Search Results

The intended use for the G-Tech Patch System is to serve as a tool that provides gastrointestinal myoelectrical activity measurements to be used at the discretion of the Physician or Clinical User to aid in the diagnosis and evaluation of gastrointestinal disorders.

The G-Tech System is a non-invasive wireless gastrointestinal monitoring system for use by the physician or staff in the hospital setting, clinic, or physician's office. The device is intended for use while the patient is under direct supervision (e.g., in hospital or clinic) as well as after discharge to home or to intermediate setting. The device is for prescription use only. The device consists of the G-Tech Patch, the G-Tech Patch Monitor, an iOS application, a secure cloud storage server and computer-based data analysis algorithms. The G-Tech Patch is a single use, wearable electrode patch that reads the electrical signals of the gastrointestinal tract from the abdominal skin surface. The Patch transmits the acquired electrical signals via Bluetooth to the G-Tech Patch Monitor. The G-Tech Patch Monitor receives the raw data, encrypts it, and periodically uploads it to a secure cloud server. Additionally, the Patch Monitor has a patient interface to allow the patient to manual enter events such as meals, bowel movements, pain or the taking of medications. Data analysis algorithms process the uploaded data to measure and report myoelectrical activity for the stomach and intestines. These measurements are made available to physicians for aid in clinical evaluation of their patient.

The provided document describes the G-Tech Wireless Patch System (WPS), a non-invasive device for measuring gastrointestinal myoelectrical activity. The information details the device's characteristics, intended use, and substantial equivalence to predicate devices, through various studies. However, the document does not explicitly state acceptance criteria in a quantitative, measurable format with corresponding reported device performance values in a table. It also does not present a traditional "study" in the sense of a clinical trial with a defined test set, ground truth experts, adjudication, or MRMC design for regulatory clearance.

Instead, the submission focuses on demonstrating substantial equivalence through a combination of:

• Bench Performance Study: Functional testing with known signals to assess the system's ability to extract rhythmic motor activity.
• Animal Study: Comparison of G-Tech Wireless Patches with internal electrodes in mini-pigs to show fidelity of surface-measured signals to internal ones.
• Clinical Studies (Primary Evidence): Simultaneous measurements with manometry and Smart Pill to demonstrate concordance in frequency and time of myoelectrical activity with internal measurements. These appear to be more akin to validation or correlation studies rather than a pass/fail performance study against pre-defined acceptance criteria for diagnostic accuracy.
• Supplemental Evidence: Four peer-reviewed publications demonstrating correlation with clinical markers and reproducibility.

Given the information provided, I will interpret "acceptance criteria and the study that proves the device meets the acceptance criteria" in the context of the available data, which focuses on demonstrating substantial equivalence through various aspects of performance and comparison to established methods, rather than a single, pre-defined algorithm performance study with specific cutoffs.

Here's a breakdown of the requested information based on the provided text, with explicit notes where information is not present:

1. Table of Acceptance Criteria and Reported Device Performance

As mentioned, the document does not provide a quantitative table of acceptance criteria with corresponding reported device performance values for the overall device's diagnostic capability (e.g., sensitivity, specificity for a specific disorder). The "acceptance criteria" for this 510(k) submission appear to be implicit and qualitative: demonstrating that the device performs as intended, is safe, and is substantially equivalent to predicate devices.

The "performance" is described broadly:

• G-Tech Patches are reliable and provide consistent electrical measurement devices.
• Data processing algorithms are efficient engines for extracting rhythmic motor activity signals. |
  | Fidelity to Internal Signals | Detectable slow wave signals at the skin surface (stomach, small intestine, colon) with fidelity to internally measured signals in frequency and spectral shape. | Animal Study demonstrated:
• Slow wave signals associated with the three major GI organs (as observed with internal electrodes) are detectable at the skin surface with the G-Tech Wireless Patches.
• Signals have fidelity to the internally measured signals in frequency and spectral shape. |
  | Clinical Concordance | Records signals identifiable as originating from Stomach, Duodenum/Jejunum, and Colon, with concordance to manometry and Smart Pill data. | Clinical Studies (Primary Evidence) demonstrated:
• Non-invasive patch system records signals identifiable as being from the Stomach, Duodenum/Jejunum, and Colon by concordance with frequency and time of spectral analysis of high-resolution manometry data.
• Concordance in frequency and time observed between G-Tech Wireless Patch measurements and Smart Pill measurements. |
  | Clinical Utility/Correlation | (Implicit) Utility in understanding measurement capabilities and potential clinical use; correlation with clinical markers of postoperative recovery; reproducibility; sensitivity to external stimulus. | Clinical Studies (Supplemental Evidence - Peer-Reviewed Publications) demonstrated:
• Correlation between G-Tech Wireless Patch Measurements and clinical markers of postoperative recovery in adult and pediatric populations.
• Reproducibility of the gastric and intestinal myoelectrical profile over multi-year periods.
• Sensitivity of measurements to an external stimulus in specific examples. |
  | Safety | Electrical safety, EMC, biocompatibility, mechanical strength. | Compliance with IEC 60601-1, IEC 60601-1-11, IEC 60601-1-6, IEC 60601-1-2, ANSI C63.27-2017.
  Biocompatibility assessed with ISO 10993-1:2009 (compliant for intact skin up to 30 days).
  Mechanical strength testing (wear adhesion, tensile, impact, drop, shock, vibration) demonstrates performance as intended and safety. |
  | Software Performance | Software acquires and processes myoelectrical signals, generates reports as intended. | Software documentation developed in consideration of "Guidance for the Content of Premarket Submissions for Software Contained in Medical Devices" and "Content of Premarket Submissions for the Management of Cybersecurity in Medical Devices". (Implies verification and validation were done to meet these guidelines). |

2. Sample Size Used for the Test Set and Data Provenance

The document does not explicitly define a "test set" in the context of an AI/ML model for a diagnostic study, but rather describes various validation/correlation studies.

