Introduction Inflammatory bowel disorders may result in abnormal Bowel Sound (BS) characteristics during auscultation. We employ pattern spotting to detect rare BS events continuous abdominal recordings using a smart T-shirt with embedded miniaturised microphones. Subsequently, we investigate the clinical relevance of classification task distinguish patients diagnosed inflammatory disease (IBD) and healthy controls. Methods Abdominal were obtained from 24 IBD varying activity 21 controls...

We analyse pretrained and non-pretrained deep neural models to detect 10-seconds Bowel Sounds (BS) audio segments in continuous data streams. The include MobileNet, EfficientNet, Distilled Transformer architectures. Models were initially trained on AudioSet then transferred evaluated 84 hours of labelled eighteen healthy participants. Evaluation was recorded a semi-naturalistic daytime setting including movement background noise using smart shirt with embedded microphones. collected dataset...

We present the GastroDigitalShirt, a smart T-Shirt for capturing abdominal sounds produced during digestion. The garment prototype embeds an array of eight miniaturised microphones connected to low-power wearable computer and is designed long-term recording. microphone integration shirt wiring layout. With GastroDigitalShirt we monitored different digestion phases over six hours in four healthy participants with no prior gastro-intestinal diseases. collected data were annotated by two...

Background Abdominal auscultation (i.e., listening to bowel sounds (BSs)) can be used analyze digestion. An automated retrieval of BS would beneficial assess gastrointestinal disorders noninvasively. Objective This study aims develop a multiscale spotting model detect BSs in continuous audio data from wearable monitoring system. Methods We designed based on the Efficient-U-Net (EffUNet) architecture 10-second segments at time and spot with temporal resolution 25 ms. Evaluation were collected...

Multiprocess Additive Manufacturing (AM) offers system designers new, exciting computational tools to rapidly realise smart wearable sensing devices in two-dimensional (2D) and 3D shapes. We guide readers through the novel development fabrication process based on a digital co-design framework highlight AM techniques, functional materials, assembly procedures for designing wearables as flexible stretchable on-skin patches, e-textiles, accessories everyday use.

We propose a simulation method to evaluate the performance of garment-embedded contact sensors while performing common Activities Daily Living (ADL). Our comprises four steps: dynamic 3D human body model generation, automated smart garment design, ADL simulation, sensor fitting and displacement evaluation. generated 100 models with varying shapes virtually dressed them three differently fitted T-Shirts. then analysed sensor-body distance ADLs. An Electrocardiogram (ECG) shirt was considered...

We propose a framework to automatically extract body landmarks and related measurements from 3D scans replace manual shape estimation in fitting smart garments. Our comprises five steps: scan acquisition segmentation, 2D image conversion, extraction of using Convolutional Neural Network (CNN), back projection mapping extracted space, tailored garment generation. trained tested the algorithm on 3000 synthetic models estimated required for T-Shirt design. The results show that can successfully...

<sec> <title>BACKGROUND</title> Abdominal auscultation (i.e., listening to bowel sounds (BSs)) can be used analyze digestion. An automated retrieval of BS would beneficial assess gastrointestinal disorders noninvasively. </sec> <title>OBJECTIVE</title> This study aims develop a multiscale spotting model detect BSs in continuous audio data from wearable monitoring system. <title>METHODS</title> We designed based on the Efficient-U-Net (EffUNet) architecture 10-second segments at time and spot...