Skip to content

Big Data in Health Care

Data reuse in health care is a vast topic, one problematic being the anonymization of the data themselves. This is such a complex problem that it’s usually the one in the spotlights, but it is useful sometimes to remind ourselves why working out the data is needed. Let us imagine briefly that we have solved the matter and reflect only on the opportunities and solutions Big Data technologies bring.

Electronic_stethoscope

Before answering the question, if you see data as a set of materials, Big data is like a toolbox. It basically encompasses any technology that allows you to process data and build something based on them. When we mean any, this is literally the case, from stream processing to infrastructure.

7263382550_002a49be33_b

Data is the new oil some say, and you are probably sitting on a pile of data, which means knowledge and information, which means value. Think of 3 of the most incredible problems you’d like to solve, or 3 wishes. Odds are that at least one of them can be solved, captured or tamed. And I am willing to buy you a drink if that isn’t the case.

Let’s not get carried away though.

To put it simply, you might want to consider working out your data if you want to do any of the following :

  • Process lots of data, frequently and in real-time

  • Process different sources of information, of different quality and volume and have those results interact

  • Predict the future

  • Find patterns that could not be discovered

  • ….

It goes without saying, this is merely what I would call “a good start”. Ideally, it should not be restricted to any of the former propositions (I am not talking about the drink), dare to think outside of any of the boxes any tech guy or actually anyone has tried to lock your thoughts in. Also I guarantee that there still are some people out there who love a good challenge and that would be thrilled to be able to work on whatever you have been wishing to have for so long.

Ok now, for the more practical of you, I have two sample situations that illustrate the propositions above :

Let’s focus one problem in hospitals : resources planning for emergency services. If you could wish for anything, it might be to be able to read the future in order to redirect patient as soon as they arrive, to make sure the adequate beds are ready, that the medication is available, that the nurses are present etc… What would it mean to improve your hourly predictions by 20% how much more effective could you then become? This is one problem that could find much better solutions by using the toolbox.

Let’s consider another problem for a pharmaceutical company : What the monitoring in real time of patients’ health or clinical trials subjects could represents? What if you could not only monitor them, but if you could also detect some patterns and predict evolution to react and avoid death or other critical conditions?

Health Care is different in the sense that most of the time, everything is documented and if it’s not, there are a lot of ways to find correlated data.

This article was voluntarily kept short, but if by the end of these few words you’re skeptical at least it’s a good start. Please do not hesitate to reach out to ask questions or to try and challenge any of the assertions above. Maybe you’ll be amazed by what you possess. It’s either that, or a free drink.

Releated Posts

IEEE Big Data 2023 – A Summary

Our CTO, Sabri Skhiri, recently travelled to Sorrento for IEEE Big Data 2023. In this article, Sabri explores for you the various keynotes and talks that took place during the
Read More

Robust ML Approach for Screening MET Drug Candidates in Combination with Immune Checkpoint Inhibitors

Present study highlights the significance of dataset size in ICI microbiota models and presents a methodology to enhance the performances of a multi-cohort-based ML approach.
Read More