Search Results

EchoSoft™ is a software calculation package that is used with diagnostic ultrasound images to provide mechanical information about tendon-like tissues of interest that may be used by a physician, along with other medical data, to assist in clinical diagnosis. The software is intended to be used by trained professionals only.

The Echometrix, LLC software, EchoSoft, is an image processing algorithm that, when used with diagnostic ultrasound images, provides qualitative information about the mechanical characteristics of the deforming material by tracking motion, deformation, and ultrasonic echo magnitude change within a given region. This technology can be used by a physician to gather information about the mechanical and functional properties of soft tissues which, in conjunction with standard medical data, can be used to assist in clinical diagnosis.

The provided text does not contain detailed information about specific acceptance criteria, comprehensive device performance data, detailed study methodologies, or comparative effectiveness studies for EchoSoft™. However, based on the information provided, I can construct an answer with the available details and identify where information is missing.

Here's a breakdown of the acceptance criteria and study information, as much as can be gleaned from the text:

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

The document does not explicitly list quantitative acceptance criteria in a dedicated table format or specific performance metrics with target values. It broadly states that the software was evaluated to "verify the ability of the algorithm to distinguish between materials with different mechanical properties" and to "verify performance parameters such as resolution, sensitivity and precision." No actual numbers for these parameters are provided.

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

• Sample Size (Test Set): The exact sample size for the test set is not specified. The document mentions "multiple studies" including "laboratory measurements on phantoms," "several studies conducted on bovine tendons," and "sample clinical images." The number of phantoms, bovine tendons, or clinical images is not quantified.
• Data Provenance:
  • Phantoms: Laboratory measurements.
  • Bovine Tendons: Implies laboratory or research setting, likely not patient data.
  • Clinical Images: Described as "sample clinical images" with "previously diagnosed injuries." There is no indication of the country of origin.
  • Retrospective/Prospective: Not explicitly stated for any of the data types. "Previously diagnosed injuries" for clinical images suggests a retrospective collection, but this is not confirmed.

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

This information is not provided in the document.

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

This information is not provided in the document. Given the nature of the studies (phantom, bovine tissue, and "sample clinical images"), formal expert adjudication methods commonly seen in human-in-the-loop clinical trials are unlikely to have been detailed in this regulatory submission.

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

There is no indication that an MRMC comparative effectiveness study involving human readers with and without AI assistance was performed or reported in this submission. The device is a software calculation package designed to provide "mechanical information," not primarily to improve human reader diagnostic accuracy in a comparative setting described for MRMC studies. The software is intended to "assist in clinical diagnosis," implying its output is used by a physician in conjunction with other data, but its effect on reader performance is not quantified.

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

Yes, the studies described are characteristic of standalone algorithm testing.

• "The EchoSoft 100 software was evaluated with multiple studies to verify the ability of the algorithm to distinguish between materials with different mechanical properties."
• "The studies consisted of laboratory measurements on phantoms containing materials with varying elastic properties to verify performance parameters such as resolution, sensitivity and precision."
• "Several studies were conducted on bovine tendons to demonstrate performance on tissue samples with known defects."
• "Sample clinical images were also provided to demonstrate performance of the software on tendons with previously diagnosed injuries."

These describe direct evaluations of the algorithm's output against known properties or diagnoses, without explicit human reader interaction within the performance metrics.

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

The ground truth used varied depending on the type of study:

• Phantoms: "Materials with varying elastic properties." The ground truth would be the known mechanical properties (e.g., stiffness/elasticity) of these engineered phantom materials.
• Bovine Tendons: "Tissue samples with known defects." The ground truth would be the known presence or absence and characteristics of these defects, likely established through direct inspection, mechanical testing, or other means.
• Clinical Images: "Tendons with previously diagnosed injuries." The ground truth was the existing clinical diagnosis for these injuries. The method of establishing this clinical diagnosis (e.g., expert clinical assessment, MRI, surgery, pathology) is not specified.

8. The sample size for the training set

This information is not provided in the document. The submission focuses on the evaluation (test) studies, with no mention of the training data used to develop the algorithm.

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

This information is not provided in the document, as the training set itself is not discussed.

Ask a specific question about this device