Time-Series & Sensors

Objective FRDA Assessment in Children

Objective Assessment of Friedreich Ataxia in Children: Accounting for Developmental Deficits

Measuring disease in children without mistaking normal growing pains for symptoms.

The problem

A child’s maturing nervous system produces erratic movement and sway that look like ataxia in both clinical and sensor scores.

The approach

Device-specific correction models remove developmental effects: reinforcement learning for the cup, Bayesian optimisation for the spoon and a multilayer perceptron for the pendant.

The outcome

Severity scores that reflect the disease alone, giving clinicians more precise tracking of progression in children.

In detail

Investigated a key challenge in paediatric Friedreich ataxia: distinguishing disease-related motor impairment from movement variability caused by normal nervous-system development. Device-specific correction models were developed for AIM-C, AIM-S and AIM-P data using reinforcement learning, Bayesian optimisation and neural-network methods, with the goal of producing severity scores that better isolate the effect of FRDA itself.

Built with

Source

More in Time-Series & Sensors

Next project: Dual-Relationship Modeling of Friedreich Ataxia

By Kanishka Ranaweera, PhD, Machine Learning Engineer in Melbourne, Victoria, Australia. hello@kanishkabuilt.me