Machine learning models were able to accurately predict those patients with a diagnosis of optic pathway glioma (OPG) or attention-deficit/hyperactivity disorder (ADHD) based on readily available clinical and demographic data in children with neurofibromatosis type 1 (NF1).
Why this matters
NF1 is a common genetic condition affecting around 1 in every 3,000 births with extreme clinical variability. Currently, clinicians are unable to predict the disease course in an individual child and tailored care is challenging.
OPG and ADHD are both known manifestations of NF1. It is unclear whether any clinical or demographic factors are associated with increased risk of developing these complications, and whether a machine learning approach based on this data could help to predict disease course in an individual child.