Ensemble-imbalance-based classification for amyotrophic lateral sclerosis prognostic prediction: identifying short-survival patients at diagnosis.

Autor: Papaiz F; Federal University of Rio Grande Do Norte, Natal, Brazil. fabianopapaiz@gmail.com.; University of Coimbra, Coimbra, Portugal. fabianopapaiz@gmail.com.; Federal Institute of Rio Grande Do Norte, Natal, Brazil. fabianopapaiz@gmail.com., Dourado MET Jr; Federal University of Rio Grande Do Norte, Natal, Brazil., de Medeiros Valentim RA; Federal University of Rio Grande Do Norte, Natal, Brazil., Pinto R; Federal University of Rio Grande Do Norte, Natal, Brazil.; Federal Institute of Rio Grande Do Norte, Natal, Brazil., de Morais AHF; Federal Institute of Rio Grande Do Norte, Natal, Brazil., Arrais JP; University of Coimbra, Coimbra, Portugal.
Jazyk: angličtina
Zdroj: BMC medical informatics and decision making [BMC Med Inform Decis Mak] 2024 Mar 19; Vol. 24 (1), pp. 80. Date of Electronic Publication: 2024 Mar 19.
DOI: 10.1186/s12911-024-02484-5
Abstrakt: Prognosticating Amyotrophic Lateral Sclerosis (ALS) presents a formidable challenge due to patients exhibiting different onset sites, progression rates, and survival times. In this study, we have developed and evaluated Machine Learning (ML) algorithms that integrate Ensemble and Imbalance Learning techniques to classify patients into Short and Non-Short survival groups based on data collected during diagnosis. We aimed to identify individuals at high risk of mortality within 24 months of symptom onset through analysis of patient data commonly encountered in daily clinical practice. Our Ensemble-Imbalance approach underwent evaluation employing six ML algorithms as base classifiers. Remarkably, our results outperformed those of individual algorithms, achieving a Balanced Accuracy of 88% and a Sensitivity of 96%. Additionally, we used the Shapley Additive Explanations framework to elucidate the decision-making process of the top-performing model, pinpointing the most important features and their correlations with the target prediction. Furthermore, we presented helpful tools to visualize and compare patient similarities, offering valuable insights. Confirming the obtained results, our approach could aid physicians in devising personalized treatment plans at the time of diagnosis or serve as an inclusion/exclusion criterion in clinical trials.
(© 2024. The Author(s).)
Databáze: MEDLINE
Externí odkaz:
Nepřihlášeným uživatelům se plný text nezobrazuje