Previously we talked about logical structuring medical application for mobile or web. Here Are Some GitHub Projects Around Machine Learning in Medical Diagnosis. Few current applications of AI in medical diagnostics are already in use. Machine Learning and AI is relatively slower growing compared to usage in core technical matters because of mess with data, lack of free data and somehow modern medicine has not much logical progress around standardized way of debugging. It is probably practical for many developers to know about few GitHub projects, not always we can easily search and find details faster.
Common part usually installing CUDA, Python libraries for data sciences, Jupyter Notebook, Anaconda so on.
Machine Learning in Medical Diagnosis : GitHub Projects
First one is of OpenCV, it is actually illustrative project for a book. It is great for the beginners who are somewhat used with Jupyter Notebook :
---
1 2 | https://github.com/mbeyeler/opencv-machine-learning https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/05.00-Using-Decision-Trees-to-Make-a-Medical-Diagnosis.ipynb |
Second one is basic but probably helpful as we lack a general purpose algorithm as this moment :
1 | https://github.com/mp2893/doctorai |
Here is a kind of chatbot :
1 | https://github.com/ebezzam/DoctorBot |
This is an example of algorithm with Melanoma :
1 | https://github.com/udacity/dermatologist-ai |
This is example of pulmonary nodule :
1 | https://github.com/JenifferWuUCLA/pulmonary_nodules_AI_diagnosis |
Example of prediction of chronic diseases using a patient’s previous history :
1 | https://github.com/Naresh1318/DiagnosisPredictor |
Here is a list of other resources of machine learning around healthcare :
1 | https://github.com/isaacmg/healthcare_ml |
Here, you’ll get some resources which are not directly any project but useful as data resource :
1 | https://github.com/beamandrew/medical-data |
F/OSS Imaging and Machine Learning in Medical Diagnosis
In that article I talked about logical structuring medical application for mobile or web and shown that diagnosis commonly not so difficult. Developing more such those flow charts for differential diagnosis will make the work easier. Machine Learning & AI more practical for image matching, which does have higher importance when combined with open source USG like projects. After data collection, things can be piped through some deep learning tools, like described in this project :
1 | https://spandan-madan.github.io/DeepLearningProject/ |