Related Experiment Video
Updated: Jan 13, 2026
06:37
Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
Published on: December 15, 2023
5.3K
Computer intelligence based model for mental health detection among Indian farming communities
Jyoti Agarwal1, Sachin Sharma2, Parul Madan1
1Department of Computer Science and Engineering, Graphic Era Deemed to be University, Dehradun, 248002, India.
Scientific Reports
|October 30, 2025
Summary
This study developed a voice-based convolutional neural network (CNN) model to assess mental health in Indian farmers. The scalable system achieved 99.67% accuracy, offering a viable solution for rural mental healthcare.
Area of Science:
- Public Health
- Artificial Intelligence
- Rural Healthcare
Background:
- Mental health issues among Indian farmers are a significant public health concern, exacerbated by limited access to mental health professionals in rural areas.
- Traditional stress assessment methods are subjective, labor-intensive, and difficult to scale for widespread rural deployment.
- Factors like crop failure, price volatility, debt, and inadequate social support contribute to farmer well-being deterioration.
Purpose of the Study:
- To develop a scalable, voice-based diagnostic model using convolutional neural networks (CNNs) for mental health assessment in farmers.
- To evaluate the usability and effectiveness of this AI-driven system within rural Indian contexts.
- To address the critical need for accessible and objective mental health screening tools in underserved agricultural communities.
Main Methods:
- Collected audio responses from 350 Indian farmers in local languages to a structured questionnaire on stress, coping, and social support.
- Utilized a convolutional neural network (CNN) architecture to analyze audio spectrograms, enabling automated feature learning.
- Assessed the model's predictive accuracy for mental health status and evaluated six key usability factors.
Main Results:
- The CNN model demonstrated exceptionally high predictive accuracy for mental health classification, achieving 99.67%.
- The system exhibited strong performance across all six evaluated usability factors: learnability, efficiency, configurability, satisfaction, understandability, and effectiveness.
- These findings confirm the model's robustness and suitability for real-world application in rural settings.
Conclusions:
- A voice-based CNN model offers a highly accurate and scalable solution for mental health assessment among Indian farmers.
- The developed system shows significant promise for integration into rural healthcare outreach programs, improving access to mental health support.
- This AI-driven approach represents a feasible and effective method to overcome barriers in diagnosing and addressing mental health challenges in agricultural populations.
