| dc.description.abstract | The leading causes of the sudden cardiac events which often occur suddenly and require
immediate treatment include heart attacks and chronic obstructions to the heart walls.
They are not easy to spot at an early stage because conventional diagnostic tools
are only available for clinical use in developed countries, or in rural areas that have
poor resources. The actual purpose of this study is to create a mobile application
connected portable heart attack and heart wall blockage detection system. The system
utilizes Electrocardiogram (ECG) and Photoplethysmography (PPG) sensors to constantly
monitor cardiac activity in real-time. Support Vector Machine (SVM) is a supervised
machine learning approach and Cross Wavelet Transform (XWT) that applies in data
analysis for effective anomaly detection. The smartphone application also provides
features such as real-time visualization of anomalous cardiac states, alerts, and the
possibility to alert medical professionals. The system exhibited a high level, up to
70% endorsing the use of the combination of ECG and PPG and 85% saying that it
is pertinent in the reduction of heart attack deaths, the study provides remarkable
endorsement of using real-time cardiac monitoring. Based on the survey, 78% of the
respondents supported SVM and XWT supervised machine learning methods, 72%
preferred that the data be stored in the cloud, and 61% wanted an easy interface.
To be able to perform effective cardiac care, more improvement on the sensor cost,
privacy, and reliability is needed. More development will focus on the longevity of
operating time, incorporating more sensors for enhanced prognosis performance, and
extending the learning algorithm on larger databases. This system can reduce the
number of hospital visits, enhance cardiac care for high-risk individuals living in rural
communities, and facilitate earlier detection of cardiac events in areas where health
services are lacking, catering to an urgent need in health care. | en_US |
