Depression Detection System
Depression Detection System
Volume 8, Issue 3, March – 2023 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
Depression Detection System
1
V. Jyothi (Assistant Professor)
2
Khyati S Desai
3
T Nihal Reddy
4
V Lakshmi Sruthi
Department of CSE
Anurag Group of Institutions
Abstract:- The majority of people deal with depression facilitating prompt intervention and early warning.
on a daily basis, which is a prevalent and serious mental
illness. Depression has an impact on a person's physical, Existing System
psychological, and mental health in addition to their The existing mental health system focuses only
emotional state. In contrast to other illnesses, depression unimodal development of a person’s mental health, and few
cannot be diagnosed through laboratory testing, goes systems use only PHQ-9 tests, resulting in inadequate
unnoticed due to a lack of knowledge and awareness, treatment and misdiagnosis. A more comprehensive system is
and can deteriorate to the point of suicide. Physicians needed, using accurate tools to evaluate all aspects of a
are currently using self-reported questionnaires and in- person’s mental health, leading to personalized and effective
person interactions as part of their diagnostic process for treatment plans that can improvethe quality of life for people
identifying depression. A psychiatric evaluation of social living with depression.
interactions and human behaviour is required for the
diagnosis of depression. The patient's audio and video Proposed System
recordings show how people with depression behave The proposed application has two user categories: user
differently than average people do. The user and the and admin. Users can take the PHQ-9 test to evaluate their
admin are the two different user categories that can mental health or choose a more comprehensive assessment
interact with the application. The user has two choices: with a questionnaire, video, and audio detection. Based on
the PHQ-9 exam or an evaluation that consists of three the results, the system can suggest treatment options for the
components-a questionnaire, a video, and an audio user's level of depression. Admins have access to all
detection, each weighted at 33% and used to determine assessment reports to track user progress over time. This
the user’s level of depression. The findings are also used system aims to improve mental health care quality and
to suggest treatment alternatives. In order to combat the provide personalized support for those living with
condition sooner, computer vision and machine learning depression.
havebeen employed to diagnose depression.
I. INTRODUCTION
Millions of people all over the world suffer from
depression, which is a common mental health disorder.
Unfortunately, due to obstacles like stigma, limited access to
mental health professionals, and a lack of resources, many
people do not receive timely diagnoses and treatments.
Depression can therefore have a significant effect on a
person's quality of life and raise the risk of suicide. To solve
this problem, scientists and programmers have been
developing depression detection systems that use machine
learning and artificial intelligence to identify people who
may be at risk for depression. These systems can examine a
variety of data sources, including social media activity, speech
patterns, and physiological signals, to identify depressive
symptoms and give individuals, their families, and medical
professionals early warning. The goal of this project is to
create a depression detection system that can correctly Fig 1 Proposed System Architecture
identify people who might be depressed. To provide a more
thorough and precise evaluation of a person's mental health
status, the system will make use of a variety of data sources
and cutting-edge machine learning techniques. This system
could enhance mental health outcomes and lessen the effects
of depression on individuals, families, and society by
IJISRT23MAR1654 www.ijisrt.com 1436
Volume 8, Issue 3, March – 2023 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
II. EXPERMENTAL TOOLS options.
Visual Studio Code Tensorflow
Microsoft created the well-known open-source code TensorFlow is an open-source library for large-scale
editor known as Visual Studio Code (VS Code). For writing machine learning and numerical computation that makes it
and debugging code, developers and programmers use it easier for Google Brain TensorFlow to gather data, train
frequently. The popularity of VS Code among developers is models, deliver predictions, and improve future outcomes.
largely due to its many features, which include code Machine learning and deep learning models and algorithms
highlighting, autocomplete, and debugging tools. Numerous are bundled together by TensorFlow. Python is used as a
programming languages, such as Java, Python, JavaScript, practical front-end.
C++, and others are supported.
