impact factor, citescore
logo
In press
In Press
Recent issue
Recent issue
Recent Supplement
Recent supplement
archives
Archives
articles free
Free to view
Search
Search
 

Reviews

 

New criteria and new methodological tools for devising criteria sets of inflammatory rheumatic diseases

1, 2, 3, 4

 

  1. UMR1227, Lymphocytes B et Autoimmunité, Université de Brest, Inserm, CHU Brest, LabEx IGO, Brest; and LATIM, Laboratoire de Traitement de l’Information Médicale, UMR 1101, IBRBS, Université de Brest, Inserm, CHU, Brest, France.
  2. LATIM, Laboratoire de Traitement de l’Information Médicale, UMR 1101, IBRBS, Université de Brest, Inserm, CHU, Brest, France.
  3. UMR1227, Lymphocytes B et Autoimmunité, Université de Brest, Inserm, CHU Brest, LabEx IGO, Brest, France.
  4. UMR1227, Lymphocytes B et Autoimmunité, Université de Brest, Inserm, CHU Brest, LabEx IGO, Brest; and Rheumatology Unit, Centre National de Référence des Maladies Auto-Immunes Rares (CERAINO), CHU, Brest, France. alain.saraux@chu-brest.fr

CER12828
2020 Vol.38, N°4
PI 0776, PF 0782
Reviews

Free to view
(click on article PDF icon to read the article)

PMID: 32105592 [PubMed]

Received: 01/10/2019
Accepted : 09/12/2019
In Press: 14/02/2020
Published: 28/07/2020

Abstract

Rheumatologists use classification criteria to separate patients with inflammatory rheumatic diseases (IRD). They change over time, and the concepts of the diseases also change. The paradigm is currently moving as the goal of classification in the future will be more to select which patients may be relevant for a specific treatment rather than to describe their characteristics. Therefore, the challenge will be to reclassify multifactorial diseases on the basis of their biological mechanisms rather than their clinical phenotype. Currently, various projects are trying to reclassify diseases using bioinformatics approaches and in the near future the use of advanced machine learning algorithms with large omics datasets could lead to new classification models not only based on a clinical phenotype but also on complex biological profile and common sensitivity to targeted treatment. These models would highlight common biological pathways between patients classified in the same cluster and provide a deep understanding of the mechanisms involved in the patient’s clinical phenotype. Such approaches would ultimately lead to classification models that rely more on biological causes than on symptoms. This overview on current classification of subgroups of IRD summarises the classification criteria that we use routinely, and how we will classify IRD in the future using bioinformatics and artificial intelligence techniques.

Rheumatology Article