Search Results

The devices are intended to measurement the human body temperature, and specifically indicated to operate in either predictive mode or regular mode orally, rectally or under the arm.

The device is an electronic thermometer with a hinged probe and an LCD display. The device was designed to measure human body predictive temperature (fast mode) around four seconds with predictive algorithm technology. If the predictive temperature cannot be measured, the device will take actual temperature (regular mode) automatically.

Here's an analysis of the provided text regarding the acceptance criteria and study for the Digital Clinical Thermometer Model KD-193. It's important to note that the provided text is a 510(k) summary for a digital clinical thermometer, not an AI/ML device. Therefore, many of the typical acceptance criteria and study details associated with AI/ML devices (like sample sizes for test sets, expert ground truth, adjudication methods, MRMC studies, standalone performance, training sets, etc.) are not applicable or not present in this type of submission.

The "study" here refers to demonstrating compliance with established performance standards for medical devices, rather than a clinical trial validating an AI algorithm's diagnostic accuracy.

Acceptance Criteria and Device Performance

The provided 510(k) summary for the Digital Clinical Thermometer Model KD-193 focuses on compliance with established medical device standards. Explicit numerical acceptance criteria are not detailed in the given text for various performance metrics, but rather the device's ability to meet specific standards is the "acceptance criteria." The reported device performance is its compliance with these standards.

Note: The ASTM E1112-00 standard for electronic thermometers includes specific accuracy requirements (e.g., within ±0.1°C or ±0.2°F for certain ranges), but these specific numerical criteria are not explicitly stated in the provided text, only the compliance with the standard is mentioned.

Study Details (as inferable from a non-AI/ML device submission)

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

   • Not applicable / Not explicitly stated for this type of device. For a basic clinical thermometer, "testing" involves verification that the device meets the physical and electrical requirements of the cited standards. This typically involves calibration against known standards and environmental testing, not a "test set" in the context of an AI algorithm. The text doesn't provide details on the number of units tested or the origin of any "data" beyond its design and manufacturing.

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

   • Not applicable. Ground truth in the context of AI refers to accurate labels for data. For a thermometer, the "ground truth" is established by highly accurate reference thermometers used in calibration laboratories, and the expertise lies in metrology and engineering, not medical interpretation by human experts in a clinical setting for a test set.

3. Adjudication method for the test set:

   • Not applicable. Adjudication methods like 2+1 or 3+1 are used for resolving discrepancies in expert labeling of medical images or data for AI model training/testing. This is not relevant to the performance testing of a physical electronic thermometer.

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:

   • Not applicable. MRMC studies are specific to evaluating the impact of AI on human readers (e.g., radiologists, pathologists). This device is a standalone measurement tool, not an AI diagnostic aid for human interpretation.

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

   • Yes, in spirit, but not in the AI sense. The device itself is standalone in its function as a temperature measurement device. Its "predictive algorithm" is an internal function to estimate temperature quickly. The performance data is the standalone performance of the device without human interpretation of its output beyond reading the LCD display. However, this is not "standalone AI algorithm performance" in the context of diagnostic AI.

6. The type of ground truth used:

   • Reference standards/Metrology. For a thermometer, the "ground truth" for its accuracy is established by comparing its readings to highly accurate, calibrated reference thermometers or temperature standards in a controlled laboratory environment.

7. The sample size for the training set:

   • Not applicable / Not explicitly stated. For the "predictive algorithm" mentioned, there would have been some form of data collection (e.g., simultaneous readings of the probe and a reference thermometer over time) to develop and "train" the algorithm to predict final temperature rapidly. However, the document does not provide details on the sample size of such data.

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

   • Not explicitly stated, but inferred to be reference temperature readings. For the predictive algorithm, "ground truth" would be the actual stable body temperature measured over a longer period by a highly accurate reference thermometer, against which the rapid, predictive readings of the device would be compared and the algorithm would be tuned. This would be established through controlled human or animal subject studies, or simulations, where stable temperatures are accurately measured via reference instruments.

Ask a specific question about this device