Female).
    • Short View Aorta: 35 subjects (11 Male, 24 Female).
  • Data Provenance: Retrospective, from Japan (15-16 subjects) and USA (20 subjects).
• Auto Common Bile Duct (CBD) Measure Assistant:
  • Test Set Sample Size: 25 subjects (11 Male, 14 Female).
  • Data Provenance: Retrospective, from USA (40%) and Japan (60%).
• Ultrasound Guided Fat Fraction (UGFF):
  • Test Set Sample Size (Primary Study): 582 participants.
  • Data Provenance (Primary Study): Retrospective, Japan.
  • Test Set Sample Size (Confirmatory Study 1): 15 US patients + 5 EU patients (total 20).
  • Data Provenance (Confirmatory Study 1): Retrospective, USA and EU.
  • Test Set Sample Size (Confirmatory Study 2): 24 EU patients.
  • Data Provenance (Confirmatory Study 2): Retrospective, EU.

3. Number of Experts and Qualifications for Ground Truth

• Auto Abdominal Color Assistant 2.0: Not explicitly stated, but implies multiple "readers" to ground truth anatomical visibility. No specific qualifications are mentioned beyond "readers."
• Auto Aorta Measure Assistant: Not explicitly stated, but implies multiple "readers" for measurements and an "arbitrator" to select the most accurate measurement. No specific qualifications are mentioned beyond "readers" and "arbitrator."
• Auto Common Bile Duct (CBD) Measure Assistant: Not explicitly stated, but implies multiple "readers" for measurements and an "arbitrator" to select the most accurate measurement. No specific qualifications are mentioned beyond "readers" and "arbitrator."
• Ultrasound Guided Fat Fraction (UGFF): Ground truth for the primary study was MRI Proton Density Fat Fraction (MRI-PDFF %). No human experts were involved in establishing the ground truth for UGFF, as it relies on MRI-PDFF as the reference. The correlation between UGFF and UDFF also used UDFF as a reference, not human experts.

4. Adjudication Method for the Test Set

• Auto Abdominal Color Assistant 2.0: Not explicitly mentioned, however, the process described as "Readers to ground truth the 'anatomy' visible in static B-Mode image. (Before running AI)" and then comparing to AI predictions does not suggest an adjudication process for the ground truth generation itself beyond initial reader input. Confusion matrices were generated later.
• Auto Aorta Measure Assistant: An "Arbitrator" was used to "select most accurate measurement among all readers" for the initial ground truth, which was then compared to AI baseline. This implies a 1 (arbitrator) + N (readers) adjudication method for measurement accuracy. For keystroke comparison, readers measured with and without AI.
• Auto Common Bile Duct (CBD) Measure Assistant: An "Arbitrator" was used to "select most accurate measurement among all readers" for the initial ground truth, which was then compared to AI baseline. This implies a 1 (arbitrator) + N (readers) adjudication method for measurement accuracy. For keystroke comparison, readers measured with and without AI.
• Ultrasound Guided Fat Fraction (UGFF): Ground truth was established by MRI-PDFF or comparison to UDFF. No human adjudication method was described for these.

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

• Auto Aorta Measure Assistant: Yes, a comparative study was performed by comparing keystroke counts with and without AI assistance for human readers.
  • Effect Size:
    • Long View Aorta AP Measurement: Average reduction from $4.132 \pm 0.291$ keystrokes (without AI) to $1.236 \pm 0.340$ keystrokes (with AI).
    • Short View Aorta AP and Trans Measurement: Average reduction from $7.05 \pm 0.158$ keystrokes (without AI) to $2.307 \pm 1.0678$ keystrokes (with AI).
• Auto Common Bile Duct (CBD) Measure Assistant: Yes, a comparative study was performed by comparing keystroke counts with and without AI assistance for human readers.
  • Effect Size: Average reduction of $1.62 \pm 0.375$ keystrokes (mean and standard deviation) from manual to AI-assisted measurements.
• Other features (Auto Abdominal Color Assistant 2.0, UGFF): The documentation does not describe a MRMC study for improved human reader performance with AI assistance for these features.

6. Standalone (Algorithm Only) Performance Study

• Auto Abdominal Color Assistant 2.0: Yes, the model's accuracy (detection accuracy, sensitivity, specificity, DICE score) was evaluated in a standalone manner against the human-annotated ground truth.
• Ultrasound Guided Fat Fraction (UGFF): Yes, the correlation and agreement of the UGFF algorithm's values were tested directly against an established reference standard (MRI-PDFF) and another device's derived fat fraction (UDFF).

