How to code it all?

Course Q&A AI For Medicine AI for Medical Diagnosis
1

Hello

I have completed the AI in Medical Diagnosis course and have been able to successfully complete all the assignments. However this is just filling in some of the blanks in a much larger set of code. I am wondering how one learns to code the whole thing like the Week 3 assignment on Tumour Segmentation or this is done by large team of programmers?

Thanks

Eric

2

You should be improving with time and experience Eric.

3

Hi @Eric5,

As Gent quite rightly said, you will widen your perspective with time and experience.

There are always two aspects to programming. One is the pure programming bits such as language (such as Python) and tech domain (such as AI/Data Science/Machine Learning). The other is the business bits (here Clinical/Medicine).

The focus of these labs is more on the latter aspects i.e. to teach how technology is applied to medicine. That is the focus of the subset of lines you coded.

And yes, you are correct. It is probably true that more programmers working as a team would have crafted the larger framework of tutorial code. But that is software engineering in general, and if you are keen, you could pick that up as well. And apply it not just to Medicine but to any business domain.

Cheers.

4

Thanks so much, that makes sense

5

The other thing to point out is that no-one builds everything from scratch in python these days. If you are solving real world problems, you should use one of the deep learning platforms like TensorFlow or PyTorch or one of the others. The State of the Art is so deep these days that no-one has the time or expertise to build all the required algorithms themselves from scratch. I have not taken AI4M, so I don’t know what they cover there other than Pandas, but there are plenty of other courses here that will introduce you to TensorFlow and/or PyTorch. E.g. the DLS specialization would be a good one to take to learn about all the different types of networks and how to apply them. That uses TF extensively in the later courses (C4 and C5). And there are several specializations specifically about TF. The GANs specialization, which is interesting in its own right, also has the side benefit that it will introduce you to PyTorch.