Using machine learning to understand neuromorphological change and image-based biomarker identification in Cavalier King Charles Spaniels with Chiari-like malformation-associated pain and syringomyeli

Spiteri M, Knowler SP, Rusbridge C, Wells K. Using machine learning to understand neuromorphological change and image-based biomarker identification in Cavalier King Charles Spaniels with Chiari-like malformation-associated pain and syringomyelia. J Vet Intern Med. 2019. doi:10.1111/jvim.15621.

This retrospective study applied machine learning to analyze neuromorphological changes and identify imaging biomarkers in Cavalier King Charles Spaniels (CKCS) with Chiari-like malformation-associated pain (CM-P) and symptomatic syringomyelia (SM-S). Using T2W midsagittal MRI images and Demons image registration, deformation features were extracted and classified via a kernelized Support Vector Machine. The results showed an area under the curve (AUC) of 0.78 for CM-P biomarkers and 0.82 for SM biomarkers, indicating good sensitivity and specificity. These findings suggest machine learning can aid in diagnosing CM/SM and understanding morphological abnormalities in affected dogs.