7. Type of Ground Truth Used

• Auto Abdominal Color Assistant 2.0: Expert consensus/annotations on B-Mode images, followed by comparison to AI predictions.
• Auto Aorta Measure Assistant: Expert consensus on measurements (human readers with arbitrator selection) and keystroke counts from these manual measurements and AI-assisted measurements.
• Auto Common Bile Duct (CBD) Measure Assistant: Expert consensus on measurements (human readers with arbitrator selection) and keystroke counts from these manual measurements and AI-assisted measurements.
• Ultrasound Guided Fat Fraction (UGFF): Established clinical reference standard: MRI Proton Density Fat Fraction (MRI-PDFF %). For one confirmatory study, another cleared device's derived fat fraction (UDFF) was used as a comparative reference.

8. Sample Size for the Training Set

• The document states that "The exams used for test/training validation purpose are separated from the ones used during training process" but does not provide the sample size for the training set itself for any of the AI features.

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

• The document implies that the ground truth for training data would have been established similarly to the test data ground truth (e.g., expert annotation for Auto Abdominal Color Assistant, expert measurements for Auto Aorta/CBD Measure Assistants). However, the specific methodology for the training set's ground truth establishment (e.g., number of experts, adjudication, qualifications) is not detailed in the provided text. It only explicitly states that "Before the process of data annotation, all information displayed on the device is removed and performed on information extracted purely from Ultrasound B-mode images" for annotation. Independence of test and training data by exam site origin or overall separation is mentioned, but not the process for creating the training set ground truth.

Ask a specific question about this device

The LOGIQ E10s is intended for use by a qualified physician for ultrasound evaluation.

Specific clinical applications and exam types include: Fetal / Obstetrics; Abdominal (including Renal, Gynecology / Pelvic); Pediatric; Small Organ (Breast, Testes, Thyroid); Neonatal Cephalic; Adult Cephalic; Cardiac (Adult and Pediatric); Peripheral Vascular; Musculo-skeletal Conventional and Superficial; Urology (including Prostate); Transrectal; Transvaginal; Transesophageal and Intraoperative (Abdominal and Vascular).

Modes of operation include: B, M, PW Doppler, CW Doppler, Color Doppler, Color M Doppler, Power Doppler, Harmonic Imaging, Coded Pulse, 3D/4D Imaging mode, Elastography, Shear Wave Elastography, Attenuation Imaging and Combined modes: B/M, B/Color, B/PWD, B/Color/PWD, B/Power/PWD.

The LOGIQ E10s is intended to be used in a hospital or medical clinic.

The LOGIQ E10s is a full featured, Track 3, general purpose diagnostic ultrasound system which consists of a mobile console approximately 585 mm wide (keyboard), 991 mm deep and 1300 mm high that provides digital acquisition, processing and display capability. The user interface includes a computer keyboard, specialized controls, 12-inch high-resolution color touch screen and 23.8-inch High Contrast LED LCD monitor.

The provided text describes three AI features: Auto Abdominal Color Assistant 2.0, Auto Aorta Measure Assistant, and Auto Common Bile Duct (CBD) Measure Assistant, along with a UGFF Clinical Study.

Here's an analysis of the acceptance criteria and study details for each, where available:

1. Table of Acceptance Criteria and Reported Device Performance

For Auto Abdominal Color Assistant 2.0

For Auto Aorta Measure Assistant

For Auto Common Bile Duct (CBD) Measure Assistant

For UGFF Clinical Study

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

Auto Abdominal Color Assistant 2.0:

• Sample Size: 49 individual subjects (1186 annotation images)
• Data Provenance: Retrospective, from the USA (100%).

Auto Aorta Measure Assistant:

• Sample Size:
  • Long View Aorta: 36 subjects
  • Short View Aorta: 35 subjects
• Data Provenance: Retrospective, from Japan and USA.

Auto Common Bile Duct (CBD) Measure Assistant:

• Sample Size: 25 subjects
• Data Provenance: Retrospective, from USA (40%) and Japan (60%).

UGFF Clinical Study:

• Sample Size:
  • Original study: 582 participants
  • Confirmatory study (US/EU): 15 US patients and 5 EU patients (total 20)
  • Confirmatory study (UGFF vs UDFF): 24 EU patients
• Data Provenance: Retrospective and Prospective implicitly (clinical study implies data collection).
  • Original Study: Japan (Asian population)
  • Confirmatory Study (US/EU): US and EU (demographic info unavailable for EU patients, US patients: BMI 21.0-37.5, SCD 13.9-26.9)
  • Confirmatory Study (UGFF vs UDFF): EU

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

Auto Abdominal Color Assistant 2.0:

• Number of Experts: Not specified. The text mentions "Readers to ground truth the 'anatomy'".
• Qualifications of Experts: Not specified.

