Getting a job doing ML in Health

Part 1 (2019)
1

I would love to get a job doing ML in the health sector but sometimes it feels like it’s a difficult industry to crack.

I have a pretty heavy software engineering background, and I’m not a domain expert in medicine or health. I have been working with Fastai for the past year now, and I’ve completed a few health related Kaggle competitions and posting the related Jupyter notebooks as project pieces on my website.

I’d love to get advice on how best to approach getting a job in this field. In particular I’m trying to figure out:

  1. How should I keep building my folio?
  2. What kind of work I need to be doing?
  3. Who should I be speaking to?
  4. Or if I should be working on getting certified somehow?
  5. Should I be working with more health related Kaggle datasets and competitions?
  6. Or should is there another type of project I could be doing instead?
  7. Or is this just too difficult for someone without a medical background to do?

Any advice around this would be super helpful!

Thanks Fastai community :slight_smile:

2

I think you should first get into the industry . What I mean by that is :-

  1. Understand the format of data dealt there
  2. Pick up companies that work on AI+Med. Find what sectors are they targetting.[Look up angel.co,glassdoor / Google Search on Medical AI startups]
  3. At last, follow your favorite use cases. For eg:- You could only be interested in Kidney Tumor Segmentation.
  4. Go laterally. Forget Step 2 . Skip to Step 3 and find companies that match your favorite use case.

Step 1 is a must.

1 Like
3

Awesome thanks a lot @ucalyptus!

I’ve been trying to understand data formats like DICOM etc as well and I think that’s probably the next project I would need to work on, and in an area of interest. It can be difficult to find out exactly what I need to build on so thank you for sharing!