Depression Detection Using Multimodal Analysis with Chatbot Support | IEEE Conference Publication | IEEE Xplore

Depression Detection Using Multimodal Analysis with Chatbot Support


Abstract:

Depression, a widespread psychiatric disorder affecting people globally, spans all age groups, predominantly impacting adults. This bipolar disorder characterized by symp...Show More

Abstract:

Depression, a widespread psychiatric disorder affecting people globally, spans all age groups, predominantly impacting adults. This bipolar disorder characterized by symptoms including pessimism, hopelessness, anhedonia, and sadness, significantly influences lives, contributing to depression. Our paper proposes a multi-model approach for depression detection, utilizing facial expression analysis, audio evaluation, and user text input through deep learning algorithms, alongside an intelligent chatbot for personalized support. This hybrid model integrates facial expressions, audio features, and textual input for a comprehensive approach to depression detection. The methodology includes four key objectives: a CNN model for real-time or pre-recorded video facial expression analysis, audio evaluation using an NLP algorithm to transcribe users' voices, text-based analysis uncovering linguistic patterns and emotional context, and Multimodal Fusion integrating outputs for a unified multimodal approach. The intelligent chatbot encourages users to share emotions openly, enhancing the system's accuracy in identifying individuals at risk of depression. Results demonstrate the fusion's contribution to early depression detection, enabling timely interventions and improving accuracy, efficiency, and overall performance.
Date of Conference: 15-16 March 2024
Date Added to IEEE Xplore: 11 April 2024
ISBN Information:
Conference Location: Greater Noida, India

I. Introduction

Depression is one of the most popular debilitating mental health problems that affects millions of people globally. WHO (World Health Organization) defines it as a persistent mental health condition that involves continuous sadness and unhappiness in various activities. [1]. According to the survey of 2021 more than 280 million people are affected by depression and for it, early detection of depression is needed.[2] Depression, being a multifaceted disorder, often manifests through an intricate tapestry of signs and symptoms encompassing diverse channels of communication. It mostly affects an individual's ability to learn anything and also causes various types of mood swings and it also reduces various capabilities of human beings as the working capacity of human beings becomes reduced in a depressed person [3]. These include verbal expressions, vocal intonations, and written text, each serving as a distinct medium through which the emotional landscape of an individual can be deciphered. As a result, the ability to recognize and interpret this diverse array of cues, spanning both linguistic and non-linguistic dimensions of communication, holds the potential to offer profound insights into the mental state of individuals Various traditional therapies and methods for detecting depression are time-consuming and expensive and most of them are ineffective. Some of these old therapies are such as psychotherapy or pharmacological [4]. One of the major problems with these types of traditional therapies is that they require more information related to the patient's need, and patient history for providing various types of information and require continuous monitoring of people's health activities. And secondly, patients feel uncomfortable telling doctors about their mental condition due to their fear of society [5]. Various automatic systems are used for detecting depression in the early stages so that doctors can treat patients as soon as possible. Some of the tools needed earlier are an assessment system and an interview -style system. The assessment process includes methods such as the Hamilton Rating scale and self-reporting techniques that contain the Beck Depression Inventory tool and structured clinical interviews, and PHQ-8 scores that detect symptoms of depression and common action in patients for detecting depression in the early stage of life [6]. With the rise of various technologies Like Artificial Intelligence, Machine Learning, Blockchain, and Fuzzy Logic various techniques are formed that detect emotions in human beings very easily. Some papers detect depression using a Text-based system by using sentiment analysis for users' various types of tweets and posts on various types of social networking sites [7]. Various machine learning algorithms like Support Vector Machines SVM, KNN, LSTM, and Naive-Bayes. They are used for detecting depression and these algorithms need a confusion matrix for evaluating results without using a confusion matrix they face problems in analyzing results. In some research papers machine learning algorithms such as PCA (Principal Component Analysis) [8], and KNN are used that helps to extract various facial features for detecting depression these algorithms only extract facial features LSTM, Linear Support Vector, RNN, and Logistic Regression these algorithm helps to detect depression through text from various tweets [9]. Depression detection from textual features or only from facial expressions does not provide the correct level of depression. It does not provide high accuracy about the mental state of people and to remove this problem we provide an approach that provides a fusion of CNN and NLP that provides a comprehensive understanding of an individual's mental state and also provides a chatbot that provides a personalized health support. Here In this paper, we are using CNN and NLP for detecting depression which help in increasing accuracy for detecting depression and also helps in the early detection of depression.

Contact IEEE to Subscribe

References

References is not available for this document.