• 

  Our logo

• 

  Accuracy Percentage

• 

  Web App Front Page

• 

  Cancer Detector

AccurateDetect

AccurateDetect is a web-based software that uses machine learning to predict various types of cancer through a simple blood test. It is built using Python, Ruby-on-Rails, HTML, CSS, and Javascript, and utilizes Tensorflow and Keras for its deep learning capabilities. The project was evaluated using ten-fold cross-validation and optimized by adjusting hyperparameters and the output threshold, resulting in a specificity of above 99% and a median accuracy of 82.30%. The development team encountered challenges in web development and time constraints, but were able to learn and improve their skills in AI, ML, web development, Python, and research. The team is currently looking into deploying AccurateDetect in the real world and collecting more training data to increase the accuracy of the model.

Built With

• css3
• flask
• html5
• java
• keras
• python
• tensorslow
• web-app