Deep learning methods to predict amyotrophic lateral sclerosis disease progression.

Area of Science:

• Neuroscience
• Computational Biology
• Medical Informatics

Background:

• Amyotrophic Lateral Sclerosis (ALS) is a complex neurodegenerative disease impacting motor neurons, necessitating accurate disease progression prediction for timely treatment.
• The Pooled Resource Open-Access ALS Clinical Trials (PRO-ACT) repository provides extensive data for ALS research.
• Predictive models are crucial for understanding ALS progression due to the disease's low life expectancy.

Purpose of the Study:

• To investigate the efficacy of deep learning models in predicting ALS disease progression using the PRO-ACT cohort.
• To explore novel deep learning architectures for forecasting the slope of the ALSFRS score.
• To establish deep learning as a viable alternative for ALS progression prediction.

Main Methods:

• Development of three distinct deep learning models with varied architectures.
• Application of these models to predict ALS progression using the PRO-ACT dataset.
• Comparison of deep learning model performance against existing state-of-the-art methods.

Main Results:

• Deep learning models achieved performance comparable or superior to existing state-of-the-art approaches.
• The developed models demonstrated effectiveness in predicting ALS progression via ALSFRS slope.
• The study validates deep learning as a powerful tool for ALS research.

Conclusions:

• Deep learning models present a robust and effective alternative for predicting ALS disease progression.
• The findings highlight the potential of deep learning in advancing ALS patient care and treatment strategies.
• Further exploration of deep learning in the PRO-ACT cohort can yield significant insights into ALS heterogeneity.