Search Results

The SpectralMD™ DeepView™ Wound Imaging System 2.0 is an optical imaging device intended for studies of blood flow in the microcirculation. The DeepView system is suitable for a wide variety of clinical applications including plastic surgery, diabetes, dermatology, vascular surgery, wound healing, neurology, neurosurgery and anesthetics. In particular, it can be used for measuring perfusion of healthy and injured skin including burn wounds, skin flaps (plastic and reconstructive surgery), chronic wounds, decubitus ulcers and diabetic ulcers.

The DeepView System 2.0 is a prescription device that utilizes the principles of non-contact photoplethysmography (PPG) to capture images of tissue blood perfusion. This is accomplished by measuring the optical properties of tissues and blood as they vary in response to changing hemodynamic conditions. The device's software combines real-time digital analysis based on the interaction of light with vascular tissues below the skin's surface to produce 2-D color images on a touch-screen display depicting relative blood perfusion. The DeepView System consists of a Camera Head with LED optics, an Articulating Arm for Camera Head positioning, a Touch-Screen Display for image viewing, and for accessing and interacting with the Graphical User Interface (GUI). All components are integrated on a Mobile Cart that houses the hardware/software, uninterruptable power supply (UPS), and allows for transport between use environments. The DeepView System 2.0 is AC powered with a backup UPS, and is for use in healthcare/hospital facilities.

The provided text describes the SpectralMD DeepView Wound Imaging System 2.0 and its 510(k) submission for substantial equivalence to predicates.

Here's an analysis of the acceptance criteria and the study proving the device meets them, based on the provided document:

Crucially, the document states: "No clinical performance data were needed to support substantial equivalence." This means that the device's acceptance was not based on human reader studies, AI assistance, or the analysis of detailed clinical performance metrics like sensitivity, specificity, or AUC for diagnostic purposes. Instead, the focus was on demonstrating technological equivalence and safety.

Therefore, many of the typical criteria for evaluating AI/ML-based medical devices (especially those involving diagnostic performance improvement or standalone AI performance) are not applicable in this context, as the device is an imaging system measuring blood flow, not a diagnostic AI.

However, I can extract information related to the device's functional performance acceptance criteria and the engineering studies performed.

Acceptance Criteria and Reported Device Performance

The "acceptance criteria" here are framed around demonstrating equivalence to the predicate device in terms of fundamental functional performance related to blood flow detection and safety.

• Ability to identify the 2% alternating change (AC) modulation consistent with tissue-volume change from blood flow.
• Ability to identify AC modulations at various frequencies within human heart rate frequencies.
• Capability to detect fluid flow beneath the surface of an optically dense medium. These results "demonstrate that the DeepView System 2.0 performs in an equivalent manner to the original DeepView System (primary predicate)." |
  | Equivalent Frequency Detection of Pulsatile Flow | (See above) Demonstrated ability to identify AC modulations at various frequencies within human heart rate range. |
  | Equivalent Capability under Simulated Physiological Conditions | (See above) Demonstrated ability to detect fluid flow beneath the surface of an optically dense medium. |
  | Electrical Safety | Met AAMI/ANSI ES60601-1:2005/(R) 2012 & A1:2012; IEC 60601-1:2005 (Third Edition) + CORR. 1:2006 + CORR. 2:2007. |
  | Electromagnetic Compatibility (EMC) | Met IEC 60601-1-2 (2007)/(R) 2012. |
  | Human Factors and Usability Engineering Compliance & Validation | Designed "in accordance with FDA guidance on human factors and usability engineering" and "subjected to usability testing validation." |

Study Details (Based on the provided text)

Given the stated "No clinical performance data were needed," the "study" is primarily a series of engineering and bench testing studies to demonstrate functional and safety equivalence, not a traditional clinical trial.

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

   • Test Set Sample Size: Not specified in terms of number of "cases" or "patients" as this was not a clinical performance study. The tests likely involved physical phantoms or controlled bench setups to simulate blood flow conditions.
   • Data Provenance: Not applicable in the sense of patient data from specific countries. This was likely laboratory/bench testing.
   • Retrospective/Prospective: Not applicable in a clinical sense. These were controlled engineering experiments.

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

   • Not applicable. The ground truth for these engineering tests would be derived from the controlled experimental setup (e.g., known fluid flow rates, known material properties of optically dense media, precise electrical and EMC standards). No human "experts" establishing "ground truth" in terms of clinical interpretation were needed for these specific tests.

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

   • Not applicable. This type of adjudication is for clinical ground truth establishment, which was not performed here. The "adjudication" would be based on instrument readings and engineering standards.

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 MRMC study was done. The document explicitly states: "No clinical performance data were needed to support substantial equivalence." The device is an imaging system, and its software provides "specific wound modules for facilitating patient/wound documentation," but it's not described as having an AI component that assists human interpretation for a diagnostic outcome that would be subject to an MRMC study.

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

   • No standalone AI performance study was done in the diagnostic sense. The "algorithm" here processes light interaction with tissue to produce 2D perfusion images. The performance assessed was its ability to detect pulsatile flow equivalently to the predicate device, not its ability to make a diagnostic determination on its own.

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

   • The "ground truth" for the functional tests was engineered and controlled experimental conditions (e.g., known AC modulation percentages, known frequency inputs, known optical properties of test media). For safety tests, it was adherence to recognized international electrical safety and EMC standards.

7. The sample size for the training set:

   • Not applicable. The document describes a traditional medical device submission based on predicate equivalence, not an AI/ML device that requires a separate training set for algorithm development. While there's a "Proprietary Software Algorithm" for data analysis, there's no mention of a machine learning training phase or associated dataset.

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

   • Not applicable. As no training set was described for an AI/ML model, no ground truth establishment for such a set is relevant here.

In summary, the DeepView System 2.0's acceptance was based on demonstrating technical and functional equivalence to its predicate device through bench and engineering testing, alongside adherence to safety and usability standards. It was not cleared as an AI-enabled diagnostic device requiring clinical performance studies with human readers or standalone AI performance metrics.

Ask a specific question about this device