We keep on hearing about how artificial intelligence and machine learning is going to revolutionise Medicine.
But what’s hype, and what’s realistic? And how can you get involved?
The first step is to understand the technology - where it’s well-suited to healthcare (and where it isn’t).
Thankfully, there are several online courses that can get us up to speed:
AI For Medicine, by DeepLearning.AI
This course is taught by Pranav Rajpurkar, a PhD student at Stanford under Andrew Ng.
It covers AI for medicine across three domains:
- AI For Medical Diagnosis (where you’ll code a neural network to diagnose lung and brain disorders)
- AI For Medical Prognosis (where you’ll code risk models and prognosis predictors)
- AI for Medical Treatment (where you’ll write code to predict the effects of treatment)
Pranav and Andrew are prominent researchers in the field, so you’re being taught by the best.
The course is pretty hands-on and technical. Therefore, it will be tough for people who aren’t comfortable with coding.
It’s available here on Coursera. It costs $49 per month to enrol, so total cost will depend on how long you take to complete it (they advise it will take around 3 months at 7 hours/week).
Artificial Intelligence in Healthcare, by Stanford School of Medicine
The program is made up of five components:
- Introduction to Healthcare
- Introduction to Clinical Data
- Fundamentals of Machine Learning for Healthcare
- Evaluations of AI Applications in Healthcare
- AI in Healthcare Capstone
The course is taught by a mix of faculty at Stanford University.
It’s aimed at an introductory level, and the focus is more on theory than practical (coding).
I’d recommend this one ahead of the DeepLearning.AI course if you have limited coding experience.
Links to each module on Coursera are available here. Each module is $79.
Machine Learning for Healthcare (FREE), by Dr Chris Lovejoy
I made this course as a free alternative / precursor to the more serious courses.
It’s a nine-part video series that could probably be completed in an afternoon.
It provides a less-technical and more healthcare-tailored introduction to machine learning, and the nuances of applying it to healthcare.
Specifically, it covers:
- What is machine learning and how do we train an algorithm?
- What have ‘neural networks’ and ‘deep learning’ got to do with medicine?
- How do we assess the performance and clinical impact of machine learning models?
- How to read a medical machine learning paper
- How can you combine machine learning with a medical career?
I filmed this in my living room, so it’s not quite as professional as the courses above 😅.
But I’m keen to lower the barrier-to-entry for learning about ML, so I hope this is a good introduction for those on-the-fence about paying for a multi-week course.
If you enjoy this course, definitely consider taking one of the above courses as a follow-on.
This is a hot topic, so I’m sure other courses will pop up over time.
If there’s any you’re aware of, please let me know in the comments below, and I’ll add them here!