Processing of Healthcare Data to Investigate the Correlations and the Anomalies

Fitness, safety and data monitoring wearables have become extremely prevalent. It is no wonder that sedentary lifestyle will lead to a variety of health and personal problems, including excess weight, the onset of chronic and acute illness, and perhaps low productivity at school, at work and even in everyday life. Monitoring fitness has been extensively pursued by the business and university researchers, focusing on sleep pattern, cardiac wellbeing, continuous activity tracking, overall wellness, recovery from illness & diseases, and many more. Researchers have used deep learning techniques; machine learning algorithms and a broad variety of statistical techniques to forecast and identify the trends, patterns, and deviations in the healthcare data collected through the wearable devices. The purpose of this study is to collect the data using Fitbit Inspire HR and Fitbit Versa. The information is extracted automatically using the Fitbit API. The study also explores and investigates any anomalies, trends and correlations inside the Fitbit healthcare data. The K-means algorithm is used to classify the data based on the parameters of the Fitbit healthcare data.