Empowering Healthcare: TinyML for Precise Lung Disease Classification DOI Creative Commons

Youssef Abadade,

Nabil Benamar, Miloud Bagaa

и другие.

Future Internet, Год журнала: 2024, Номер 16(11), С. 391 - 391

Опубликована: Окт. 25, 2024

Respiratory diseases such as asthma pose significant global health challenges, necessitating efficient and accessible diagnostic methods. The traditional stethoscope is widely used a non-invasive patient-friendly tool for diagnosing respiratory conditions through lung auscultation. However, it has limitations, lack of recording functionality, dependence on the expertise judgment physicians, absence noise-filtering capabilities. To overcome these digital stethoscopes have been developed to digitize record sounds. Recently, there growing interest in automated analysis sounds using Deep Learning (DL). Nevertheless, execution large DL models cloud often leads latency, dependency internet connectivity, potential privacy issues due transmission sensitive data. address we Tiny Machine (TinyML) real-time detection by sound recordings, deployable low-power, cost-effective devices like stethoscopes. We trained three machine learning models—a custom CNN, an Edge Impulse LSTM—on publicly available dataset. Our data preprocessing included bandpass filtering feature extraction Mel-Frequency Cepstral Coefficients (MFCCs). applied quantization techniques ensure model efficiency. CNN achieved highest performance, with 96% accuracy 97% precision, recall, F1-scores, while maintaining moderate resource usage. These findings highlight TinyML provide accessible, reliable, tools, particularly remote underserved areas, demonstrating transformative impact integrating advanced AI algorithms into portable medical devices. This advancement facilitates prospect screening

Язык: Английский

Comparative Analysis of DL Models for Early Detection of COVID-19 Using Cough Audio Data DOI
Jagat Ram,

P. Sidharth,

S. Sunil

и другие.

Lecture notes on data engineering and communications technologies, Год журнала: 2025, Номер unknown, С. 209 - 218

Опубликована: Янв. 1, 2025

Язык: Английский

Процитировано

0
6G digital twin and CPS system promote the development of rural architectural planning DOI
Zhai Binqing,

Yicong Yao,

Mohammad Khishe

и другие.

Evolving Systems, Год журнала: 2025, Номер 16(2)

Опубликована: Апрель 16, 2025

Язык: Английский

Процитировано

0
Explainable AI for Respiratory Disease Detection: Leveraging Deep Learning on Patient Audio Data DOI

S.V.R. Madiraju,

Manjula Shenoy K,

Dhanya

и другие.

Research Square (Research Square), Год журнала: 2025, Номер unknown

Опубликована: Май 7, 2025

Abstract Respiratory diseases affect millions of people around the world, making it necessary for reliable and interpretable diagnosis. Lung sound analysis is a non- invasive cost-effective approach detecting respiratory abnormalities such as wheezes crackles, which are critical indicators conditions like Chronic Obstructive Pulmonary Disease (COPD). This study uses machine learn- ing techniques to detect crackles from lung sounds automatically. Leveraging database, 13 Mel-Frequency Cepstral Coeffi- cients (MFCCs) were extracted audio recordings classify abnormalities. While deep learning models achieve high accuracy, their black-box nature limits transparency. proposes an explainable AI (XAI) solu- tion disease classification using signals. ensures interpretability by identifying features influencing predictions train- on publicly available datasets incorporating Local Interpretable Model Agnostic Explanations (LIME). Explainability revealed criti- cal predictions, ensuring model research advances development trustworthy AI-driven diagnostic tools, contributing enhanced transparency in healthcare.

Язык: Английский

Процитировано

0
Empowering Healthcare: TinyML for Precise Lung Disease Classification DOI Creative Commons

Youssef Abadade,

Nabil Benamar, Miloud Bagaa

и другие.

Future Internet, Год журнала: 2024, Номер 16(11), С. 391 - 391

Опубликована: Окт. 25, 2024

Respiratory diseases such as asthma pose significant global health challenges, necessitating efficient and accessible diagnostic methods. The traditional stethoscope is widely used a non-invasive patient-friendly tool for diagnosing respiratory conditions through lung auscultation. However, it has limitations, lack of recording functionality, dependence on the expertise judgment physicians, absence noise-filtering capabilities. To overcome these digital stethoscopes have been developed to digitize record sounds. Recently, there growing interest in automated analysis sounds using Deep Learning (DL). Nevertheless, execution large DL models cloud often leads latency, dependency internet connectivity, potential privacy issues due transmission sensitive data. address we Tiny Machine (TinyML) real-time detection by sound recordings, deployable low-power, cost-effective devices like stethoscopes. We trained three machine learning models—a custom CNN, an Edge Impulse LSTM—on publicly available dataset. Our data preprocessing included bandpass filtering feature extraction Mel-Frequency Cepstral Coefficients (MFCCs). applied quantization techniques ensure model efficiency. CNN achieved highest performance, with 96% accuracy 97% precision, recall, F1-scores, while maintaining moderate resource usage. These findings highlight TinyML provide accessible, reliable, tools, particularly remote underserved areas, demonstrating transformative impact integrating advanced AI algorithms into portable medical devices. This advancement facilitates prospect screening

Язык: Английский

Процитировано

2