Dataset Information

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

ABSTRACT: 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.

SUBMITTER: Papaiz F

PROVIDER: S-EPMC10949816 | biostudies-literature | 2024 Mar

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

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

Papaiz Fabiano F Dourado Mario Emílio Teixeira MET de Medeiros Valentim Ricardo Alexsandro RA Pinto Rafael R de Morais Antônio Higor Freire AHF Arrais Joel Perdiz JP

BMC medical informatics and decision making 20240319 1

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 sympto ...[more]

PMID: 38504285

Similar Datasets

Project description:IntroductionAmyotrophic lateral sclerosis (ALS) is a rare, devastating neurodegenerative disease that affects upper and lower motor neurons. To date, no effective treatment or reliable biomarker for ALS has been developed. In recent years, many factors have been proposed as possible biomarkers of ALS; however, no consensus has been reached. Therefore, a reliable biomarker is urgently needed. Eosinophils may play a crucial role in healthy humans and diseases, and serve as a biomarker for many chronic diseases.MethodsRoutine blood test results were collected from 66 healthy controls and 59 patients with ALS. The percentages and total numbers of each cell population were analyzed, and the correlation between these indicators and patient ALS functional rating scale-revised (ALSFRS-R) score or disease progression rate (ΔFS score) was analyzed.ResultsCompared to healthy controls, the number of blood leukocytes, neutrophils, monocytes, and basophils was significantly decreased in patients with ALS (p = 0.002, p = 0.001, p = 0.049, and p < 0.0001, respectively). There was an increase in the number of eosinophils (p < 0.0001), but no difference in the number of lymphocytes between patients with ALS and healthy controls was found (p = 0.563). Compared to healthy controls, the percentage of neutrophils was decreased and the percentage of lymphocytes and eosinophils was increased in patients with ALS (p = 0.01, p = 0.012, and p = 0.001, respectively). There was no difference between patients with ALS and healthy controls in the percentage of monocytes and basophils (p = 0.622 and p = 0.09, respectively). However, only the percentage and number of eosinophils had a correlation with the ΔFS score. Further multivariate analysis revealed a significant correlation between the disease duration, eosinophil count and percentage, and the disease progression rate (p < 0.0001, p = 0.048, and p = 0.023, respectively). The neutrophil-to-eosinophil ratio (NER), lymphocyte-to-eosinophil ratio (LER), and monocyte-to-eosinophil ratio (MER) were significantly lower in patients with ALS than in healthy controls. However, only the LER was significantly correlated with the ΔFS score.ConclusionThese observations implicate neutrophils, lymphocytes, and eosinophils as important factors, and increasing eosinophil counts were negatively correlated with the ΔFS score in patients with ALS.

Dataset Information

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

Publications

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

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets