New Trends in Advanced Analytics and AI for Patient Care

I'm a data scientist in the Advanced Analytics and AI group at Teva.
The three trends I'm interested in are:
Developing digital biomarkers
Discovering insights from data
And drug discovery
The first trend is digital biomarker development. A digital biomarker is a measurable characteristic collected from digital technology, which indicates a biological process, for example, sleep efficiency detected by an activity watch or a motor stage of a Parkinson's disease patient for example. I'm excited about digital biomarkers because it could provide early continuous and objective measurements that could help us see if a patient is responding to his treatment. So I think the potential for the future is that you could use digital biomarkers to personalize the treatment and better manage the disease of the patients.
The second trend is discovering insights from data, finding something interesting and possibly actionable in a data set. So for example, we can try to detect tremor intervals in a 2-weeks’ recording of an activity watch worn by a Parkinson's disease patient. This will allow us to evaluate his tremor severity. I'm excited about analyzing challenging data which is very noisy and contains irrelevant data and the possibility to pick the effective parts of it.
The third trend is drug discovery. Drug discovery is basically the process of identifying new medicines. So for example, a drug discovery process starts with target and identification. Target identification for a specific disease is finding an entity within the biological process of the disease, which can bind to a drug. There are numerous numbers of drug to target as well as disease to target interactions known in the literature. And we could utilize AI machine learning in order to identify new interactions from the known ones. So if we shorten the time to identify targets, we can shorten the time to the whole drug discovery process.