Machine learning approaches to mental stress detection: a review

Purpose of Review: Machine Learning has shown exponential growth in ingesting a huge amount of data and give accurate outcomes equivalent to the human level. It provides a glance at the future where complex data, analysis and analytical model together help innumerable people suffering from health issues. This paper reviews the current application of ML in the health sector, their limitation, predictive analysis, and areas that are hard-to- diagnose and need advance research. New Findings: We have reviewed 30 papers on mental stress detection using ML that used Social networking sites, blogs, discussion forums, student’s record, Questioner technique, clinical dataset, real-time data (video, driving task, audio), Bio-signal technology (ECG, EEG), a wireless device and suicidal tendency. Collectively, these studies show high accuracy and potential of ML algorithms in mental health, and which ML algorithm yields the best result. Summary: With the advancement of ML, it has unfolded many areas like traditional clinical trials which are not sufficient to collect all the information about a person. Currently, define under DSM-V stage to detect these illnesses at the preliminary stage, diagnosing and treating before any mishap. It has re-defined the mental health practicing reducing cost and time, making it easier and convenient for patients to reach better health care whenever they need it.