Abstract—In recent years, with rapid advancement in Artificial Intelligence technology, several intelligent systems have been developed for human emergency prediction under ambient intelligence. Automatic pain recognition through state-of-the-art deep learning algorithms has attracted much attention recently in smart healthcare informatics. This research presents a Convolutional Neural Network (CNN) approach for detecting audio-based emergency identification during Myocardial Infarction. For evaluation, simulated emergency distress audio signals are recorded during possible myocardial infarction as a private dataset. This work demonstrates an approach to train the deep learning CNN model for multiclass audio samples and deploy it on an edge embedded Artificial Intelligence device Jetson Nano for real-time recognition.
 

Index Terms—ambient intelligence, myocardial infarction, convolutional neural network, emergency distress signal

Cite: H. M. Mohan and S. Anitha, "Real Time Audio-Based Distress Signal Detection as Vital Signs of Myocardial Infarction Using Convolutional Neural Networks," Journal of Advances in Information Technology, Vol. 13, No. 2, pp. 106-116, April 2022.

Copyright © 2022 by the authors. This is an open access article distributed under the Creative Commons Attribution License (CC BY-NC-ND 4.0), which permits use, distribution and reproduction in any medium, provided that the article is properly cited, the use is non-commercial and no modifications or adaptations are made.