English
 
User Manual Privacy Policy Disclaimer Contact us
  Advanced SearchBrowse

Item

ITEM ACTIONSEXPORT
  Inference and prediction diverge in biomedicine

Bzdok, D., Engemann, D. A., & Thirion, B. (2020). Inference and prediction diverge in biomedicine. Patterns, 100119. doi:10.1016/j.patter.2020.100119.

Item is

Basic

show hide
Item Permalink: http://hdl.handle.net/21.11116/0000-0007-450E-D Version Permalink: http://hdl.handle.net/21.11116/0000-0007-450F-C
Genre: Journal Article

Files

show Files
hide Files
:
Bzdok_2020.pdf (Publisher version), 5MB
Name:
Bzdok_2020.pdf
Description:
-
Visibility:
Public
MIME-Type / Checksum:
application/pdf / [MD5]
Technical Metadata:
Copyright Date:
-
Copyright Info:
-
License:
-

Locators

show

Creators

show
hide
 Creators:
Bzdok, Danilo1, 2, Author
Engemann, Denis A.3, 4, Author              
Thirion, Bertrand3, Author
Affiliations:
1Mila – Quebec Artificial Intelligence Institute, Montréal, QC, Canada, ou_persistent22              
2McConnell Brain Imaging Centre, Montreal Neurological Institute and Hospital, McGill University, Montréal, QC, Canada, ou_persistent22              
3Institut national de recherche en informatique et en automatique (INRIA), Gif-sur-Yvette, France, ou_persistent22              
4Department Neurology, MPI for Human Cognitive and Brain Sciences, Max Planck Society, ou_634549              

Content

show
hide
Free keywords: Explainable AI; Scientific discovery; Data science; Variable importance; Reproducibility
 Abstract: In the 20th century, many advances in biological knowledge and evidence-based medicine were supported by p values and accompanying methods. In the early 21st century, ambitions toward precision medicine place a premium on detailed predictions for single individuals. The shift causes tension between traditional regression methods used to infer statistically significant group differences and burgeoning predictive analysis tools suited to forecast an individual's future. Our comparison applies linear models for identifying significant contributing variables and for finding the most predictive variable sets. In systematic data simulations and common medical datasets, we explored how variables identified as significantly relevant and variables identified as predictively relevant can agree or diverge. Across analysis scenarios, even small predictive performances typically coincided with finding underlying significant statistical relationships, but not vice versa. More complete understanding of different ways to define “important” associations is a prerequisite for reproducible research and advances toward personalizing medical care.

Details

show
hide
Language(s): eng - English
 Dates: 2020-08-172020-06-162020-09-142020-10-08
 Publication Status: Published online
 Pages: -
 Publishing info: -
 Table of Contents: -
 Rev. Type: -
 Identifiers: DOI: 10.1016/j.patter.2020.100119
 Degree: -

Event

show

Legal Case

show

Project information

show hide
Project name : -
Grant ID : NIH R01 AG068563A
Funding program : -
Funding organization : National Institutes of Health
Project name : -
Grant ID : CIHR 438531
Funding program : -
Funding organization : Canadian Institutes of Health Research
Project name : -
Grant ID : BZ2/2-1, BZ2/3-1, and BZ2/4-1
Funding program : -
Funding organization : Deutsche Forschungsgemeinschaft
Project name : -
Grant ID : 126/16
Funding program : -
Funding organization : Faculty of Medicine, RWTH Aachen
Project name : -
Grant ID : OPSF449
Funding program : -
Funding organization : Exploratory Research Space, RWTH Aachen
Project name : Seventh Framework Programme
Grant ID : 604102
Funding program : (FP7/2007-2013)
Funding organization : European Union

Source 1

show
hide
Title: Patterns
Source Genre: Journal
 Creator(s):
Affiliations:
Publ. Info: Amsterdam : Elsevier
Pages: - Volume / Issue: - Sequence Number: 100119 Start / End Page: - Identifier: ISSN: 2666-3899
CoNE: https://pure.mpg.de/cone/journals/resource/2666-3899