Dive deeper into the aspects covered in this lesson with these readings that expand on the topics discussed. Each reading includes reflective questions to help you focus on key learning points and strengthen your understanding.

Reading: Deep Learning in Medicine.
By Heilbroner, Samuel P.; Miotto, Riccardo.

This reading offers an insightful discussion on the potential future impact of deep learning in healthcare. It introduces key concepts we’ll explore in upcoming lectures, particularly how deep learning can enhance clinical decision-making by analyzing large, complex datasets like electronic health records (EHRs), medical images, -omics data, and wearable devices. The challenges of these datasets—such as high dimensionality, sparsity, and bias—make traditional methods challenging, highlighting the growing role of machine learning (ML) and deep learning as powerful solutions.

Reading time: ~45 min

   Download the reading

Reflective questions

  • How can deep learning overcome the hurdles of dealing with different kinds of data at once?
  • What are the problems with annotating data with labels for supervised learning and how deep learning can help?
  • What is self-supervised learning?

Source: Heilbroner SP, Miotto R. Deep Learning in Medicine. Clin J Am Soc Nephrol. 2023 Mar 1;18(3):397-399. doi: 10.2215/CJN.0000000000000080. Epub 2023 Jan 20. PMID: 36735512; PMCID: PMC10103223.  Link

Reading: Python AI: How to Build a Neural Network & Make Predictions.
By Déborah Mesquita.

Once again, for those of you getting excited about the idea of getting your hands dirty, we recommend checking out this very basic Deep Learning practical course in Python. The tutorial covers a bit of everything: from understanding what artificial intelligence is, to how both machine learning and deep learning contribute to AI, and how a neural network works internally. For the bravest among you, it even walks you through building a neural network from scratch using Python!

Source: Mesquita. Python AI: How to Build a Neural Network & Make Predictions. Website (Accessed 28 Jan 2025)  Link