Auto Aorta Measure Assistant:

• Number of Experts: Not specified. The text mentions "Readers to ground truth the AP measurement..." and an "Arbitrator to select most accurate measurement among all readers." This implies multiple readers and a single arbitrator.
• Qualifications of Experts: Not specified.

Auto Common Bile Duct (CBD) Measure Assistant:

• Number of Experts: Not specified. The text mentions "Readers to ground truth the diameter..." and an "Arbitrator to select most accurate measurement among all readers." This implies multiple readers and a single arbitrator.
• Qualifications of Experts: Not specified.

UGFF Clinical Study:

• Number of Experts: Not applicable, as ground truth was established by MRI-PDFF measurements, not expert consensus on images.

4. Adjudication method for the test set

Auto Abdominal Color Assistant 2.0:

• Adjudication Method: Not explicitly described as a specific method (e.g., 2+1). The process mentions "Readers to ground truth" and then comparison to AI predictions, but no specific adjudication among multiple readers' initial ground truths.

Auto Aorta Measure Assistant:

• Adjudication Method: Implies an arbitrator-based method. "Arbitrator to select most accurate measurement among all readers." This suggests multiple readers provide measurements, and a single arbitrator makes the final ground truth selection.

Auto Common Bile Duct (CBD) Measure Assistant:

• Adjudication Method: Implies an arbitrator-based method. "Arbitrator to select most accurate measurement among all readers." Similar to the Aorta assistant.

UGFF Clinical Study:

• Adjudication Method: Not applicable. Ground truth was established by MRI-PDFF measurements.

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

Auto Abdominal Color Assistant 2.0:

• MRMC Study: Not explicitly stated as a comparative effectiveness study showing human improvement. The study focuses on the algorithm's performance against ground truth.
• Effect Size (Human Improvement with AI): Not reported.

Auto Aorta Measure Assistant:

• MRMC Study: Yes, an implicit MRMC study comparing human performance with and without AI. Readers performed measurements with and without AI assistance.
• Effect Size (Human Improvement with AI):
  • Long View Aorta (Keystrokes): Average keystrokes reduced from 4.132 (without AI) to 1.236 (with AI).
  • Short View Aorta (Keystrokes): Average keystrokes reduced from 7.05 (without AI) to 2.307 (with AI).
  • (No specific improvement in diagnostic accuracy for human readers with AI is stated, primarily focuses on efficiency via keystrokes).

Auto Common Bile Duct (CBD) Measure Assistant:

• MRMC Study: Yes, an implicit MRMC study comparing human performance with and without AI. Readers performed measurements with and without AI assistance.
• Effect Size (Human Improvement with AI):
  • Porta Hepatis CBD (Keystrokes): Average reduction in keystrokes for measurements with AI vs. manually is 1.62 +/- 0.375.
  • (No specific improvement in diagnostic accuracy for human readers with AI is stated, primarily focuses on efficiency via keystrokes).

UGFF Clinical Study:

• MRMC Study: No, this was a standalone algorithm performance study compared to a reference standard (MRI-PDFF) and a predicate device (UDFF). It did not involve human readers using the AI tool.

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

Auto Abdominal Color Assistant 2.0:

• Standalone Performance: Yes. The reported accuracy, sensitivity, specificity, and DICE score are for the algorithm's performance.

Auto Aorta Measure Assistant:

• Standalone Performance: Yes, implicitly. The "AI baseline measurement" was compared for accuracy against the arbitrator-selected ground truth. While keystrokes involved human interaction to use the AI, the measurement accuracy is an algorithm output.

Auto Common Bile Duct (CBD) Measure Assistant:

• Standalone Performance: Yes, implicitly. The "AI baseline measurement" was compared for accuracy against the arbitrator-selected ground truth.

UGFF Clinical Study:

• Standalone Performance: Yes. The study directly assesses the correlation and agreement of the UGFF algorithm's output with MRI-PDFF and another ultrasound-derived fat fraction algorithm.

7. The type of ground truth used

Auto Abdominal Color Assistant 2.0:

• Ground Truth Type: Expert consensus for anatomical visibility ("Readers to ground truth the 'anatomy' visible in static B-Mode image.")

