Poster (Scientific congresses and symposiums)
TOWARDS AN ACCURATE CANCER DIAGNOSIS MODELIZATION:COMPARISON OF RANDOM FOREST STRATEGIES
Debit, Ahmed; Poulet, Christophe; JOSSE, Claire et al.
201919th BeSHG meeting 15th March 2019 Liege, Belgium
 

Files


Full Text
BeSHG_2019.pdf
Author preprint (27.13 kB)
Download

All documents in ORBi are protected by a user license.

Send to



Details



Keywords :
Biomarker Signature; Random Forest; Cancer Diagnosis; TCGA RNA-seq data
Abstract :
[en] Machine learning approaches are heavily used to produce models that will one day support clinical decisions. To be reliably used as a medical decision, such diagnosis and prognosis tools have to harbor a high-level of precision. Random Forests have been already used in cancer diagnosis, prognosis, and screening. Numerous Random Forests methods have been derived from the original random forest algorithm from Breiman et al. in 2001. Nevertheless, the precision of their generated models remains unknown when facing biological data. The precision of such models can be therefore too variable to produce models with the same accuracy of classification, making them useless in daily clinics. Here, we perform an empirical comparison of Random Forest based strategies, looking for their precision in model accuracy and overall computational time. An assessment of 15 methods is carried out for the classification of paired normal -tumor patients, from 3 TCGA RNA-Seq datasets: BRCA (Breast Invasive Carcinoma), LUSC (Lung Squamous Cell Carcinoma), and THCA (Thyroid Carcinoma). Results demonstrate noteworthy differences in the precisions of the model accuracy and the overall time processing, between the strategies for one dataset, as well as between datasets for one strategy. Therefore, we highly recommend to test each random forest strategy prior to modelization. This will certainly improve the precision in model accuracy while revealing the method of choice for the candidate data.
Research center :
Human Genetics, BIO3 GIGA-R
Disciplines :
Computer science
Life sciences: Multidisciplinary, general & others
Author, co-author :
Debit, Ahmed ;  Université de Liège - ULiège > GIGA
Poulet, Christophe  ;  Université de Liège - ULiège > Département des sciences biomédicales et précliniques > Département des sciences biomédicales et précliniques
JOSSE, Claire  ;  Centre Hospitalier Universitaire de Liège - CHU > Département de médecine interne > Service d'oncologie médicale
Azencott, Chloe-Agathe;  Mines ParisTech, Institut Curie and INSERM > Centre for Computational Biology (CBIO)
Jerusalem, Guy  ;  Université de Liège - ULiège > Département des sciences cliniques > Oncologie
Van Steen, Kristel  ;  Université de Liège - ULiège > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Bioinformatique
BOURS, Vincent ;  Centre Hospitalier Universitaire de Liège - CHU > Unilab > Service de génétique
Language :
English
Title :
TOWARDS AN ACCURATE CANCER DIAGNOSIS MODELIZATION:COMPARISON OF RANDOM FOREST STRATEGIES
Publication date :
15 March 2019
Event name :
19th BeSHG meeting 15th March 2019 Liege, Belgium
Event organizer :
Belgian Society of Human Genetics BeSHG
Event place :
Liege, Belgium
Event date :
15-03-2019
Name of the research project :
WALInnov-NACATS 1610125
Funders :
Région wallonne [BE]
Available on ORBi :
since 26 September 2020

Statistics


Number of views
90 (5 by ULiège)
Number of downloads
11 (3 by ULiège)

Bibliography


Similar publications



Contact ORBi