{"title":"Different Approaches in Depression Analysis : A Review","authors":"Swathy Krishna, Anju. J","doi":"10.1109/ComPE49325.2020.9200001","DOIUrl":null,"url":null,"abstract":"Clinical depression has been a common but a serious mood disorder nowadays affecting people of any age group. Since depression affects the mental state, the patient will find it difficult to communicate his/her condition to the doctor. Commonly used diagnostic measures are interview style assessment or questionnaires about the symptoms, laboratory tests to check whether the depression symptoms are related with other serious illness. With the emergence of machine learning and convolutional neural networks, many techniques have been developed for supporting the diagnosis of depression in the past few years. Since depression is a multifactor disorder, the diagnosis of depression should follow a multimodal approach for its effective assessment. This paper presents a review of various unimodal and multimodal approaches that have been developed with the aim of analyzing the depression using emotion recognition. The unimodal approach considers either of the attributes among facial expressions, speech, etc. for depression detection while multimodal approaches are based on the combination of one or more attributes. This paper also reviews several depression detection systems that use facial feature extraction methods that use eigenvalue algorithm, fisher vector algorithm, etc. and speech features such as spectral, acoustic feature, etc. The survey covers the existing emotion detection research efforts that use audio and visual data for depression detection. The survey shows that the depression detection using multimodal approach and deep learning techniques achieve greater performance over unimodal approaches in the depression analysis.","PeriodicalId":6804,"journal":{"name":"2020 International Conference on Computational Performance Evaluation (ComPE)","volume":"76 1","pages":"407-414"},"PeriodicalIF":0.0000,"publicationDate":"2020-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"4","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2020 International Conference on Computational Performance Evaluation (ComPE)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ComPE49325.2020.9200001","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 4
Abstract
Clinical depression has been a common but a serious mood disorder nowadays affecting people of any age group. Since depression affects the mental state, the patient will find it difficult to communicate his/her condition to the doctor. Commonly used diagnostic measures are interview style assessment or questionnaires about the symptoms, laboratory tests to check whether the depression symptoms are related with other serious illness. With the emergence of machine learning and convolutional neural networks, many techniques have been developed for supporting the diagnosis of depression in the past few years. Since depression is a multifactor disorder, the diagnosis of depression should follow a multimodal approach for its effective assessment. This paper presents a review of various unimodal and multimodal approaches that have been developed with the aim of analyzing the depression using emotion recognition. The unimodal approach considers either of the attributes among facial expressions, speech, etc. for depression detection while multimodal approaches are based on the combination of one or more attributes. This paper also reviews several depression detection systems that use facial feature extraction methods that use eigenvalue algorithm, fisher vector algorithm, etc. and speech features such as spectral, acoustic feature, etc. The survey covers the existing emotion detection research efforts that use audio and visual data for depression detection. The survey shows that the depression detection using multimodal approach and deep learning techniques achieve greater performance over unimodal approaches in the depression analysis.