Search

IntroductionThe increasing prevalence of Alzheimer’s disease (AD) has created an urgent need for rapid and cost-effective methods to diagnose and monitor people at all stages of the disease. Progressive memory impairment and hippocampal atrophy are key features of the most common so-called typical variant of AD. However, studies evaluating detailed cognitive measures combined with region of interest (ROI)-based imaging markers of progression over the long term in the AD dementia (ADD) stage are rare.MethodWe conducted a retrospective longitudinal follow-up study in patients with mild to moderate ADD (aged 60-92 years). They underwent magnetic resonance imaging (MRI; 3 Tesla, MPRAGE) as well as clinical and neuropsychological examination (Consortium to Establish a Registry for Alzheimer’s Disease [CERAD] -Plus test battery) at baseline and at least one follow-up visit. ROI-based brain structural analysis of baseline MRIs was performed using the Computational Anatomy Toolbox (CAT) 12. Clinical dementia progression (progression index [PI]) was measured by the annual decline in the Mini Mental State Examination (MMSE) scores. MRI, demographic, and neuropsychological data were included in univariate and multiple linear regression models to predict the PI.Results104 ADD patients (age 63 to 90 years, 73% female, mean MMSE score 22.63 ± 3.77, mean follow-up 4.27 ± 2.15 years) and 32 age- and gender-matched cognitively intact controls were included. The pattern of gray matter (GM) atrophy and the cognitive profile were consistent with the amnestic/typical variant of ADD in all patients. Deficits in word list learning together with temporal lobe GM atrophy had the highest predictive value for rapid cognitive decline in the multiple linear regression model, accounting for 25.4% of the PI variance.DiscussionOur results show that temporal atrophy together with deficits in the encoding of verbal material, rather than in immediate or delayed recall, is highly predictive for rapid cognitive decline in patients with mild to moderate amnestic/typical ADD. These findings point to the relevance of combining detailed cognitive and automated structural imaging analyses to predict clinical progression in patients with ADD.

Objective: To develop a machine learning-based nomogram for distinguishing between supratentorial extraventricular ependymoma (STEE) and supratentorial glioblastoma (GBM). Methods: We conducted a retrospective analysis on MRI datasets obtained from 140 patients who were diagnosed with STEE (n=48) and GBM (n=92) from two institutions. Initially, we compared seven different machine learning algorithms to determine the most suitable signature (rad-score). Subsequently, univariate and multivariate logistic regression analyses were performed to identify significant clinical predictors that can differentiate between STEE and GBM. Finally, we developed a nomogram by visualizing the rad-score and clinical features for clinical evaluation. Results: The TreeBagger (TB) outperformed the other six algorithms, yielding the best diagnostic efficacy in differentiating STEE from GBM, with area under the curve (AUC) values of 0.735 (95% CI: 0.625-0.845) and 0.796 (95% CI: 0.644-0.949) in the training set and test set. Furthermore, the nomogram incorporating both the rad-score and clinical variables demonstrated a robust predictive performance with an accuracy of 0.787 in the training set and 0.832 in the test set. Conclusion: The nomogram could serve as a valuable tool for non-invasively discriminating between STEE and GBM. [ABSTRACT FROM AUTHOR]