Auto Aorta Measure Assistant:

• Ground Truth Type: Expert consensus from multiple readers, adjudicated by an arbitrator, for specific measurements ("Arbitrator to select most accurate measurement among all readers.")

Auto Common Bile Duct (CBD) Measure Assistant:

• Ground Truth Type: Expert consensus from multiple readers, adjudicated by an arbitrator, for specific measurements ("Arbitrator to select most accurate measurement among all readers.")

UGFF Clinical Study:

• Ground Truth Type: Outcomes data / Quantitative Reference Standard: MRI Proton Density Fat Fraction (MRI-PDFF %).

8. The sample size for the training set

Auto Abdominal Color Assistant 2.0:

• Training Set Sample Size: Not specified beyond "The exams used for test/training validation purpose are separated from the ones used during training process".

Auto Aorta Measure Assistant:

• Training Set Sample Size: Not specified beyond "The exams used for regulatory validation purpose are separated from the ones used during model development process".

Auto Common Bile Duct (CBD) Measure Assistant:

• Training Set Sample Size: Not specified beyond "The exams used for regulatory validation purpose are separated from the ones used during model development process".

UGFF Clinical Study:

• Training Set Sample Size: Not specified. The study describes validation but not the training phase.

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

Auto Abdominal Color Assistant 2.0:

• Training Set Ground Truth: Not explicitly detailed, but implied to be similar to the test set ground truthing process: "Information extracted purely from Ultrasound B-mode images" and "Readers to ground truth the 'anatomy'".

Auto Aorta Measure Assistant:

• Training Set Ground Truth: Not explicitly detailed, but implied to be similar to the test set ground truthing process: "Information extracted purely from Ultrasound B-mode images" and "Readers to ground truth the AP measurement...".

Auto Common Bile Duct (CBD) Measure Assistant:

• Training Set Ground Truth: Not explicitly detailed, but implied to be similar to the test set ground truthing process: "Information extracted purely from Ultrasound B-mode images" and "Readers to ground truth the diameter...".

UGFF Clinical Study:

• Training Set Ground Truth: Not specified for the training set, but for the validation set, the ground truth was MRI-PDFF measurements.

Ask a specific question about this device

The SIGNA™ Sprint is a whole body magnetic resonance scanner designed to support high resolution, high signal-to-noise ratio, and short scan times. It is indicated for use as a diagnostic imaging device to produce axial, sagittal, coronal, and oblique images, spectroscopic images, parametric maps, and/or spectra, dynamic images of the structures and/or functions of the entire body, including, but not limited to, head, neck, TMJ, spine, breast, heart, abdomen, pelvis, joints, prostate, blood vessels, and musculoskeletal regions of the body. Depending on the region of interest being imaged, contrast agents may be used.

The images produced by SIGNA™ Sprint reflect the spatial distribution or molecular environment of nuclei exhibiting magnetic resonance. These images and/or spectra when interpreted by a trained physician yield information that may assist in diagnosis.

SIGNA™ Sprint is a whole-body magnetic resonance scanner designed to support high resolution, high signal-to-noise ratio, and short scan time. The system uses a combination of time-varying magnet fields (Gradients) and RF transmissions to obtain information regarding the density and position of elements exhibiting magnetic resonance. The system can image in the sagittal, coronal, axial, oblique, and double oblique planes, using various pulse sequences, imaging techniques and reconstruction algorithms. The system features a 1.5T superconducting magnet with 70cm bore size. The system is designed to conform to NEMA DICOM standards (Digital Imaging and Communications in Medicine).

Key aspects of the system design:

• Uses the same magnet as a conventional whole-body 1.5T system, with integral active shielding and a zero boil-off cryostat.
• A gradient coil that achieves up to 65 mT/m peak gradient amplitude and 200 T/m/s peak slew rate.
• An embedded body coil that reduces thermal and enhance intra-bore visibility.
• A newly designed 1.5T AIR Posterior Array.
• A detachable patient table.
• A platform software with various PSD and applications, including the following AI features:
  • AIRx™ (Cleared in K183231)
  • AIR™ Recon DL (Cleared in K202238)
  • Sonic DL™ (Cleared in K223523)

The provided text is a 510(k) clearance letter and summary for a new MRI device, SIGNA™ Sprint. It states explicitly that no clinical studies were required to support substantial equivalence. Therefore, the information requested regarding acceptance criteria, study details, sample sizes, ground truth definitions, expert qualifications, and MRMC studies is not available in this document.

