A binary hidden Markov model on spatial network for amyotrophic lateral sclerosis disease spreading pattern analysis.

Area of Science:

• Neurology
• Biostatistics
• Computational Biology

Background:

• Amyotrophic Lateral Sclerosis (ALS) is a progressive neurological disorder characterized by muscle weakness.
• Understanding the spatiotemporal patterns of muscle weakness is crucial for ALS research.
• Existing methods may not fully capture the complex spread of muscle impairment in ALS.

Purpose of the Study:

• To develop and validate a novel statistical model for analyzing the spread of muscle weakness in ALS.
• To estimate key epidemiological parameters such as ALS incidence rate and disease state transition probabilities.
• To incorporate both historical patient data and spatial muscle network information into the analysis.

Main Methods:

• A hidden Markov model (HMM) framework was proposed, assuming observed muscle status depends on two latent disease states.
• The model utilizes logistic autoregression to capture the spatial network dynamics of muscle susceptibility, treating it as a Markov process.
• Parameter estimation involved an iterative algorithm for sparse-penalized likelihood maximization with bias correction, and the Viterbi algorithm for hidden state labeling.

Main Results:

• The proposed HMM approach successfully estimated ALS incidence rates and disease transition probabilities from spatiotemporal muscle strength data.
• The model demonstrated flexibility in integrating historical muscle conditions and spatial relationships.
• The Viterbi algorithm effectively identified hidden disease states, providing insights into ALS progression patterns.

Conclusions:

• The developed hidden Markov model offers a robust framework for analyzing the spatiotemporal progression of muscle weakness in ALS.
• This approach enhances the understanding of ALS disease dynamics and can inform future clinical studies.
• The model's ability to incorporate spatial and temporal factors provides a more comprehensive view of ALS progression.