Jong Hun Kim

This paper shows that polygenic risk scores (PRS) overestimate predictive accuracy when datasets have overlapping subjects, using Alzheimer's disease and other traits as examples.

This study uses machine learning to predict Alzheimer's disease incidence from administrative health data, achieving moderate accuracy in 1-4 year predictions.

This paper uses multimodal MRI and machine learning to classify Alzheimer's disease and mild cognitive impairment, showing high accuracy in diagnosis and progression prediction.

This paper uses machine learning on brain morphometry and connectomes from MRI to predict Alzheimer's disease risk and progression, achieving high accuracy in both single-site and multi-site datasets.