Search Results

deepLive is intended to be used as a non-invasive imaging tool in the evaluation of external human tissue microstructure by providing three-dimensional, cross-sectional and en-face real-time depth visualization for assessment by physicians to support in forming a clinical judgment.

deepLive was designed for an easy integration into clinical practices. The device is composed of:

• A. A mobile cart, allowing the user to move the whole device and including a cart tablet for accessories.
• B. A touchscreen, fixed on the cart mast, displaying the software interfaces to the user.
• C. A hand-held probe, integrating the LC-OCT imaging system (interferometric microscope, OCT camera). The probe is connected to the CPU front panel by a sheathed cable bundle, and stored in a dedicated probe-holder fixed on the cart tablet. The probe is the interface between the device, the doctor and the patient: its measuring head (tip) must be positioned in contact with the patient's skin.
• D. A central power unit (CPU), mounted on the cart, integrating various imaging and electronic peripherals (laser, computer, electronic cards, drivers, power supplies, etc.), driving and powering the imaging probe.
• E. A software running on the device's computer, which controls the components of the system, acquires and processes images, and provides user interfaces for performing examinations and managing data.

deepLive hardware interfaces are located on the front-panel of the CPU. Input/output connections include:

• 1 Display port to connect the screen
• · 3 USB ports to connect external drives (Wifi key, hard drive disk, etc.)

deepLive software runs on a computer embedded in the CPU of the device. The computer uses Windows Enterprise LTSC operating system. The software executable and all dynamic libraries needed for program execution are deployed at a specific location in the file system.

The secured access to the computer operating system, deepLive software and data folders are managed by Windows sessions authentication system. The computer hosting deepLive is also likely to have applications installed by DAMAE Medical:

• · Synology Drive: used to retrieve device data for maintenance and software improvement purposes.
• TeamViewer: remote control software used for software manual update and software issues solving.

The provided text is an FDA 510(k) clearance letter and associated summary for the deepLive device. It outlines the device's characteristics, indications for use, and a comparison to a predicate device. However, it does not contain the detailed information necessary to fully address all parts of your request regarding acceptance criteria and the comprehensive study that proves the device meets these criteria.

Specifically, the document states: "Safety and performance of the deepLive device have been evaluated and verified in accordance with product and software specifications and applicable performance standards through verification, validation, nonclinical performance, and safety testing." and "Verification, validation, nonclinical performance, and safety test results established that the device meets its design requirements and indications for use, that it is as safe and as effective as the predicate device, and that no new questions of safety and effectiveness have been raised." However, it does not provide specific numerical acceptance criteria or the reported performance data from these tests in a detailed manner. There's also no mention of a clinical human-in-the-loop study (MRMC) or an AI-specific standalone performance study.

Given the information provided in the input, here's what can be extracted and what cannot:

Acceptance Criteria and Study for deepLive Device

The deepLive device is an imaging system, not an AI/ML diagnostic tool in the sense that would require typical AI performance metrics like sensitivity, specificity, AUC, or reader studies for decision-making support. Its "performance" in this context primarily refers to its imaging capabilities, technical specifications, and safety.

Since the document does not provide a table of acceptance criteria and reported device performance related to a diagnostic task or specific image interpretation metrics, it's impossible to create such a table in the requested format. The performance testing section broadly states that the device was evaluated according to product and software specifications and applicable standards, and that it met its design requirements.

Here's a breakdown of what can be answered based on the provided text:

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

• Acceptance Criteria (Implicit): The text states that "Safety and performance of the deepLive device have been evaluated and verified in accordance with product and software specifications and applicable performance standards through verification, validation, nonclinical performance, and safety testing." and that "Verification, validation, nonclinical performance, and safety test results established that the device meets its design requirements and indications for use..." This implies the acceptance criteria were met internally based on design requirements, but these specific requirements and corresponding performance values are not detailed in the publicly available 510(k) summary.

• Reported Device Performance: The document provides technical specifications of the deepLive device and compares them to the predicate device. These are performance characteristics of the imaging system itself, rather than performance on a diagnostic task (e.g., classifying disease).

  Parameter	Acceptance Criteria (Implied: "Meets or Exceeds Predicate/Design Specs")	Reported Device Performance (deepLive)	Predicate Device (VivoSight Dx)	Substantially Equivalent?
  Imaging Modality	Be OCT	Optical Coherence Tomography	Optical Coherence Tomography	Yes
  Near Infrared Wavelength	Yes (700-1400mm)	Yes	Yes	Yes
  Light Source Wavelength	Compatible for imaging	800 nm	1305 nm	Yes
  Frame Rate (B-scan)	Adequate for real-time imaging	8 fps	5 fps	Yes
  Frame Rate (A-scan)	Adequate for real-time imaging	8 fps	N/A (not specified for Pred.)	Yes
  Lateral Resolution	≤ Predicate Resolution for comparable detail	1.3 μm	7.5 μm	Yes (deepLive is superior)
  Axial Resolution	≤ Predicate Resolution for comparable detail	1.1 μm	10 μm	Yes (deepLive is superior)
  Lateral Scanning Range	Adequate for tissue assessment	1.2 mm	6 mm	Yes
  Axial Scanning Range	Adequate for tissue assessment	0.5 mm	1 mm	Yes
  Optical Safety	Class 1 medical device	Class 1	Class 1	Yes

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

• The document mentions "verification, validation, nonclinical performance, and safety testing" but does not provide sample sizes for any test sets (e.g., number of images, patients, or tissue samples).
• Data provenance is not specified. The type of device (optical coherence tomography for external human tissue microstructure) suggests that if human data was used for validation of imaging quality, it would likely be prospective clinical data, but this is speculative given the lack of detail. There is no indication of country of origin.

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

• This information is not provided because the nature of the device's clearance appears to be based on technical specifications and safety rather than a diagnostic performance study requiring expert ground truth beyond device design verification.

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

• This information is not applicable/not provided as no diagnostic ground truth establishment process is described.

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 study is mentioned. The device is described as an "imaging tool" for physicians to assess and support clinical judgment, but there is no mention of an AI component that assists human readers or an evaluation of such assistance.

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

• No standalone algorithm performance study is mentioned. The device provides image visualization; it is not presented as an AI algorithm making diagnoses.

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

• For the technical specifications, the "ground truth" would be engineering measurements and adherence to specified performance metrics. For safety, it would be compliance with international standards (IEC, EN, ASTM, ISO listed in the document). There is no mention of clinical ground truth (e.g., pathology, outcomes) in the context of a diagnostic performance study.

8. The sample size for the training set

• Not applicable/not provided. The document describes the device as an imaging system, not explicitly an AI/ML model that would require a "training set" in the context of machine learning.

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

• Not applicable/not provided for the same reason as point 8.

Ask a specific question about this device