Data Science and Cancer Research

Data Science and Cancer Research

January 14, 2019 DATAcated Challenge 1

Data Science has a lot of interesting applications, from financial to medical ones.

Today I want to tell you about the medical ones, specifically what can be done in cancer research. I have a bachelor degree in Bioinformatics and I am now attending the last year of a Master in Medical Bioinformatics, so the studies I did offered me the possibily to connect medical bioinformatics with data science.

There is one field of research that I find particularly interesting and it is neuroimaging. There are a lot of different applications of neuroimaging, for example I worked with fMRI data for my bachelor thesis and with dMRI data for my master thesis ( I still have to finish). fMRI is Functional Magnetic Resonance Imaging while dMRI is Diffiusion Magnetic Resonance Imaging. With fMRI you can get an idea of the activity that is present in the brain while doing a certain task or resting; with dMRI you measure the diffusion of water to estimate how the fibers, axons are structured and connect different areas of the brain.
Anyway, this was just a short introduction to other interesting topics I want to talk you about today: cancer and data science.

Some days ago I was reading an article done by Novartis about a recent work they started. This work involved AI, Machine Learning, Cancer, Images and doctors. They developed an application (PathAI) that could identify cancerous cells form a tissue image in seconds. It could seem like a normal thing to you, but if you consider that to do the same thing a pathologist would need hours, what they achieved is quite awsome. Identifying patterns and cancerous cells is not easy, there is a lot of training to be done to ensure that the algorithm is capable of distinguish between a normal tissue and tumor. This research could lead to new discoveries, for example they could understand how spcific features correlate with the response to a certain drug.

Now imagine what we could achieve bringing data science and fields like imaging and cancer research together. We could help doctors make diagnosis, make more precise diagnosis, find patterns and information never seen before and a lot more.

Machine Learning has great potentials if used in imaging or neuroimaging. We could have a dataset of images, structural brain images acquired with MRI (Magnetic Resonance Imaging) and using machine learning we could develop an application that could help surgeons identify the cancer when operating, so doctors would be more sure of removing all cancerous cells.
These are just a few of the things that data science could do applied to the fields I mentioned, but it is a very great start!

If you are interested in reading the article I talked about, I leave here the link:

Data Science and Cancer Research: Carlotta Fabris.


One Response

  1. Mahsa says:

    Sorry i couldnt access the link could you send me the paper

Leave a Reply

Your email address will not be published. Required fields are marked *