Article Text

Download PDFPDF
The whole is greater than the sum of its parts: combining classical statistical and machine intelligence methods in medicine
  1. Khader Shameer1,
  2. Kipp W Johnson2,
  3. Benjamin S Glicksberg2,
  4. Joel T Dudley2,
  5. Partho P Sengupta3
  1. 1Department of Information Services, Center for Research Informatics and Innovation, Northwell Health, New Hyde Park, New York, USA
  2. 2Department of Genetics and Genomic Sciences, Icahn Institute for Genomics and Multiscale Biology, Institute for Next Generation Healthcare, Mount Sinai Health System, Icahn School of Medicine at Mount Sinai, New York City, New York, USA
  3. 3Division of Cardiology, West Virginia Heart and Vascular Institute, Morgantown, West Virginia, USA
  1. Correspondence to Dr Partho P Sengupta, Heart and Vascular Institute, West Virginia University, Morgantown, WV 26506, USA; Partho.Sengupta{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

We kindly thank Drs Ramachandran and van den Heuvel for their interest in our article.1 2 We broadly defined machine learning (the field concerned with algorithms to find structure or patterns in data) as a set of techniques to enable artificial intelligence (the field concerned with programming computers to mimic human intelligence). Methods commonly taught in statistics classes like linear regression, discriminant analysis, principal components analysis, and so on are used to find structure or patterns in data and can be considered as algorithms for machine learning. It is thus difficult to define a fine line between where statistics ends and where machine learning begins—some of the methods have both flavours …

View Full Text


  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Patient consent Not required.

  • Provenance and peer review Not commissioned; internally peer reviewed.

Linked Articles