Depression and Mood Monitoring for Pregnant Women Using IoT-Based Wearables

Depression and Mood Monitoring for Pregnant Women Using IoT-Based Wearables

Delshi Howsalya Devi, R. Naresh, C. N. S. Vinoth Kumar, S. Senthilkumar, Asis Jovin
Copyright: © 2023 |Pages: 24
DOI: 10.4018/979-8-3693-1718-1.ch009
OnDemand:
(Individual Chapters)
Available
$37.50
Current Special Offers
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

Depression is regarded as a mental disease by the World Health Organization (WHO). In the last ten years, mental illnesses, including stress and depression, have increased in prevalence, become a serious public health issue, and had an important influence on society. A significant number of people commit suicide each year as a result of depression in a growing nation like India. Sensors are becoming a commonplace feature of daily life because of technological advancements. Many scientists are working to cure and diagnose using IOT and other technologies. In the suggested approach, a wearable is made to collect the biological. Through the application of cloud computing and other technologies, several researchers are attempting to identify and cure depression. In the proposed work, a wearable device is created to record the bodily markers a clinically depressed individual experiences when under stress. IoT is crucial for collecting, analyzing, and processing data to relieve anxiety and depression.
Chapter Preview
Top

Eeg Signal Processing

To predict the severity of depression, the author employed EEG signal processing. They processed a model using the connections between sleep and depression. Those who are sad frequently have insomnia. Insomnia and hypersomnia are among the sleep problems that affect three-quarters of depressed people[27-32]. The signs of drinking and sleep difficulties have a significant negative effect on quality of life, which raises the risk of suicide. They found that the ANFIS results were marginally superior to the classifier results.

Linear Prediction Coding (LPC)

Suicides and depression are becoming to be serious health issues. The authors in have created a model of an emotional voice recognition system employing the Tamil language using Linear Prediction Coding (LPC) and a parameters-based approach. 90% was the highest recognition rate that could be achieved with LPC algorithms.

Top

I-Vector Technique And Fuzzy Membership Functions

The I-Vector method and fuzzified functions have been chosen to determine the degrees of depression in 20 patients. Reliability, balanced categorization rate, maximum signal to noise ratio, F-Measure, and specificity are used to differentiate between the algorithms. To increase accuracy, prior processing is now required to eliminate any silence that may be included in the audio signals. With an accuracy of 97%, fuzzy membership functions showed to be significantly better [33-36].

Complete Chapter List

Search this Book:
Reset