Inspiration
Skin cancer is one of the most common yet preventable cancers. Early detection can save lives, but access to dermatologists isn't always easy. We wanted to build a tool that empowers users to identify potential skin issues quickly, get actionable insights, and take the next step toward care.
What it does
It analyzes skin lesion images to classify them as benign or malignant using a CNN model. There is presented a clear explanation using any relevant clinical data that the user may give, and it helps users locate the nearest hospital or dermatologist if a malignant result is found.
How we built it
DermaDetect is an AI-powered web app that analyzes skin lesion images to classify them as benign or malignant using a CNN model. It provides a clear explanation using relevant clinical data and helps users locate the nearest hospital or dermatologist if a malignant result is found.
Challenges we ran into
- Model accuracy with limited time and data
- Integrating clinical data explanations with model predictions in a user-friendly way
- Deploying and connecting the LLM backend with the web frontend under time constraints
Accomplishments that we're proud of
- Successfully built an end-to-end AI diagnosis and support platform in under 24 hours
- Created an explainable AI system that builds trust through transparency
- Integrated geolocation features for actionable next steps after diagnosis
What we learned
- Best practices for building fast, responsive UIs with React, Flask, and Git
- Importance of accessibility and trust in health-related applications
What's next for DermaDetect
- Improve the model with larger, more diverse datasets
- Partner with medical institutions to validate and deploy in real-world scenarios
- Enable user history tracking and integration
Log in or sign up for Devpost to join the conversation.