Search Results

The Proteus Personal Monitor is a miniaturized, wearable data-logger for ambulatory recording of heart rate, activity, body angle relative to gravity, and time-stamped, patientlogged events, including events signaled by swallowing the Ingestion Event Marker (IEM) accessory. The Proteus Personal Monitor enables unattended data collection for clinical and research applications. The Proteus Personal Monitor may be used in any instance where quantiffable analysis of event-associated heart rate, activity, and body position is desirable.

The Proteus Personal Monitor, also called the "Patch", is a body-worn sensor that collects physiological and behavioral metrics including heart rate, activity, body angle and time-stamped user-logged events generated when a user marks an event by swallowing an Ingestion Event Marker (IEM) or by manually pressing an event marker button on the Patch. The Patch stores and wirelessly sends the IEM data to a general computing device.

The Proteus Personal Monitor Ingestion Event Marker system is comprised of three main subsystems; (1) the ingestion event marker (IEM), (2) the data recorder (Patch), and (3) the Proteus software.

1. Ingestion Event Marker (IEM)
   The grain-of-sand sized IEM is designed to communicate the time-stamped confirmation of IEM device ingestion as a unique identifier to the Proteus Personal Monitor worn on the skin. The ingestion signal is communicated via volume conduction communication also known as intrabody communication. The IEM is attached to an inert pharmaceutical excipient tablet for ease of handling and swallowability.

2. Proteus Personal Monitor (Patch)
   The Proteus Personal Monitor (Patch) receives, stores, and wirelessly sends ingestion confirmation data to a general computing device.

3. Software
   The Proteus software is used to pair the Patch with a mobile computing device. The software organizes and displays ingestion events.

Here's the breakdown of the acceptance criteria and the study details for the Proteus Personal Monitor including Ingestion Event Marker:

Acceptance Criteria and Device Performance

Study Information

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

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

• Human Clinical Test Set:

  • Number of subjects: 219 subjects ingested IEMs across multiple studies, with 254 subjects wearing the Proteus Personal Monitor (Patch) in total.
  • Number of IEM ingestions: 11,655 cumulative IEM ingestions.
  • Subject/days of system use: 3,811 cumulative subject-days.
  • Data Provenance: Not explicitly stated, but the mention of "The PROMITTER substudy was conducted on the Proteus campus and enrolled 5 healthy volunteer subjects" suggests some studies were conducted in the US. The overall clinical data likely comes from studies conducted by Proteus Biomedical, Inc. (based in Redwood City, CA). The data appears to be prospective as it involves active participation in clinical trials.

• Animal Test Set:

  • Number of animals: Forty-two (42) in-vivo studies in rodent, canine, and porcine models. The specific number of animals per study is not detailed.
  • Data Provenance: Not explicitly stated, but likely from studies commissioned by Proteus Biomedical, Inc. and conducted in laboratory settings. These are prospective animal studies.

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

• Clinical Studies (IEM Ingestion Detection): The document does not specify the number of experts or their qualifications for establishing the ground truth of IEM ingestion events in the clinical studies. Typically, this would involve direct observation, patient logs, or other objective measures as the ground truth for "ingestion event" rather than expert interpretation of a signal.
• ECG Performance Testing: For the MIT-BIH arrhythmia database, the ground truth is established by cardiologists and is inherent to the database itself. For the PROMITTER substudy, it is not specified how the ECG ground truth was established, but typically this would involve a reference ECG device and interpretation by cardiology experts.

4. Adjudication method for the test set:

• The document does not explicitly describe an adjudication method (like 2+1 or 3+1) for disagreement among experts, which typically applies to subjective interpretations. For IEM ingestion events, the ground truth is likely directly observed or logged, rather than relying on interpretation that would require adjudication.

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, a multi-reader multi-case (MRMC) comparative effectiveness study focusing on human readers improving with or without AI assistance was not conducted or reported in this document. The device's primary function is to automatically detect ingestion events, not to assist human readers in interpreting medical images or other data that typically involves MRMC studies.

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

• Yes, the core performance metrics "Positive Detection Accuracy (PDA)" and "Negative Detection Accuracy (NDA)" for IEM ingestion detection represent the standalone (algorithm only) performance of the Proteus Personal Monitor system. The system automatically detects and logs these events without direct human-in-the-loop intervention for each individual ingestion detection.

7. The type of ground truth used:

• IEM Ingestion Detection: The ground truth for IEM ingestion events in human clinical studies is implicitly based on documented ingestions by the subjects. This would likely involve patient logs, direct observation during controlled settings, or other objective confirmation of "patient-logged events" as described in the Indications for Use.
• Physiological Measurements (ECG, Accelerometer, Respiratory Rate): For these, the ground truth was established using:
  • Standard databases: MIT-BIH arrhythmia database for ECG performance.
  • Reference measurements: Applied values for accelerometers, controlled breathing rates for respiratory rate.
• Safety (Toxicity, Biocompatibility): Ground truth established through established laboratory testing protocols, histopathology, clinical observations, and established toxicology assessments by experts (e.g., Gradient Corp for metal toxicology).

8. The sample size for the training set:

• The document does not provide details on the sample size used for training the algorithms. This information is typically proprietary or not included in regulatory summaries focused on validation.

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

• The document does not provide details on how the ground truth for the training set was established. Similar to the test set, it would likely involve objective data related to IEM ingestion, physiological signals with known parameters, or established clinical diagnoses for specific outcomes targeted by any machine learning components.

Ask a specific question about this device