• Bench Performance Study: Known signals were applied to the system, but the specific number/type of signals is not quantified as a "sample size."
• Animal Study: Data from "a study" undertaken to compare measurements in "adolescent Yucatan mini-pigs." The number of animals is not specified. Data provenance: likely laboratory/research setting (not specified country).
• Clinical Studies (Primary Evidence): Describes "simultaneous measurements between G-Tech Wireless Patches and antroduodenal and colonic manometry" and "concordance... between G-Tech Wireless Patch measurements and Smart Pill measurements." The number of human subjects/patients involved in these "primary evidence" studies is not specified. Data provenance: "IRB-approved studies and well-documented case histories," implying prospective or retrospective for specific cases. Country of origin is not specified, but the applicant is US-based.
• Clinical Studies (Supplemental Evidence - Peer-Reviewed Publications): These are external studies. The document only lists the publications without detailing their sample sizes or methodology beyond their titles. To get this info, one would need to refer to the publications themselves.

3. Number of Experts Used to Establish Ground Truth and Qualifications of Those Experts

This information is not provided in the document. The "ground truth" for the device's performance appears to be established through:

• Known signals (bench testing).
• Internal electrodes (animal study).
• Established clinical methods like manometry and Smart Pill (clinical studies).

There is no mention of human expert consensus or adjudication being used as a "ground truth" reference for this device's output. The device provides "measurements" to be used at the "discretion of the Physician or Clinical User," suggesting the physician interprets the device's data rather than the device providing a definitive "diagnosis."

4. Adjudication Method for the Test Set

This information is not applicable/provided. No "test set" in the context of expert review with adjudication is described.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study was Done, What was the Effect Size of How Much Human Readers Improve with AI vs. Without AI Assistance?

This information is not provided and such a study design is not described for this device. The G-Tech WPS provides "measurements" for physicians to use in their diagnosis; it is not described as an AI-powered diagnostic aid that re-interprets medical images or provides a direct diagnostic output that a human then uses. Therefore, a study on human reader improvement with/without AI assistance isn't relevant to the information presented for this device.

6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) was Done?

The document states: "Data analysis algorithms process the uploaded data to measure and report myoelectrical activity for the stomach and intestines." The "Bench Performance Study" and "Animal Study" would represent aspects of standalone algorithm performance, as they assess how well the system, through its algorithms, extracts and measures signals compared to known inputs or internal direct measurements.

• Bench Performance Study: Confirmed algorithms are "efficient engines for extracting rhythmic motor activity signals."
• Animal Study: Confirmed surface signals "have fidelity to the internally measured signals in frequency and spectral shape."

7. The Type of Ground Truth Used (Expert Consensus, Pathology, Outcomes Data, etc.)

The types of "ground truth" or reference standards used are:

• For Bench Testing: Known, synthetic electrical signals with realistic background, artifact, and noise characteristics.
• For Animal Study (Fidelity): Simultaneous measurements using "multi-element electrodes sutured to serosal surfaces of stomach, small intestine, and colon." (i.e., direct internal physiological measurements).
• For Clinical Studies (Concordance):
  • High-resolution manometry data (for antroduodenal and colonic myoelectrical activity).
  • Smart Pill measurements (for gastrointestinal transit and related activity).
• For Clinical Utility/Correlation (Supplemental Evidence): Clinical markers of postoperative recovery.

8. The Sample Size for the Training Set

This information is not provided. The document describes the device as a measurement system, not a machine learning model developed with explicit training and test sets in the typical AI/ML sense (though data analysis algorithms are present). If the algorithms involved machine learning, the training set details would be critical, but they are not mentioned here.

9. How the Ground Truth for the Training Set Was Established

This information is not provided, as details regarding a specific "training set" are absent.

Summary of Missing Information critical for a typical AI/ML Diagnostic Study:

• Quantitative acceptance criteria for diagnostic performance (e.g., sensitivity, specificity, AUC).
• Sample sizes for human clinical "test sets."
• Details on expert involvement, qualifications, and adjudication methods for any human interpretation studies.
• MRMC study results for human-AI collaboration.
• Explicit training set details, including size and ground truth establishment, for the data analysis algorithms (if they are AI/ML-based algorithms requiring such, which is not explicitly stated).

The provided text focuses on the device's technical performance (signal acquisition, processing, safety) and its correlation/concordance with existing physiological measurement techniques, supporting its claim of substantial equivalence as an electrogastrography system.

Ask a specific question about this device