China 
Africa 
Explore chapters and articles related to this topic
Spectrogram Image Textural Descriptors for Lung Sound Classification
Published in Om Prakash Jena, Bharat Bhushan, Nitin Rakesh, Parma Nand Astya, Yousef Farhaoui, Machine Learning and Deep Learning in Efficacy Improvement of Healthcare Systems, 2022
Bhakti Kaushal, Smitha Raveendran, Mukesh D. Patil, Gajanan K. Birajdar
Various respiratory conditions are assessed through lung auscultation and spirometry [1]. Spirometry is one of the simple and useful lung function measurement techniques. It helps in measuring the time taken and the volume of the air inhaled and exhaled by an individual. One major limitation of this technique is the cooperation provided by an individual, and there is a chance of high potential error if the effort put by the individual is substandard. Also, due to its high cost and the challenges faced by clinicians to guide patients, not many clinical settings can use it [1]. Lung auscultation is another method to assess respiratory sounds, measured with the help of a stethoscope. It is placed on the patient's anterior and posterior chest by a trained expert clinician so that they can listen and understand the sounds, which can help in correct detection between normal and abnormal sounds. The main limitation of the conventional auscultation method is that it does not continuously monitor the respiratory sounds. It requires an expert clinician to understand the findings, which can vary individually [1]. These limitations can be overcome by an automated diagnostic system for respiratory sounds [7]. Hence, we have used the International Conference on Biomedical and Health Informatics (ICBHI) 2017 challenge data set [1] and built an algorithm in this work such that digital auscultation can also be used at primary care centers and homes.
Assessing the accuracy of artificial intelligence enabled acoustic analytic technology on breath sounds in children
Published in Journal of Medical Engineering & Technology, 2022
Zai Ru Cheng, Huiyu Zhang, Biju Thomas, Yi Hua Tan, Oon Hoe Teoh, Arun Pugalenthi
In this study, the single label classification was performed instead of a multi-label one, even though in a real-life paediatric clinical context, adventitious sounds can occur in the “multi-label mode”, consisting of both crackles with wheeze, and even additional respiratory sounds such as stridor, stertor and other transmitted sounds. The advanced (multi-label) classifier with greater capability in detecting and interpreting such clinical scenarios has been considered for upcoming research and development.