The document highlights the device's technical equivalence to a predicate device (SIGNA™ Premier) and reference devices (SIGNA™ Artist, SIGNA™ Champion) and relies on non-clinical tests and sample clinical images to demonstrate acceptable diagnostic performance.

Here's a breakdown of what can be extracted from the document regarding testing, and why other requested information is absent:

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

• Acceptance Criteria (Implicit): The document states that the device's performance is demonstrated through "bench testing and clinical testing that show the image quality performance of SIGNA™ Sprint compared to the predicate device." It also mentions "acceptable diagnostic image performance... in accordance with the FDA Guidance 'Submission of Premarket Notifications for Magnetic Resonance Diagnostic Devices' issued on October 10, 2023."
  • Specific quantitative acceptance criteria (e.g., minimum SNR, CNR, spatial resolution thresholds) are not explicitly stated in this document.
• Reported Device Performance: "The images produced by SIGNA™ Sprint reflect the spatial distribution or molecular environment of nuclei exhibiting magnetic resonance. These images and/or spectra when interpreted by a trained physician yield information that may assist in diagnosis."
  • No specific quantitative performance metrics (e.g., sensitivity, specificity, accuracy, or detailed image quality scores) are provided in this regulatory summary. The statement "The image quality of the SIGNA™ Sprint is substantially equivalent to that of the predicate device" is the primary performance claim.

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

• Test Set Sample Size: Not applicable/Not provided. The document explicitly states: "The subject of this premarket submission, the SIGNA™ Sprint, did not require clinical studies to support substantial equivalence."
• Data Provenance: Not applicable/Not provided for a formal clinical test set. The document only mentions "Sample clinical images have been included in this submission," but does not specify their origin or nature beyond being "sample."

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)

• Not applicable. Since no formal clinical study was conducted for substantial equivalence, there was no "test set" requiring ground truth established by experts in the context of an effectiveness study. The "interpretation by a trained physician" is mentioned in the Indications for Use, which is general to MR diagnostics, not specific to a study.

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

• Not applicable. No clinical test set requiring adjudication was conducted for substantial equivalence.

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. The document explicitly states: "The subject of this premarket submission, the SIGNA™ Sprint, did not require clinical studies to support substantial equivalence." While the device incorporates AI features cleared in other submissions (AIRx™, AIR™ Recon DL, Sonic DL™), this specific 510(k) for the SIGNA™ Sprint system itself does not include an MRMC study or an assessment of human reader improvement with these integrated AI features. The focus is on the substantial equivalence of the overall MR system.

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

• No, not for the SIGNA™ Sprint as a whole system. This 510(k) is for the MR scanner itself, not for a standalone algorithm. Any standalone performance for the integrated AI features (AIRx™, AIR™ Recon DL, Sonic DL™) would have been part of their respective clearance submissions (K183231, K202238, K223523), not this one.

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

• Not applicable. No formal clinical study requiring ground truth was conducted for this submission.

8. The sample size for the training set

• Not applicable/Not provided. This submission is for the SIGNA™ Sprint MR system itself, not a new AI algorithm requiring a training set developed for this specific submission. The AI features mentioned (AIRx™, AIR™ Recon DL, Sonic DL™) were cleared in previous 510(k)s and would have had their own training and validation processes.

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

• Not applicable/Not provided. As explained in point 8, this submission does not detail the training of new AI algorithms.

Ask a specific question about this device

The system is intended to produce cross-sectional images of the body by computer reconstruction of x-ray transmission projection data from the same axial plane taken at different angles. The system may acquire data using Axial, Cine, Helical, Cardiac, and Gated CT scan techniques from patients of all ages. These images may be obtained either with or without contrast. This device may include signal analysis and display equipment, patient and equipment supports, components and accessories.

This device may include data and image processing to produce images in a variety of trans-axial and reformatted planes. Further, the images can be post processed to produce additional imaging planes or analysis results.

The system is indicated for head, whole body, cardiac, and vascular X-ray Computed Tomography applications.

The device output is a valuable medical tool for the diagnosis of disease, trauma, or abnormality and for planning, guiding, and monitoring therapy.

If the spectral imaging option is included on the system, the system can acquire CT images using different kV levels of the same anatomical region of a patient in a single rotation from a single source. The differences in the energy dependence of the attenuation coefficient of the different materials provide information about the chemical composition of body materials. This approach enables images to be generated at energies selected from the available spectrum to visualize and analyze information about anatomical and pathological structures.

