Subscribe to DSC Newsletter

Data mining for clinical research - why so little use of machine learning?

As a bioinformatician, I am gradually drawn into analyzing clinical data. I noticed that the field is dominated by a small number of statistical tools. Any thoughts as of why?

Eitan

Tags: Clinical, Data, Machine, Medicine, Statistics, learning, vs.

Views: 2508

Reply to This

Replies to This Discussion

I had the same reflection as you... and started DNAlytics (http://www.dnalytics.com), a Belgian SME focused on Machine Learning (mainly feature selection and classification/regression tasks) applied to clinical research, personalized medicine and genomics. Our main activity is service, but we just released this week our first product as well: http://www.rheumakit, a new diagnostic solution for patient with arthritis, combining rheumatology, molecular biology, machine learning and cloud computing.

It may be difficult to provide a comfortable answer to the question. There are tools; however their use is to be with caution as clinical trials are not coming under usual statistical framework. If there is any specific question, possibly some useful answer could be given.

Actually, It's pretty obvious why data mining is NOT part of *clinical trials*. In clinical trials you are NOT trying to discover something new, you are trying to test a hypothesis (that treatment A is better than B, or something like this). 

My question was about clinical research: why are so few new and meaningful hypothesis generated through machine learning? Why does t-test, multiple logistic regression and generalized linear models so dominant in clincal research?

ER

Kalyanaraman K said:

It may be difficult to provide a comfortable answer to the question. There are tools; however their use is to be with caution as clinical trials are not coming under usual statistical framework. If there is any specific question, possibly some useful answer could be given.

I do not know if I could give any reason. However, I remember a number of reports using very complicated techniques to answer different types of problems. Please see the following reference..

Khattree. R and Naik. D.N (Ed.) “Computational Methods in Biomedical Research”, Chapman & Hall/CRC Biostatistics Series, 2008.



Eitan Rubin said:

Actually, It's pretty obvious why data mining is NOT part of *clinical trials*. In clinical trials you are NOT trying to discover something new, you are trying to test a hypothesis (that treatment A is better than B, or something like this). 

My question was about clinical research: why are so few new and meaningful hypothesis generated through machine learning? Why does t-test, multiple logistic regression and generalized linear models so dominant in clincal research?

ER

Kalyanaraman K said:

It may be difficult to provide a comfortable answer to the question. There are tools; however their use is to be with caution as clinical trials are not coming under usual statistical framework. If there is any specific question, possibly some useful answer could be given.

RSS

Videos

  • Add Videos
  • View All

© 2020   Data Science Central ®   Powered by

Badges  |  Report an Issue  |  Privacy Policy  |  Terms of Service