A Machine Learning Approach for Early COVID-19 Symptoms Identification

Symptom identification and early detection are the first steps towards a health condition diagnosis. The COVID-19 virus causes pneumonia-like symptoms such as fever, cough, shortness of breath. Many contraction tests necessitate extensive clinical protocols in medical settings. Clinical studies help with accurate analysis COVID-19, where has already spread to lungs most patients. majority existing supervised machine learning-based disease techniques based on data like x-rays computerized tomography. This is heavily reliant larger study does not emphasize symptom detection. aim this investigate anomalies patient physiological for identification. In context, two prevalent symptoms, fever were examined two-fold manner utilizing an unsupervised learning model. To examine progression, features from chest-worn device analyzed. First, Single Vector Activity Index (SVAI) parameter proposed monitor breathing cough patterns. Second, dataset's variance using DBSCAN method clustering outlier Finally, model accuracy evaluated identify outliers real-time feature dissimilarities, yielding overall 90.34%.