Search Results

The ThermPix Thermovisual Camera is intended to view, measure, and record heat patterns and variations. It is intended for use as adjunctive diagnostic imaging for themally significant indications stemming from heat emitted from the human body. The significance of these thermal pariations is determined by professional investigation. This device is intended for use by qualified technical personnel. Clinical judgement and experience are required to review and interpret the information transmitted.

The ThermPix Thermovisual Camera is only for use in addition to other diagnostic medical devices. It does not provide any absolute measurement of temperature and should not be used for sole screening or diagnosis for any disease or condition. The system is not intended to be used as a thermometry device.

The ThermPix Thermovisual Camera is a Telethermographic system which consists of infrared and visual cameras and a touch-screen display. It is a non-contact, non-invasive, and non-radiating infrared system capable of imaging and storing thermal patterns generated by the human body. These images along with other patient-specific data are uploaded via Wifi to the ThermPix Cloud, a secure repository where clinicians can access and augment data from a 3rd party device or computer.

It employs passive infrared emission sensing technology to capture the thermal data and uses proprietary software to display the temperature distribution pattern as an image. It is suitable for imaging adult human targets and can be used in hospitals, acute and sub-acute healthcare settings, clinics, and any environment where is provided by a healthcare professional.

The provided text focuses on the 510(k) summary for the "ThermPix Thermovisual Camera" and its comparison to a predicate device. It details technical specifications and non-clinical testing performed to establish substantial equivalence. However, it explicitly states that clinical testing was not required and therefore does not contain information about a study proving the device meets acceptance criteria related to human performance, ground truth, expert adjudication, or MRMC studies.

I will extract the available information and highlight what is not present based on your request.

Here's the breakdown of the acceptance criteria and the study that was provided (non-clinical):

1. Table of Acceptance Criteria and Reported Device Performance

The acceptance criteria are not explicitly stated as distinct "acceptance criteria" but are implied through the comparison with the predicate device and the non-clinical performance testing. The "reported device performance" is the result of these non-clinical tests.

The study described is non-clinical testing only.

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

• Non-Clinical Testing: Relative Temperature Difference (Thermal Sensitivity)
  • Sample Size: "image data collected from 2 cameras"
  • Data Provenance: Not specified regarding country of origin, but it is laboratory-based data from the manufacturer's testing. It is a prospective test conducted by the manufacturer for the submission.

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

• N/A. The provided text explicitly states: "Human Clinical Performance Testing: Clinical testing was not required to demonstrate the safety and effectiveness of the device." Therefore, no human experts were involved in establishing ground truth for clinical performance. The "ground truth" for the non-clinical tests was established using traceable certified reference black body calibrators and thermocouples.

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

• N/A. As no clinical testing with human readers/experts was performed, there was no adjudication method.

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. No clinical or MRMC studies were performed or described. The device is a "Thermovisual Camera" and the submission does not describe an AI component that would assist human readers in interpretation.

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

• This device is a camera system. Its performance (e.g., accuracy, thermal sensitivity) was tested non-clinically as a standalone device. While there's "proprietary software to display the temperature distribution pattern," the submission doesn't describe an "algorithm" making diagnostic assessments. The output is an image to be interpreted by "qualified technical personnel" requiring "clinical judgement and experience." So, yes, non-clinical standalone performance was assessed (camera system's ability to measure and display thermal patterns), but not in the context of an "algorithm" making diagnostic decisions to be evaluated similar to an AI diagnostic aid.

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

• For the non-clinical thermal accuracy and sensitivity tests, the "ground truth" was established by traceable certified reference black body calibrators and thermocouples, which are highly precise and accurate physical measurement standards.

8. The sample size for the training set:

• N/A. The provided text does not describe any machine learning/AI component that requires a training set. The device's software appears to be for image display and processing, not for diagnostic interpretation or classification.

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

• N/A. As there's no mention of a training set or AI model, this is not applicable.

Ask a specific question about this device