The future of Deep Learning in Healthcare

The future of Deep Learning in Healthcare

Which field in digital healthcare will be revolutionised Deep Learning first? Histo-Pathology? Radiology? Lung-, Breast or Skin-Cancer diagnostics? Mobile and Monitoring Apps? Or prediction and personalised medicine on the basis of Biobank-data?

Our Feedback for some time has been, that Lung-, Breast and Skin-cancer-diagnosis has the highest potential for applications of AI, as the diagnosis in these fields makes up a lot of time and a huge market share. Also, it is technically doable already at this point, to develop Software for diagnosis on the basis of AI, that is able to give a diagnostic second opinion with high accuracy.

But it always depends on your business-model: There are obviously loads of other potential fields in healthcare, where AI can lead to tremendous improvements in terms of personalised medicine.

As Pearse Keane on the other hand suggested at the "Deep Learning in Healthcare Summit", Ophthalmology will be the first field to be revolutionized in healthcare, because of one institution and its huge resources of data: Moofields Eye Hospital. Moorfield treats 400 patients per day and has 300.000 hospital visits per year, which has no comparison in the world. In addition they are working with Google Deep Mind.

Using OCT – which is supposed to produce a data base, filled with data, comparable to histo-pathology slides – the hospital can produce 3000 scans per week and is thus buried beneath data. This brings to mind, what “The Economist” calls the “data deluge” and Neil Lawrence the “data delusion”.

The goals of Pearse Keane sounded similar to our vision in diagnostics:

1. General algorithm for diagnosis of retinal disease
2. New clinical and scientific insights
3. Reinventing Eye Examination

Do you think, Ophthalmology is the future of Deep Learning in Healthcare? Let us know in the comments or contact us directly!

Comments are closed.