Salvato in:
Dettagli Bibliografici
Autori principali: Matynia, Igor, Nowak, Robert
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:https://arxiv.org/abs/2504.08659
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866917982656004096
author Matynia, Igor
Nowak, Robert
author_facet Matynia, Igor
Nowak, Robert
contents Sound events representing intestinal activity detection is a diagnostic tool with potential to identify gastrointestinal conditions. This article introduces BowelRCNN, a novel bowel sound detection system that uses audio recording, spectrogram analysys and region-based convolutional neural network (RCNN) architecture. The system was trained and validated on a real recording dataset gathered from 19 patients, comprising 60 minutes of prepared and annotated audio data. BowelRCNN achieved a classification accuracy of 96% and an F1 score of 71%. This research highlights the feasibility of using CNN architectures for bowel sound auscultation, achieving results comparable to those of recurrent-convolutional methods.
format Preprint
id arxiv_https___arxiv_org_abs_2504_08659
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle BowelRCNN: Region-based Convolutional Neural Network System for Bowel Sound Auscultation
Matynia, Igor
Nowak, Robert
Sound
Audio and Speech Processing
68
J.3
Sound events representing intestinal activity detection is a diagnostic tool with potential to identify gastrointestinal conditions. This article introduces BowelRCNN, a novel bowel sound detection system that uses audio recording, spectrogram analysys and region-based convolutional neural network (RCNN) architecture. The system was trained and validated on a real recording dataset gathered from 19 patients, comprising 60 minutes of prepared and annotated audio data. BowelRCNN achieved a classification accuracy of 96% and an F1 score of 71%. This research highlights the feasibility of using CNN architectures for bowel sound auscultation, achieving results comparable to those of recurrent-convolutional methods.
title BowelRCNN: Region-based Convolutional Neural Network System for Bowel Sound Auscultation
topic Sound
Audio and Speech Processing
68
J.3
url https://arxiv.org/abs/2504.08659