GSI provides information of the chemical composition of renal calculi by calculation and graphical display of the spectrum of effective atomic number. GSI Kidney stone characterization provides additional information to aid in the characterization of uric acid versus non-uric acid stones. It is intended to be used as an adjunct to current standard methods for evaluating stone etiology and composition.

The CT system is indicated for low dose CT for lung cancer screening. The screening must be performed within the established inclusion criteria of programs/ protocols that have been approved and published by either a governmental body or professional medical society.

This proposed device Revolution Vibe is a general purpose, premium multi-slice CT Scanning system consisting of a gantry, table, system cabinet, scanner desktop, power distribution unit, and associated accessories. It has been optimized for cardiac performance while still delivering exceptional imaging quality across the entire body.

Revolution Vibe is a modified dual energy CT system based on its predicate device Revolution Apex Elite (K213715). Compared to the predicate, the most notable change in Revolution Vibe is the modified detector design together with corresponding software changes which is optimized for cardiac imaging providing capability to image the whole heart in one single rotation same as the predicate.

Revolution Vibe offers an accessible whole heart coverage, full cardiac capability CT scanner which can deliver outstanding routine head and body imaging capabilities. The detector of Revolution Vibe uses the same GEHC's Gemstone scintillator with 256 x 0.625 mm row providing up to 16 cm of coverage in Z direction within 32 cm scan field of view, and 64 x 0.625 mm row providing up to 4 cm of coverage in Z direction within 50 cm scan field of view. The available gantry rotation speeds are 0.23, 0.28, 0.35, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, and 1.0 seconds per rotation.

Revolution Vibe inherits virtually all of the key technologies from the predicate such as: high tube current (mA) output, 80 cm bore size with Whisper Drive, Deep Learning Image Reconstruction for noise reduction (DLIR K183202/K213999, GSI DLIR K201745), ASIR-V iterative recon, enhanced Extended Field of View (EFOV) reconstruction MaxFOV 2 (K203617), fast rotation speed as fast as 0.23 second/rot (K213715), and spectral imaging capability enabled by ultrafast kilovoltage(kv) switching (K163213), as well as ECG-less cardiac (K233750). It also includes the Auto ROI enabled by AI which is integrated within the existing SmartPrep workflow for predicting Baseline and monitoring ROI automatically. As such, the Revolution Vibe carries over virtually all features and functionalities of the predicate device Revolution Apex Elite (K213715).

This CT system can be used for low dose lung cancer screening in high risk populations*.

The provided FDA 510(k) clearance letter and summary for the Revolution Vibe CT system does not include detailed acceptance criteria or a comprehensive study report to fully characterize the device's performance against specific metrics. The information focuses more on the equivalence to a predicate device and general safety/effectiveness.

However, based on the text, we can infer some aspects related to the Auto ROI feature, which is the only part of the device described with specific performance testing details.

Here's an attempt to extract and describe the available information, with clear indications of what is not provided in the document.

Acceptance Criteria and Device Performance for Auto ROI

The document mentions specific performance testing for the "Auto ROI" feature, which utilizes AI. For other aspects of the Revolution Vibe CT system, the submission relies on demonstrating substantial equivalence to the predicate device (Revolution Apex Elite) through engineering design V&V, bench testing, and a clinical reader study focused on overall image utility, rather than specific quantitative performance metrics meeting predefined acceptance criteria for the entire system.

1. Table of Acceptance Criteria and Reported Device Performance (Specific to Auto ROI)

Note: The document does not provide the explicit numerical value for the "pre-established acceptance criteria" or the actual "success rate" achieved for the Auto ROI feature.

2. Sample Size and Data Provenance for the Test Set (Specific to Auto ROI)

• Sample Size: 1341 clinical images
• Data Provenance: "real clinical practice" (Specific country of origin not mentioned). The images were used for "Auto ROI performance" testing, which implies retrospective analysis of existing clinical data.

3. Number of Experts and Qualifications to Establish Ground Truth (Specific to Auto ROI)

• Number of Experts: Not specified for the Auto ROI ground truth establishment.
• Qualifications of Experts: Not specified for the Auto ROI ground truth establishment.

Note: The document mentions 3 readers for the overall clinical reader study (see point 5), but this is for evaluating the diagnostic utility and image quality of the CT system and not explicitly for establishing ground truth for the Auto ROI feature.

4. Adjudication Method for the Test Set (Specific to Auto ROI)

• Adjudication Method: Not specified for the Auto ROI test set.

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

• Was an MRMC study done? Yes, a "clinical reader study of sample clinical data" was carried out. It is described as a "blinded, retrospective clinical reader study."

