profile image of Cancer


A web-app that detects 8 different types of cancer

Detects with an accuracy of over 96%!

Created using Tensorflow and Keras

AccurateDetect is prediction software focusing on the development of a web based multi-platform solution for augmenting prognostic strategies to diagnose different types of cancer, from a variety of inexpensive blood tests, all in an automated fashion.

It uses deep learning to detect 8 different types of cancers with great accuracy. AccurateDetect consists of neural network with 2 hidden layers and considers various parameters obtained from a blood test.

AccurateDetect was evaluated using ten-fold cross validation, and finally optimized by adjusting hyper-parameters and the output threshold. While keeping the specificity above 99%, the median accuracy was increased from 70.80% in the Detect A model to 82.30% using AccurateDetect. The accuracy of results of the model as determined by the testing set, was competent enough or outperformed other existing and similar projects like Detect-A.

Upload a .csv file with 40 columns of protein data to check diagnosis 👇