Fig. 5From: Comparing a pre-defined versus deep learning approach for extracting brain atrophy patterns to predict cognitive decline due to Alzheimer’s disease in patients with mild cognitive symptomsCorrelation between predicted slope and corresponding ground truth. Models using a) clinical data (age, sex, and education, baseline cognition [MMSE score, ADAS delayed word recall] and APOE genotype), b) clinical data, intracranial volume and hippocampal volume, c) clinical data, intracranial volume and FreeSurfer, d) clinical data, MRI and JD. The blue data shows the BioFINDER test sets and the red data shows the external evaluation on ADNI dataBack to article page