Hey Coner, it really depends on what you're interested in. I think the most common now is image processing, and that's where I started. Wrote an article about image segmentation [towardsdatascience.com/biomedical-...], which is part of the process to detect glaucoma disease.
X-ray and MRI are also pretty popular. There is a survey call "A Survey on Deep Learning in Medical Image Analysis" if you are interested to find out about other image processing applications.
Sounds a great place to start to be honest. I've always had an interest in medical field although not being professionally involved in it, is there some projects I could possibly contribute towards while developing my skills?
Also, from the perspective of learning machine learning, what language would be best? I know JS has tensorflow but I hear python is great for this stuff. Any thoughts?
There are past competitions on Kaggle such as Diabetic Retinopathy Detection and APTOS 2019 Blindness Detection, probably many others. Where there are datasets, and codes to learn from others how they did it. That would likely develop machine learning knowledge. Open-source project to contribute, unlikely, not that I know of. Most medical data are not open source.
Python will certainly be the way to go for ML. For me, I use the PyTorch framework. With Python, you get to use the system's resources and GPU. TensorFlow.js is only mainly used for presentation, after training a model (with Python), port the model into a "lite" version, and present model capability in web browser (with TFJS). Although end-to-end training can be done in TFJS, it is very limited to data size and computation. See my end-to-end training article to predict time series with TFJS
Thanks for the overview. As you can expect my knowledge of machine learning at the minute is quite primitive. But, I'll definitely be looking into some of the technologies you mentioned and thr article you linked. Thanks once again and I wish you the best of luck with your work. 👍
For further actions, you may consider blocking this person and/or reporting abuse
We're a place where coders share, stay up-to-date and grow their careers.
Hey Coner, it really depends on what you're interested in. I think the most common now is image processing, and that's where I started. Wrote an article about image segmentation [towardsdatascience.com/biomedical-...], which is part of the process to detect glaucoma disease.
X-ray and MRI are also pretty popular. There is a survey call "A Survey on Deep Learning in Medical Image Analysis" if you are interested to find out about other image processing applications.
Sounds a great place to start to be honest. I've always had an interest in medical field although not being professionally involved in it, is there some projects I could possibly contribute towards while developing my skills?
Also, from the perspective of learning machine learning, what language would be best? I know JS has tensorflow but I hear python is great for this stuff. Any thoughts?
There are past competitions on Kaggle such as Diabetic Retinopathy Detection and
APTOS 2019 Blindness Detection, probably many others. Where there are datasets, and codes to learn from others how they did it. That would likely develop machine learning knowledge. Open-source project to contribute, unlikely, not that I know of. Most medical data are not open source.
Python will certainly be the way to go for ML. For me, I use the PyTorch framework. With Python, you get to use the system's resources and GPU. TensorFlow.js is only mainly used for presentation, after training a model (with Python), port the model into a "lite" version, and present model capability in web browser (with TFJS). Although end-to-end training can be done in TFJS, it is very limited to data size and computation. See my end-to-end training article to predict time series with TFJS
Thanks for the overview. As you can expect my knowledge of machine learning at the minute is quite primitive. But, I'll definitely be looking into some of the technologies you mentioned and thr article you linked. Thanks once again and I wish you the best of luck with your work. 👍