A mathematical model of bowel sound generation

Multiscale Bowel Sound Event Spotting in Highly Imbalanced Wearable Monitoring Data: Algorithm Development and Validation Study

BACKGROUND

Abdominal auscultation (i.e., listening to bowel sounds (BSs)) can be used to analyze digestion. An automated retrieval of BS would be beneficial to assess gastrointestinal disorders noninvasively.

OBJECTIVE

This study aims to develop a multiscale spotting model to detect BSs in continuous audio data from a wearable monitoring system.

METHODS

We designed a spotting model based on the Efficient-U-Net (EffUNet) architecture to analyze 10-second audio segments at a time and spot BSs with a temporal resolution of 25 ms. Evaluation data were collected across different digestive phases from 18 healthy participants and 9 patients with inflammatory bowel disease (IBD). Audio data were recorded in a daytime setting with a smart T-Shirt that embeds digital microphones. The data set was annotated by independent raters with substantial agreement (Cohen κ between 0.70 and 0.75), resulting in 136 hours of labeled data. In total, 11,482 BSs were analyzed, with a BS duration ranging between 18 ms and 6.3 seconds. The share of BSs in the data set (BS ratio) was 0.0089. We analyzed the performance depending on noise level, BS duration, and BS event rate. We also report spotting timing errors.

RESULTS

Leave-one-participant-out cross-validation of BS event spotting yielded a median F1-score of 0.73 for both healthy volunteers and patients with IBD. EffUNet detected BSs under different noise conditions with 0.73 recall and 0.72 precision. In particular, for a signal-to-noise ratio over 4 dB, more than 83% of BSs were recognized, with precision of 0.77 or more. EffUNet recall dropped below 0.60 for BS duration of 1.5 seconds or less. At a BS ratio greater than 0.05, the precision of our model was over 0.83. For both healthy participants and patients with IBD, insertion and deletion timing errors were the largest, with a total of 15.54 minutes of insertion errors and 13.08 minutes of deletion errors over the total audio data set. On our data set, EffUNet outperformed existing BS spotting models that provide similar temporal resolution.

CONCLUSIONS

The EffUNet spotter is robust against background noise and can retrieve BSs with varying duration. EffUNet outperforms previous BS detection approaches in unmodified audio data, containing highly sparse BS events.

A Human Defecation Prediction Method Based on Multi-Domain Features and Improved Support Vector Machine

The difficulty of defecation seriously affects the quality of life of the bedridden elderly. To solve the problem that it is difficult to know the defecation time of the bedridden elderly, this paper proposed a human pre-defecation prediction method based on multi-domain features and improved support vector machine (SVM) using bowel sound as the original signal. The method includes three stages: multi-domain features extraction, feature optimization, and defecation prediction. In the stage of multi-domain features extraction, statistical analysis, fast Fourier transform (FFT), and wavelet packet transform are used to extract feature information in the time domain, frequency domain, and time-frequency domain. The symmetry of the bowel sound signal in the time domain, frequency domain, and time-frequency domain will change when the human has the urge to defecate. In the feature optimization stage, the Fisher Score (FS) algorithm is introduced to select meaningful and sensitive features according to the importance of each feature, aiming to remove redundant information and improve computational efficiency. In the stage of defecation prediction, SVM is optimized by the gray wolf optimization (GWO) algorithm to realize human defecation prediction. Finally, experimental analysis of the bowel sound data collected during the study is carried out. The experimental result shows that the proposed method could achieve an accuracy of 92.86% in defecation prediction, which proves the effectiveness of the proposed method.

Body Acoustics for the Non-Invasive Diagnosis of Medical Conditions

In the past few decades, many non-invasive monitoring methods have been developed based on body acoustics to investigate a wide range of medical conditions, including cardiovascular diseases, respiratory problems, nervous system disorders, and gastrointestinal tract diseases. Recent advances in sensing technologies and computational resources have given a further boost to the interest in the development of acoustic-based diagnostic solutions. In these methods, the acoustic signals are usually recorded by acoustic sensors, such as microphones and accelerometers, and are analyzed using various signal processing, machine learning, and computational methods. This paper reviews the advances in these areas to shed light on the state-of-the-art, evaluate the major challenges, and discuss future directions. This review suggests that rigorous data analysis and physiological understandings can eventually convert these acoustic-based research investigations into novel health monitoring and point-of-care solutions.

Automated Bowel Sound Analysis: An Overview

Despite technological progress, we lack a consensus on the method of conducting automated bowel sound (BS) analysis and, consequently, BS tools have not become available to doctors. We aimed to briefly review the literature on BS recording and analysis, with an emphasis on the broad range of analytical approaches. Scientific journals and conference materials were researched with a specific set of terms (Scopus, MEDLINE, IEEE) to find reports on BS. The research articles identified were analyzed in the context of main research directions at a number of centers globally. Automated BS analysis methods were already well developed by the early 2000s. Accuracy of 90% and higher had been achieved with various analytical approaches, including wavelet transformations, multi-layer perceptrons, independent component analysis and autoregressive-moving-average models. Clinical research on BS has exposed their important potential in the non-invasive diagnosis of irritable bowel syndrome, in surgery, and for the investigation of gastrointestinal motility. The most recent advances are linked to the application of artificial intelligence and the development of dedicated BS devices. BS research is technologically mature, but lacks uniform methodology, an international forum for discussion and an open platform for data exchange. A common ground is needed as a starting point. The next key development will be the release of freely available benchmark datasets with labels confirmed by human experts.

Noninvasive Diagnosis of Irritable Bowel Syndrome via Bowel Sound Features: Proof of Concept

Irritable bowel syndrome (IBS) is a common and debilitating disorder estimated to affect approximately 11% of the world's population. Typically, IBS is a diagnosis of exclusion after patients undergo a costly and invasive colonoscopy to exclude organic disease. Clinician's and researchers have identified a need for a new cost-effective, accurate, and noninvasive diagnostic test for IBS.

Bowel Sounds Identification and Migrating Motor Complex Detection with Low-Cost Piezoelectric Acoustic Sensing Device

Interpretation of bowel sounds (BS) provides a convenient and non-invasive technique to aid in the diagnosis of gastrointestinal (GI) conditions. However, the approach's potential is limited by variation between BS and their irregular occurrence. A short, manual auscultation is sufficient to aid in diagnosis of only a few conditions. A longer recording has the potential to unlock additional understanding of GI physiology and clinical utility. In this paper, a low-cost and straightforward piezoelectric acoustic sensing device was designed and used for long BS recordings. The migrating motor complex (MMC) cycle was detected using this device and the sound index as the biomarker for MMC phases. This cycle of recurring motility is typically measured using expensive and invasive equipment. We also used our recordings to develop an improved categorization system for BS. Five different types of BS were extracted: the single burst, multiple bursts, continuous random sound, harmonic sound, and their combination. Their acoustic characteristics and distribution are described. The quantities of different BS during two-hour recordings varied considerably from person to person, while the proportions of different types were consistent. The sensing devices provide a useful tool for MMC detection and study of GI physiology and function.