A novel scalable electrode array and system for non-invasively assessing gastric function using flexible electronics

Abstract

Background: Disorders of gastric function are highly prevalent, but diagnosis often remains symptom-based and inconclusive. Body surface gastric mapping is an emerging diagnostic solution, but current approaches lack scalability and are cumbersome and clinically impractical. We present a novel scalable system for non-invasively mapping gastric electrophysiology in high-resolution (HR) at the body surface.

Methods: The system comprises a custom-designed stretchable high-resolution "peel-and-stick" sensor array (8 × 8 pre-gelled Ag/AgCl electrodes at 2 cm spacing; area 225 cm2 ), wearable data logger with custom electronics incorporating bioamplifier chips, accelerometer and Bluetooth synchronized in real-time to an App with cloud connectivity. Automated algorithms filter and extract HR biomarkers including propagation (phase) mapping. The system was tested in a cohort of 24 healthy subjects to define reliability and characterize features of normal gastric activity (30 m fasting, standardized meal, and 4 h postprandial).

Key results: Gastric mapping was successfully achieved non-invasively in all cases (16 male; 8 female; aged 20-73 years; BMI 24.2 ± 3.5). In all subjects, gastric electrophysiology and meal responses were successfully captured and quantified non-invasively (mean frequency 2.9 ± 0.3 cycles per minute; peak amplitude at mean 60 m postprandially with return to baseline in <4 h). Spatiotemporal mapping showed regular and consistent wave activity of mean direction 182.7° ± 73 (74.7% antegrade, 7.8% retrograde, 17.5% indeterminate).

Conclusions and inferences: BSGM is a new diagnostic tool for assessing gastric function that is scalable and ready for clinical applications, offering several biomarkers that are improved or new to gastroenterology practice.

Keywords: bioelectronics; diagnostics; functional gastrointestinal disorders; gastric motility.

Conflict of interest statement

AG, PD, CNA, and GO hold grants and intellectual property in the field of GI electrophysiology and are members of University of Auckland spin‐out companies: The Insides Company (GO), FlexiMap (PD), and Alimetry (AG, SC, YY, SW, JSTW, PD, CNA and GO). DAC, TCLH, and CV have no relevant conflicts to declare.

© 2022 The Authors. Neurogastroenterology & Motility published by John Wiley & Sons Ltd.

Figures

FIGURE 1

Body Surface Gastric Mapping (BSGM) workflow. (1) Setup includes placement of the sensor array personalized by measurements between anatomic landmarks, skin preparation with conductive gel, and a signal check using live impedance data displayed on the companion App. (2) Gastric activity is captured in HR continuously throughout the recording period (30 m fasted, standardized meal, up to 4 h postprandially). (3) A report of gastric activity is generated following automated signal processing and analyses. This includes a heat map used to infer gastric position by spatial distribution of amplitude, traditional gastric biomarkers including frequency and amplitude, along with novel meal response and spatial wave propagation biomarkers

FIGURE 2

Components of Body Surface Gastric Mapping (BSGM) system. (A) Assembled BSGM system including sensor array, connector clamp, flexible printed circuit cable, and wearable data logger. (B) Close‐up of sensor array mating panel, demonstrating convergence of all 64 conductive tracks. This is opposed with the cable using the connector clamp shown in (C): thumb screws are loosened and the top piece removed to expose the connector piece (middle), which facilitates secure connection between the mating panel and the connector cable (bottom). (D) Companion App used to register test and participant details, customize recording variables, and guide setup of the sensor array and data logger. (D. i) Measurement input interface, where distances between the xiphoid and umbilicus, xiphoid and anterior superior iliac spine (ASIS), and abdominal circumference are recorded to guide personalized array positioning. (D. ii) Signal quality check shown on the App. The size and color of each electrode button represents the impedance measured for each channel

FIGURE 3

High‐resolution sensor array (8 × 8 pre‐gelled Ag/AgCl electrode grid; 2 cm spacing; 64 channels; area 225 cm2) printed upon a flexible TPU substrate, enabling comfort and optimal electrode contact across contours of the abdominal wall. (A) Backing layer partially peeled off, exposing adhesive layer and hydrogel discs. (B) Sensor array placed on a subject's epigastrium while reclined at 45 degrees

FIGURE 4

(A) Spatial heat maps for three participants, demonstrating the estimated signal power in the gastric range over the recording duration. Each electrode is represented by a gray circle, and the subject's left (L) and right (R) are indicated. Approximate stomach location can be inferred by the region of greatest amplitude (yellow). (B) Average heat map for all 24 participants. (C) Box and whisker plot of mean impedance for all 24 subjects

FIGURE 5

(A) Frequency‐amplitude spectrograms (top) and amplitude over time (bottom) for three subjects demonstrating features of the typical meal response. Start of meal time is marked by the blue line. Variability in the meal response was observed, with near‐immediate postprandial increases in amplitude (e.g., (i) and (ii)), or following a lag phase (e.g., [iii]). (B) Average frequency‐amplitude spectrogram for all 24 subjects, after normalization for amplitude. Amplitude steadily increased following the meal to reach a plateau at 1 h postprandially, before gradual return towards baseline (fasting) amplitude by the end of the 4 h postprandial period

FIGURE 6

(A) Box and whisker plots of cohort mean amplitude over 1 h time periods. (B) Scatter plot of mean amplitude and BMI with a linear trend line applied

FIGURE 7

(A) Representative animations of postprandial periods of subject with antegrade activity; (B) representative animations of postprandial periods of subject with (i) retrograde activity immediately postprandially, which returned to antegrade activity soon after (ii); (C) Average polar histogram of all 24 subjects; (D) Average phase map of all 24 subjects