Overview / Usage

In the face of challenges, innovation can emerge that transforms negative experiences into powerful catalysts for positive change. The unfortunate incident of a person being falsely predicted and administered cancer drugs for years serves as a stark reminder of the critical responsibility that rests upon those venturing into the field of medical technology. While such an incident highlights the potential risks of inaccurate predictions, it also underscores the urgency to develop ethical and accurate solutions in skin disease prediction. Hence building a skin disease prediction application can be a valuable contribution to the field of healthcare and dermatology. Such an application could help individuals identify potential skin issues early on and encourage them to seek professional medical advice.

Here's an overview of how this application works and the additional information it could provides:

• Input of dermoscopic image or normal image.

• Prediction: The application uses a trained convolutional neural network (CNN) model to predict

• Output: Provides the user with a prediction of the skin disease or condition.

• Gives the disease overview

• Provides available medical treatments

• Provides home remedies.

Methodology / Approach

• Imported essential libraries

• Understand the data

• Train Test split

• Train the model using CNN to get better results and faster computation (Intel oneAPI Deep Neural Network (oneDNN))

• Model Fitting

• Attempt to gain maximum accuracy

• Save the model

Technologies Used

• Intel Devcloud

• onrDNN

• HTML, CSS, JS

• Django

Repository

https://github.com/Dishanth-Prabhakar/DermNet.git

Collaborators