Skin Detect

The Login Page
The Home Page
The About Us Page
The Upload Image Page
The Upload Result

Inspiration

Our inspiration stemmed from the need to bridge the gap between technology and healthcare, providing a tool that could make a tangible difference in early diagnosis and treatment.

What it does

Skin Detect allows users to upload images of skin issues, detect potential diseases, determine if a tumor is benign or malignant, and identify specific types of cancer. It also includes a chatbot to answer user queries and provide real-time support.

How we built it

The website was built using a combination of technologies:

CNN Neural Network for disease detection and cancer classification.
TensorFlow & Keras for model development.
FastAPI to serve the models' backend.
Django with MongoDB for managing user information.
Firebase for additional services.
HTML & Tailwind CSS for the frontend design.
Vercel for hosting the website.
FAISS vector store and LLaMA-2 dense embedding for the chatbot.
Figma for designing the user interface.

Challenges we ran into

Integrating different components of the project seamlessly.
Ensuring the accuracy of the detection models.
Addressing the ethical and legal implications of deploying a healthcare-related tool.
Hosting the main app

Accomplishments that we're proud of

Successfully developing a functional prototype that integrates machine learning with a user-friendly interface.
Creating a chatbot capable of providing real-time support and enhancing the user experience.
Laying the groundwork for a tool that can significantly impact early disease detection.

What we learned

The importance of interdisciplinary collaboration between tech and healthcare professionals.
The challenges of deploying machine learning models in real-world applications.
The value of continuous feedback and iteration in improving the accuracy and usability of healthcare tools.

What's next for Skin Detect

Educational Validation: We plan to validate the tool further in educational settings with medical students.
Public Deployment: Upon validation and compliance with healthcare regulations, we aim to deploy the website in the public sector, making it widely accessible.
Feature Expansion: We are considering additional features such as a mobile app, more advanced diagnostic capabilities, and integration with healthcare providers.