• Effect Size of Human Readers Improvement with AI vs. without AI assistance: The document states the purpose of this reader study was to validate that "Revolution Vibe are of diagnostic utility and is safe and effective for its intended use." It does not report an effect size or direct comparison of human readers' performance with and without AI assistance (specifically for the Auto ROI feature within the context of reader performance). The study seemed to evaluate the CT system's overall image quality and clinical utility, possibly implying that the Auto ROI is integrated into this overall evaluation, but a comparative effectiveness study of the AI's impact on human performance is not described.

  • Details of MRMC Study:
    • Number of Cases: 30 CT cardiac exams
    • Number of Readers: 3
    • Reader Qualifications: US board-certified in Radiology with more than 5 years' experience in CT cardiac imaging.
    • Exams Covered: "wide range of cardiac clinical scenarios."
    • Reader Task: "Readers were asked to provide evaluation of image quality and the clinical utility."

6. Standalone (Algorithm Only) Performance

• Was a standalone study done? Yes, for the "Auto ROI" feature, performance was tested "using 1341 clinical images from real clinical practice," and "the tests results in success rates exceeding the pre-established acceptance criteria." This implies an algorithm-only evaluation of the Auto ROI's ability to successfully identify and monitor ROI.

7. Type of Ground Truth Used (Specific to Auto ROI)

• Type of Ground Truth: Not explicitly stated for the Auto ROI. Given the "success rates" metric, it likely involved a comparison against a predefined "true" ROI determined by human experts or a gold standard method. It's plausible that this was established by expert consensus or reference standards.

8. Sample Size for the Training Set

• Sample Size: Not provided in the document.

9. How Ground Truth for the Training Set Was Established

• Ground Truth Establishment: Not provided in the document.

In summary, the provided documentation focuses on demonstrating substantial equivalence of the Revolution Vibe CT system to its predicate, Revolution Apex Elite, rather than providing detailed, quantitative performance metrics against specific acceptance criteria for all features. The "Auto ROI" feature is the only component where specific performance testing (standalone) is briefly mentioned, but key details like numerical acceptance criteria, actual success rates, and ground truth methodology for training datasets are not disclosed. The human reader study was for general validation of diagnostic utility, not a comparative effectiveness study of AI assistance.

Ask a specific question about this device

The Venue, Venue Go, Venue Fit and Venue Sprint are general purpose diagnostic ultrasound systems for use by qualified and trained healthcare professionals or practitioners that are legally authorized or licensed by law in the country, state or other local municipality in which he or she practices, for ultrasound imaging, measurement, display and analysis of the human body and fluid. The users may or may not be working under supervision or authority of a physician. Users may also include medical students working under the supervision or authority of a physician during their education / training.

Venue, Venue Go and Venue Fit are intended to be used in a hospital or medical clinic. Venue, Venue Go and Venue Fit clinical applications include: abdominal (GYN and Urology), thoracic/pleural, ophthalmic, Fetal/OB, Small Organ (including breast, testes, thyroid), Vascular/Peripheral vascular, neonatal and adult cephalic, pediatric, musculoskeletal (conventional and superficial), cardiac (adults and pediatric), Transrectal, Transvaginal, Transesophageal, Intraoperative (vascular) and interventional guidance (includes tissue biopsy, fluid drainage, vascular and non-vascular access). Modes of operation include: B, M, PW Doppler, CW Doppler, Color Doppler, Color M Doppler, Power Doppler, Harmonic Imaging, Coded Pulse and Combined modes: B/M, B/Color M, B/PWD, B/Color/PWD, B/Power/PWD, B/CWD, B/Color/CWD.

The Venue Sprint is intended to be used in a hospital, medical clinic, home environment and road/air ambulance. Venue Sprint clinical applications include: abdominal (GYN and Urology), thoracic/pleural, ophthalmic, Fetal/OB, Small Organ (including breast, testes, thyroid), Vascular/Peripheral vascular, neonatal and adult cephalic, pediatric, musculoskeletal (conventional and superficial), cardiac (adults and pediatric, 40 kg and above) and interventional guidance (includes free hand tissue biopsy, fluid drainage, vascular and non-vascular access). Modes of operation include: B, M, PW Doppler, Color Doppler and Harmonic Imaging.

Venue, Venue Go, Venue Fit and Venue Sprint are general-purpose diagnostic ultrasound systems intended for use by qualified and trained healthcare professionals to evaluate the body by ultrasound imaging and fluid flow analysis.

