Accuracy of CT scan measurements of the medial temporal lobe in routine dementia diagnostics

Classification of CT brain images based on deep learning networks

While computerised tomography (CT) may have been the first imaging tool to study human brain, it has not yet been implemented into clinical decision making process for diagnosis of Alzheimer's disease (AD). On the other hand, with the nature of being prevalent, inexpensive and non-invasive, CT does present diagnostic features of AD to a great extent. This study explores the significance and impact on the application of the burgeoning deep learning techniques to the task of classification of CT brain images, in particular utilising convolutional neural network (CNN), aiming at providing supplementary information for the early diagnosis of Alzheimer's disease. Towards this end, three categories of CT images (N = 285) are clustered into three groups, which are AD, lesion (e.g. tumour) and normal ageing. In addition, considering the characteristics of this collection with larger thickness along the direction of depth (z) (~3-5 mm), an advanced CNN architecture is established integrating both 2D and 3D CNN networks. The fusion of the two CNN networks is subsequently coordinated based on the average of Softmax scores obtained from both networks consolidating 2D images along spatial axial directions and 3D segmented blocks respectively. As a result, the classification accuracy rates rendered by this elaborated CNN architecture are 85.2%, 80% and 95.3% for classes of AD, lesion and normal respectively with an average of 87.6%. Additionally, this improved CNN network appears to outperform the others when in comparison with 2D version only of CNN network as well as a number of state of the art hand-crafted approaches. As a result, these approaches deliver accuracy rates in percentage of 86.3, 85.6 ± 1.10, 86.3 ± 1.04, 85.2 ± 1.60, 83.1 ± 0.35 for 2D CNN, 2D SIFT, 2D KAZE, 3D SIFT and 3D KAZE respectively. The two major contributions of the paper constitute a new 3-D approach while applying deep learning technique to extract signature information rooted in both 2D slices and 3D blocks of CT images and an elaborated hand-crated approach of 3D KAZE.

Is there any role for computed tomography measurements of medial temporal lobe atrophy in dementia? A review of the literature and case series from a memory clinic

Neuroimaging in dementia has focused on documenting any burden of vascular disease or excluding any reversible intracranial pathology. We review the use of computed tomography to examine for medial temporal lobe atrophy in dementia and compare this with a case series of such measurements from our memory clinic. Measures of medial temporal lobe atrophy were used to separate patients with Alzheimer's disease from those with normal cognition, mood disorders or other forms of early dementia.

Predicting the outcome of cholinesterase inhibitor treatment in Alzheimer's disease

OBJECTIVE

To investigate the possibility that response to cholinesterase inhibitor therapy could be predicted by easily measurable variables that are known to change as a result of treatment (such as the Mini Mental State Examination), measures of function (such as the instrumental activities of daily living and the social behaviour subscales of the Nurse's Observation Scale for Geriatric Patients), and measures of attention (such as the Digit Symbol Substitution Test; DSST), or that might influence response through structural (for example, age, cerebrovascular disease, medial temporal lobe (MTL) atrophy, hypertension) or chemical (for example, smoking) mechanisms.

METHOD

This was a cohort study of 160 consecutive outpatients with probable Alzheimer's disease who commenced cholinesterase inhibitor treatment over a 3 year period in a semi-rural area of Scotland.

RESULTS

The overall response rate was 42.1%. Stratification of response between good and poor responders was possible using baseline DSST and a measure of MTL thickness using CT. Among the patients, 60.4% of those above the cut off point for both DSST and MTL thickness (29/48 subjects) were classified as good responders, compared with 6.3% of subjects below the cut off point for both (1/16 subjects). Subjects above the cut off point for both measures were more likely to be classified as good responders than subjects with only one or no values above the respective cut off points (chi(2) = 10.61, df = 1, p = 0.001)

CONCLUSIONS

The DSST and a measure of MTL thickness derived from CT scanning may be useful in improving the prediction of response to cholinesterase inhibitors in subjects with AD. Subjects with low DSST scores and more severe MTL atrophy are unlikely to respond to treatment. These preliminary data justify a prospective trial of the usefulness of our suggested predictive measures.

Nanoparticle-based detection in cerebral spinal fluid of a soluble pathogenic biomarker for Alzheimer's disease

The recently developed ultrasensitive bio-barcode assay was used to measure the concentration of amyloid-beta-derived diffusible ligands (ADDLs), a potential soluble pathogenic Alzheimer's disease (AD) marker, in the cerebrospinal fluid (CSF) of 30 individuals. ADDL concentrations for the subjects diagnosed with AD were consistently higher than the levels in the CSF taken from nondemented age-matched controls. Studies of ADDLs or for any other potential pathogenic AD markers in CSF have not been possible because of their low concentration in CSF (<1 pM). This study is a step toward a diagnostic tool, based on soluble pathogenic markers for the debilitating disease.