Enhancing Disease Diagnoses Using Machine Learning
Keywords: Artificial intelligence License: MIT
Prototype
To test the speed, practicality, and efficiency of the model in tuberculosis diagnosis, I devised a prototype written in Python. The Python language's concise, expressive, and dynamic nature makes it well suited for prototyping tasks. Both models utilize PyTorch for training and validation.
Inspiration
In a recent project, I explored how accelerators like GPUs (Graphics Processing Unit) and TPUs (Tensor Processing Unit) would compare to the standard computer vision simulation, and discovered that the new accelerators displayed a boost in accuracy and generalization.
Conclusion and Further Research
As excited as I was to explore this new field and see results, due to the hackathons short timeframe, I was not able to train the model in the time provided. I had spent the entirety of Saturday building/designing the model and left the model to train all Saturday night to Sunday morning, but had found the model untrained. Provided more time to properly train on 8 epoches, the model has the potential to display significant results.
Log in or sign up for Devpost to join the conversation.