z-logo
Premium
Median Filter Data Analysis of Bowel Activity using Wireless Intracolonic Sensor
Author(s) -
Cabal Dario,
Majerus Steve,
Hacohen Yaneev,
Hanzlicek Brett,
Damaser Margot,
Bourbeau Dennis
Publication year - 2022
Publication title -
the faseb journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.709
H-Index - 277
eISSN - 1530-6860
pISSN - 0892-6638
DOI - 10.1096/fasebj.2022.36.s1.r4866
Subject(s) - bluetooth , wireless , wearable computer , battery (electricity) , pressure sensor , medicine , computer science , biomedical engineering , engineering , embedded system , telecommunications , mechanical engineering , power (physics) , physics , quantum mechanics
Hypothesis The mechanical and neurological mechanisms contributing to intractable chronic constipation and fecal incontinence are not yet fully understood, in part because of challenges in measuring the slow actions of the bowel. To enable multimodal data collection with a route to endoscopic implantation, we developed the Colonic Monitor of Conscious Activity (ColoMOCA). The ColoMOCA was designed for use with electrophysiology tools and to be affixed to the colon wall. After implant, the ColoMOCA wirelessly transmits colonic pressures and electrical properties of stool over multiple days. In this analysis, we are determining the capability of the ColoMOCA device in qualitatively observing slower bowel activity through the use of a median filter analysis. Study Design and Methods The ColoMOCA included battery‐powered pressure‐sensing electronics, stainless‐steel electrodes for measuring salinity of bowel contents. The ColoMOCA measured pressure from two locations 63 mm apart, and electrically measured stool conductivity in 3 overlapping regions (approximately 36, 28, and 20 cm long). The ColoMOCA transmitted data 10 times per second to a pager‐like wearable radio for ambulatory data recording. Colonic pressures and electrode readings were made with animals fully conscious and untethered. Data was received by a wireless radio worn by the animal. The radio recorded data to an onboard memory card, and forwarded data over Bluetooth to a computer that plotted data in real time. During recordings the animal ate food, walked, urinated, and defecated freely. Recorded raw pressure and electrode data were analyzed in MATLAB. The raw pressure data was downsampled to match the sampling rate of the electrode data. Following this, all data were passed through a 13th order median filter in order to observe slower colonic activity (occurring over the span of several seconds) in the animal bowel. Peak locations were then annotated by eye, where the annotator had no prior knowledge of physiological expectations of colonic activity when making annotations. The difference between the peaks was taken as period data for colonic slow waves and placed into histograms containing 10 bins. Results After applying a downsample and median filter analysis to the data, slower waves of colon activity were visible in both the pressure and electrode data. Within these plots, there exist portions of time where the slow wave activity peaks seem to align relatively well between sensor probes. Once the slow waves were available, manual annotation by eye (Figure 1) showed a median pressure wave period that ranged from roughly 48‐92 sec and a median length of 76 sec and 59 sec for the respective pressure probes: P1 and P2 (Figure 2). Electrode wave activity showed longer activity and ranged from 48‐142 sec, with a median period length of roughly 83 sec, 93 sec, 99 sec for electrode probes: E1, E2, and E3. Conclusion These results display the viability of qualitatively observing longer trends in the ColoMOCA pressure and electrode data to observe median filter analysis. These results provide a prelude to a more quantitative analysis with regards the ColoMOCA device capability of observing slower wave activity of the bowel.

This content is not available in your region!

Continue researching here.

Having issues? You can contact us here