The systems utilize a variety of linear, convex, and phased array transducers which provide high imaging capability, supporting all standard acquisition modes.

The systems have a small footprint that easily fits into tight spaces and positioned to accommodate the sometimes-awkward work settings of the point of care user.

The Venue is a mobile system, the Venue Go and Venue Fit are compact, portable systems that can be hand carried using an integrated handle, placed on a horizontal surface, attached to a mobile cart or mounted on the wall. Venue, Venue Go and Venue Fit have a high-resolution color LCD monitor, with a simple, multi-touch user interface that makes the systems intuitive.

The Venue Sprint is used together with the Vscan Air probes and provides the user interface for control of the probes and the needed software functionality for analysis of the ultrasound images and saving/storage of the related images and videos.

The Venue, Venue Go, Venue Fit and Venue Sprint systems can be powered through an electrical wall outlet for long term use or from an internal battery for a short time with full functionality and scanning. A barcode reader and RFID scanner are available as additional input devices. The systems meet DICOM requirements to support users image storage and archiving needs and allows for output to printing devices.

The Venue, Venue Go and Venue Fit systems are capable of displaying the patient's ECG trace synchronized to the scanned image. This allows the user to view an image from a specific time of the ECG signal which is used as an input for gating during scanning. The ECG signal can be input directly from the patient or as an output from an ECG monitoring device. ECG information is not intended for monitoring or diagnosis. Compatible biopsy kits can be used for needle-guidance procedures.

The provided document, a 510(k) Clearance Letter and Submission Summary, primarily focuses on the substantial equivalence of the GE Healthcare Venue series of diagnostic ultrasound systems to previously cleared predicate devices. It specifically details the "Auto Bladder Volume (ABV)" feature as an AI-powered component and provides a summary of its testing.

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

1. Table of Acceptance Criteria and Reported Device Performance (for Auto Bladder Volume - ABV)

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

• Test Set (Verification Dataset) Sample Size: 1874 images from 101 individuals.
• Data Provenance:
  • Country of Origin: USA and Israel.
  • Retrospective or Prospective: Not explicitly stated as either retrospective or prospective. However, the description of "data collected from several different Console variants" for training and verification suggests pre-existing data, which often leans towards a retrospective collection.

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

• Number of Experts: Not explicitly stated. The document refers to "annotators" who performed manual annotation.
• Qualifications of Experts: Not explicitly stated. The annotators are described as performing "manual annotation," implying they are skilled in this task, but specific qualifications (e.g., radiologists, sonographers, years of experience) are not provided.

4. Adjudication Method for the Test Set

• Adjudication Method: Not explicitly stated. The document mentions "annotators performed manual annotation," but does not detail if multiple annotators were used for each case or any specific adjudication process (e.g., 2+1, 3+1 consensus).

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

• Was an MRMC study done? No. The document states: "The subjects of this premarket submission, Venue, Venue Go, Venue Fit and Venue Sprint, did not require clinical studies to support substantial equivalence." The testing described for ABV is a standalone algorithm performance validation against established ground truth, not a comparative human-AI study.
• Effect Size of Human Readers Improvement: Not applicable, as no MRMC study was performed.

6. Standalone (Algorithm Only) Performance Study

• Was a standalone study done? Yes. The "AI Summary of Testing" section describes a study for the Auto Bladder Volume (ABV) feature, which assesses the algorithm's "automatic caliper placement success rate" against manually established ground truth. This is a standalone performance evaluation of the algorithm.

7. Type of Ground Truth Used (for ABV Test Set)

• Ground Truth Type: Expert consensus/manual annotation. The document states: "Ground truth annotations of the verification dataset were obtained as follows: In all Training/Validation and Verification datasets, annotators performed manual annotation on images converted from DICOM files." They identified "landmarks, which represent the bladder edges," corresponding to standard measurement locations.

8. Sample Size for the Training Set (for ABV)

• Training Set Sample Size: Total dataset included 8,392 images from 496 individuals. Of these, 1,874 were used for the verification dataset, and "the rest" were used for training/validation. This implies the training/validation set would be 8392 - 1874 = 6518 images from the remaining individuals not included in the verification set.

9. How the Ground Truth for the Training Set Was Established (for ABV)

• Ground Truth Establishment: Similar to the verification dataset, "annotators performed manual annotation on images converted from DICOM files" for both Training/Validation and Verification datasets. They chose "4-6 images that represent different bladder volume status" for each individual and annotated "4 different landmarks" per view (transverse and longitudinal) representing bladder edges.

Ask a specific question about this device