Antemortem differential diagnosis of dementia pathology using structural MRI: Differential-STAND

Abstract

The common neurodegenerative pathologies underlying dementia are Alzheimer's disease (AD), Lewy body disease (LBD) and frontotemporal lobar degeneration (FTLD). Our aim was to identify patterns of atrophy unique to each of these diseases using antemortem structural MRI scans of pathologically confirmed dementia cases and build an MRI-based differential diagnosis system. Our approach of creating atrophy maps using structural MRI and applying them for classification of new incoming patients is labeled Differential-STAND (Differential Diagnosis Based on Structural Abnormality in Neurodegeneration). Pathologically confirmed subjects with a single dementing pathologic diagnosis who had an MRI at the time of clinical diagnosis of dementia were identified: 48 AD, 20 LBD, 47 FTLD-TDP (pathology-confirmed FTLD with TDP-43). Gray matter density in 91 regions-of-interest was measured in each subject and adjusted for head size and age using a database of 120 cognitively normal elderly. The atrophy patterns in each dementia type when compared to pathologically confirmed controls mirrored known disease-specific anatomic patterns: AD-temporoparietal association cortices and medial temporal lobe; FTLD-TDP-frontal and temporal lobes and LBD-bilateral amygdalae, dorsal midbrain and inferior temporal lobes. Differential-STAND based classification of each case was done based on a mixture model generated using bisecting k-means clustering of the information from the MRI scans. Leave-one-out classification showed reasonable performance compared to the autopsy gold standard and clinical diagnosis: AD (sensitivity: 90.7%; specificity: 84%), LBD (sensitivity: 78.6%; specificity: 98.8%) and FTLD-TDP (sensitivity: 84.4%; specificity: 93.8%). The proposed approach establishes a direct a priori relationship between specific topographic patterns on MRI and "gold standard" of pathology which can then be used to predict underlying dementia pathology in new incoming patients.

Fig 1

A) Schematic of the Differential-STAND system: atrophy in the new incoming scan is compared to the library of scans of pathologically confirmed cases and classified as AD, LBD or FTLD-TDP. B) Illustration of the clustering and classification approach. The dementia subtypes (or equivalently, patient clusters) are represented by ovals: white for AD, red for LBD and green for FTLD-TDP. An incoming patient, depicted as the square is diagnosed unambiguously as suffering from FTLD-TDP because he falls into the patient cluster #3 which is a subtype of FTLD-TDP. However an incoming patient depicted as a triangle will be diagnosed as belonging to cluster #2 which is a mixture of AD and LBD cases with similar atrophy patterns. Some inevitable overlap remains, mostly due to similarities in atrophy patterns between the dementia subtypes such as clusters 1 and 2.

Fig. 2

Differential-STAND Map patterns of gray matter loss that are specific to each dementia (AD, LBD, FTLD-TDP) identified when compared to pathology confirmed CN. Colorbar represents the t-statistic (p<0.01 uncorrected).

Fig. 3

Typical Differential-STAND Maps of three different patients: AD, LBD and FTLD-TDP (Z-scores<−1). Colorbar represents the atrophy level relative to the CN reference image database and absolute value of the Z-score is shown.

Fig. 4

The multi-dimensional scaled dimensions 1 and 2 of a specific instance of the estimated stable clusters and their corresponding labels using the proposed differential STAND approach. The multi-dimensional scaled dimensions represent a transformation of the high dimensional Z-score GM densities and the transformation preserves the distances between the cluster centers and the cluster radii.