NLTK
Opencv NLTK(Natural Language Toolkit) is the favored
A sizable open-source library for image processing, Python API for natural language processing. To prepare text
machine learning, and computer vision is called OpenCV. data for further analysis, such as with ML models, it is a
Python, C++, and Java are just a few of the many very potent tool. It assists in turning text into numbers,
programming languages that OpenCV is compatible with. In which the model can use to its advantage.
order to identify objects, faces, and even human Moviepy
handwriting, it can analyze images and videos. The number It is a Python package for video editing that can be
of weapons in your arsenal increases when OpenCV is used with straightforward operations (like cuts,
combined with other libraries, such as NumPy, a highly concatenations, and title insertions), video processing, non-
optimized library for numerical operations, because any linear video compositing or sophisticated effects. The bulk
operation that can be done in NumPy can also be merged of widely used video formats, including GIF, can be read
withOpenCV. and written.
Django Textblob
The rapid development of trustworthy websites is made A Python (2 and 3) library called TextBlob is used to
possible by the high-level Python web platform known as process textual data. It offers a straightforward API for
Django. Django, which was created by seasoned getting started with typical natural language processing
programmers, handles a lot of the hassle associated with web (NLP) tasks like part-of-speech tagging, noun phrase
development, allowing you to concentrate on writing your extraction, sentiment analysis, classification, translation, and
app without having to invent the wheel. It is open source others.
and free, has a thriving community, excellent
documentation, and a variety of free and paid support Output
Fig 2 User Interface
It displays the user's home page, where they can register if it's their first time, log in, and also loginas an administrator.
IJISRT23MAR1654 www.ijisrt.com 1437
Volume 8, Issue 3, March – 2023 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
Fig 3 Registration Page
The age field on the registration page is the most crucial one because the questions will bedisplayed according to the
age group.
Fig 4 Selection Page
The user is given the choice on this page between the PHQ-9, a self-analysis tool, and an evaluation that comprises of a
questionnaire and an audio/video detection.
IJISRT23MAR1654 www.ijisrt.com 1438
Volume 8, Issue 3, March – 2023 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
Fig 5 PHQ-9 Test
All age groups take the same 9 questions on the test, and each question is worth a certain number of points. Results include
the overall score, the level of depression, and a suggestion.
Fig 6 Questionnaire the Above Questionnaire Changes According to Age Groups
IJISRT23MAR1654 www.ijisrt.com 1439
Volume 8, Issue 3, March – 2023 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
Fig 7 Audio/Video Detetcion Session will be Started and the Audio/Video will be Detected
Fig 8 Depression Level Detection
Based on the questionnaire and audio/video session, the severity of depression isdetermined.
IJISRT23MAR1654 www.ijisrt.com 1440
Volume 8, Issue 3, March – 2023 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
Fig 9 Recommendation
Suggestions are displayed based on the severity of the depression i.e games, exercises or nearbypsychiatrist.
III. CONCLUSION [5] Automatic depression level analysis using visual
expressions by Artificial Intelligence
In conclusion, depression is a complex and serious [6] Adolescents' speech during family interactions can be
mental health issue that affects a significant number of examined for signs of depression\
people worldwide. While it cannot be diagnosed through [7] System of multi-channel weighted speech
laboratory tests, innovative solutions are being developed to classification for teenagers' major depression
improve the accuracy and effectiveness of the diagnostic prediction
process. The use of audio and video recordings, along with [8] A comparison of read and spontaneous speech for the
self-reported questionnaires, can provide a more purpose of detecting depression
comprehensive assessment of a patient's mental state. The [9] A study of acoustic characteristics for depression
development of an application that uses computer vision and detection
machine learning to identify depression has the potential to [10] An effective deep model for audio-based depression
revolutionize the way we diagnose and treat this condition. classification is DE AudioNet
By detecting depression early on, patients can receive the [11] Analysis of eye movement to identify depression in
appropriate treatment and support, improving their chances teenagers
of recovery and reducing the risk of suicide. The findings of [12] Teenage depression detection using facial features
this research offer hope for those struggling with depression [13] A Comprehensive Review of Effective Methods for
and highlight the importance of continued efforts to improve Depression Detection on Social Media
mental health diagnosis and treatment. [14] Determining depression using eye flutter
characteristics
REFERENCES [15] Movement of related body parts for automated
depression analysis
[1] What reveals about depression level? Role of multi-
modal features at the level of interviewquestions
[2] Using partial least square regression, depression can
be identified based on dynamic vocal and facial
expression features
[3] A systematic review of an automated test for
psychiatric disorders using speech
[4] Deep learning for recognizing depression
