ABSTRACT
There is no doubt that depression is spreading its influence all over the globe starting from teenagers to adults. No one in today’s world is free from its clutches. Human–machine interaction that is a part of artificial intelligence has paved its path in detecting this terrible disorder before the onset of its severity. With the help of machine learning (ML) and Internet of things (IoT), the IoT is closely associated with cyber–physical interaction for real-time data processing, providing quick analysis of the brain signals whereas ML helps classify the unknown samples by learning from the known ones, predicting depression in individuals from a much earlier time. In this chapter, we have made a survey of various research work done on electroencephalography (EEG) signals, magnetic resonance imaging (MRI) scans, and various techniques and ML approaches used by them in the detection of depression. The chapter also presents a survey on the number of papers published per year based on both the EEG and MRI reports. Our survey suggested that although a lot more papers have been published on MRI reports for depression prediction, EEG still holds its place due to its pocket-friendly, noninvasive and time-effective factors of accurately detecting the brain’s electrical activity within a short span. We 70have also found out that there is a need of discovering more effective ML techniques for real-time data acquisition and classification for an improved detection of depression in the earliest possible time. In this chapter, we have also tried to demonstrate various IoT–ML-related works that help the society in predicting depression and related seizures, trauma, and injuries occurring within the brain much more effectively in the earliest possible time.
