MindPatch - Mental Health Support App

• 

  Our logo

• 

  our app interface

• 

  our app interface

Inspiration

Many people struggle to clearly express how they feel, especially when they are emotionally overwhelmed. Phrases like “not good” or “I’m okay I guess” are common, yet most mental-health tools fail to respond meaningfully to them. MindPatch was inspired by the idea that emotional support should not require perfect words it should meet people where they are.

What it does

MindPatch is a lightweight mental-health support app that analyzes short user text to understand emotional tone and suggest gentle, supportive “mental patches” such as grounding, breathing, or gratitude. It does not diagnose or label users, and it is designed to respond safely even when emotions are unclear or mixed.

How we built it

We built MindPatch using Python and scikit-learn. An NLP emotion classifier was trained using TF-IDF vectorization and Logistic Regression. To handle ambiguous or negated phrases, we combined machine-learning predictions with simple rule-based logic. The app interface was developed using Streamlit for clarity and ease of use.

Challenges we ran into

One major challenge was handling short or ambiguous emotional inputs. Emotion classification models often struggle with negations and mixed signals. Instead of forcing predictions, we redesigned the logic to handle uncertainty responsibly. We also worked within a tight hackathon timeframe while ensuring ethical design choices.

Accomplishments that we're proud of

We successfully built a working mental-health support prototype that is ethical, explainable, and user-friendly. The model achieved strong validation accuracy, and the app gracefully handles emotional ambiguity instead of failing or misleading users. Completing the full pipeline within the hackathon timeframe was a major achievement.

What we learned

We learned that responsible AI design is especially important in mental-health applications. Accuracy alone is not enough — how a system handles uncertainty matters just as much. We also gained hands-on experience combining machine learning with thoughtful product logic.

What's next for MindPatch - Mental Health Support App

Future improvements include emotion intensity scoring, personalized mental patches, multilingual support, and more advanced NLP models. With further development, MindPatch could become a scalable emotional support tool for everyday mental wellbeing.

Built With

• joblib
• numpy
• pandas
• python
• scikit-learn
• streamlit