K Number
K110447
Device Name
ECHOINSIGHT
Date Cleared
2011-05-27

(100 days)

Product Code
Regulation Number
892.2050
Panel
RA
Reference & Predicate Devices
Predicate For
N/A
AI/MLSaMDIVD (In Vitro Diagnostic)TherapeuticDiagnosticis PCCP AuthorizedThirdpartyExpeditedreview
Intended Use

Intended for use by or on the order of a qualified physician for analysis of ultrasound imaging of the human heart. Cardiac (adult and pediatric) applications using B-mode (including harmonic) imaging are supported. The system provides image data analysis applications that supply information on cardiac structure and motion.

Device Description

The Echolnsight™ software system enables the production and visualization of 2D tissue motion measurements (including tissue velocities, strain rates) and cardiac structural measurement information derived from tracking speckle in tissue regions visualized in any Bmode (including harmonic) imagery loops as captured by most commercial ultrasound systems.

The Echolnsight™ software system has been designed to ingest and process ultrasound imagery stored in files with content organized and encoded in conformance to the following the following standards:

  • NEMA PS3 (3.3, 3.5, 3.6, 3.10): Digital Imaging and Communications in Medicine . (DICOM), 2008.
  • ISO/IEC 10918-1:1994-02: Information technology Digital compression and coding of . continuous-tone still images – Requirements and guidelines (JPEG standard), 1994.
  • ISO/IEC 15444-1:2004: Information technology JPEG 2000 image coding system: . Core coding system, 2004.
AI/ML Overview

The provided text describes a 510(k) premarket notification for the EchoInsight™ device. However, it does not contain the acceptance criteria or details of a study proving the device meets acceptance criteria. The document primarily focuses on:

  • Device Description: What EchoInsight™ does (processing 2D tissue motion measurements, including velocities, strain rates, and cardiac structural measurements from ultrasound B-mode imagery using speckle tracking).
  • Intended Use: For analysis of ultrasound imaging of the human heart (adult and pediatric), providing information on cardiac structure and motion.
  • Technological Comparison to Predicate Device: Stating substantial equivalence to the Siemens VVI product (K091286) based on similar speckle tracking techniques for tissue motion, strain, and strain rate, and similar visualization capabilities.
  • FDA Clearance Letter: Confirming the substantial equivalence determination.

Therefore, I cannot fulfill the request to provide a table of acceptance criteria, device performance, or study details, as these are not present in the provided text.

To illustrate what the requested information would look like if it were present, here's a hypothetical structure:


Hypothetical Acceptance Criteria and Study Details (Based on typical medical device submissions, NOT on the provided text):

1. Table of Acceptance Criteria and Reported Device Performance

Performance MetricAcceptance Criteria (e.g., Predicate Equivalence, Clinical Threshold)Reported Device PerformanceOutcome
Accuracy of Strain MeasurementMean absolute difference (MAD) ≤ 5% compared to predicate device or gold standard.MAD = 3.8% vs. predicateMet
Precision of Velocity MeasurementCoefficient of Variation (CV) ≤ 10% for repeated measurements.CV = 7.2%Met
Processing TimeMean processing time per image 0.8 for visual assessment of global and regional cardiac function.Kappa = 0.85Met
Intra-operator Variability (Quantitative Reproducibility)Intraclass Correlation Coefficient (ICC) > 0.9 for key cardiac parameters (e.g., Global Longitudinal Strain).ICC = 0.92Met

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

  • Sample Size: For quantitative accuracy/precision, 150 cardiac ultrasound studies from 100 distinct patients. For qualitative assessment, 200 distinct cardiac ultrasound studies.
  • Data Provenance: Retrospective, de-identified cardiac ultrasound studies collected from three university hospitals in the United States, Germany, and Japan.

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

  • Number of Experts: 3 independent expert cardiologists.
  • Qualifications: Each expert had over 10 years of experience in echocardiography interpretation, board-certified in cardiology, and specialized in advanced cardiac imaging techniques including speckle tracking.

4. Adjudication Method for the Test Set

  • Adjudication Method: 2+1 (Two experts' consensus was taken as ground truth; if they disagreed, a third expert provided a tie-breaking decision. All three experts were blinded to the device's output during ground truth establishment).

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

  • Was an MRMC study done? Yes.
  • Effect Size of Human Readers Improvement with AI vs. without AI assistance:
    • ROC AUC Improvement: Average increase in Area Under the Receiver Operating Characteristic Curve (ROC AUC) of 0.08 (e.g., from 0.82 to 0.90) for diagnosing early-stage cardiomyopathy.
    • Reading Time Reduction: Average reduction in reading time per case by 25% (e.g., from 120 seconds to 90 seconds).
    • Diagnostic Accuracy Increase: Increase in diagnostic accuracy by 10% (e.g., from 80% to 90%) in identifying regional wall motion abnormalities.

6. Standalone (Algorithm Only) Performance Study

  • Was a standalone study done? Yes.
  • Performance Metrics:
    • Sensitivity: 92% for detecting global longitudinal strain abnormalities.
    • Specificity: 88% for detecting global longitudinal strain abnormalities.
    • Accuracy: 90% for classifying normal vs. abnormal cardiac function based on strain.

7. Type of Ground Truth Used

  • Ground Truth Type: Expert consensus from the 3 cardiologists, validated against subsequent patient outcomes data (e.g., pathology reports, follow-up clinical events over 12 months for cases of suspected disease).

8. Sample Size for the Training Set

  • Training Set Sample Size: 15,000 cardiac ultrasound studies from 10,000 distinct patients.

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

  • Training Set Ground Truth: Established through a combination of:
    • Automated extraction of quantitative parameters from existing clinical reports (e.g., ejection fraction, global longitudinal strain measured by predicate devices or manual methods).
    • Annotation by a team of trained, certified echocardiography technicians and junior cardiologists under the supervision of senior cardiologists.
    • Randomized review of 10% of annotations by a senior cardiologist to ensure quality and consistency.

Crucially, the K110447 document is a 510(k) summary focused on substantial equivalence to a predicate device, not on detailed performance studies against specific acceptance criteria for a novel algorithm. For such details, one would typically look at the full 510(k) submission, associated peer-reviewed publications, or the manufacturer's technical documentation if it were a newer device.

§ 892.2050 Medical image management and processing system.

(a)
Identification. A medical image management and processing system is a device that provides one or more capabilities relating to the review and digital processing of medical images for the purposes of interpretation by a trained practitioner of disease detection, diagnosis, or patient management. The software components may provide advanced or complex image processing functions for image manipulation, enhancement, or quantification that are intended for use in the interpretation and analysis of medical images. Advanced image manipulation functions may include image segmentation, multimodality image registration, or 3D visualization. Complex quantitative functions may include semi-automated measurements or time-series measurements.(b)
Classification. Class II (special controls; voluntary standards—Digital Imaging and Communications in Medicine (DICOM) Std., Joint Photographic Experts Group (JPEG) Std., Society of Motion Picture and Television Engineers (SMPTE) Test Pattern).