New machine learning methods better describe the progression of neurodegenerative diseases

Neurodegenerative diseases – such as amyotrophic lateral sclerosis (amyotrophic lateral sclerosis), Alzheimer's disease, and Parkinson's disease – are complex chronic diseases that can present a variety of symptoms, worsen at different rates, and have many underlying genetic and environmental causes, some of
which are unknown.

However, assessing the evolution of disease is far from simple and
clear.

Now, a new machine learning method developed by researchers at MIT, the IBM Research Center, and other institutions aims to better describe the development pattern of ALS disease and inform clinical trial design
.

"There are groups of individuals who share a common pattern
of development.

Their technique did identify discrete and robust clinical patterns of progression to amyotrophic lateral sclerosis, many of
which are nonlinear.

Remodeling health declines

After consulting clinicians, a team of machine learning researchers and neurologists let the data speak for
itself.

First, they applied the model to five longitudinal datasets
of ALS clinical trials and observational studies.

New processes and utility mechanisms

When their population-level models were trained and tested on these metrics, four major disease patterns jumped out of many trajectories — S-type rapid development, steady slow development, unstable slow development, and unstable moderate development — many of which had strong nonlinear characteristics
.

The researchers compared their method to other commonly used linear and nonlinear methods in the field to distinguish the contribution
of clustering and linearity to model accuracy.

The researchers' approach also provides insights into Alzheimer's and Parkinson's disease, both of which have a range of symptomatic manifestations and progressions
.

This work has made significant progress
in finding signals in the noise of time series of complex neurodegenerative diseases.

With the advent of new ways to understand the mechanisms of disease, the model provides another tool to distinguish diseases like ALS, Alzheimer's disease, and Parkinson's disease
from a systems biology perspective.

Fraenkel's lab understands the cause of the disease and possible therapeutic goals by looking at cellular changes, saying, "We get a lot of molecular data from the same patients, so our long-term goal is to see if there are subtypes
of this disease.