Integrating Demographics and Imaging Features for Various Stages of Dementia Classification: Feed Forward Neural Network Multi-Class Approach

BACKGROUND

MRI magnetization-prepared rapid acquisition (MPRAGE) is an easily available imaging modality for dementia diagnosis. Previous studies suggested that volumetric analysis plays a crucial role in various stages of dementia classification. In this study, volumetry, radiomics and demographics were integrated as inputs to develop an artificial intelligence model for various stages, including Alzheimer's disease (AD), mild cognitive decline (MCI) and cognitive normal (CN) dementia classifications.

METHOD

The Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset was separated into training and testing groups, and the Open Access Series of Imaging Studies (OASIS) dataset was used as the second testing group. The MRI MPRAGE image was reoriented via statistical parametric mapping (SPM12). Freesurfer was employed for brain segmentation, and 45 regional brain volumes were retrieved. The 3D Slicer software was employed for 107 radiomics feature extractions from within the whole brain. Data on patient demographics were collected from the datasets. The feed-forward neural network (FFNN) and the other most common artificial intelligence algorithms, including support vector machine (SVM), ensemble classifier (EC) and decision tree (DT), were used to build the models using various features.

RESULTS

The integration of brain regional volumes, radiomics and patient demographics attained the highest overall accuracy at 76.57% and 73.14% in ADNI and OASIS testing, respectively. The subclass accuracies in MCI, AD and CN were 78.29%, 89.71% and 85.14%, respectively, in ADNI testing, as well as 74.86%, 88% and 83.43% in OASIS testing. Balanced sensitivity and specificity were obtained for all subclass classifications in MCI, AD and CN.

CONCLUSION

The FFNN yielded good overall accuracy for MCI, AD and CN categorization, with balanced subclass accuracy, sensitivity and specificity. The proposed FFNN model is simple, and it may support the triage of patients for further confirmation of the diagnosis.

AI Deployment on GBM Diagnosis: A Novel Approach to Analyze Histopathological Images Using Image Feature-Based Analysis

BACKGROUND

Glioblastoma (GBM) is one of the most common malignant primary brain tumors, which accounts for 60-70% of all gliomas. Conventional diagnosis and the decision of post-operation treatment plan for glioblastoma is mainly based on the feature-based qualitative analysis of hematoxylin and eosin-stained (H&E) histopathological slides by both an experienced medical technologist and a pathologist. The recent development of digital whole slide scanners makes AI-based histopathological image analysis feasible and helps to diagnose cancer by accurately counting cell types and/or quantitative analysis. However, the technology available for digital slide image analysis is still very limited. This study aimed to build an image feature-based computer model using histopathology whole slide images to differentiate patients with glioblastoma (GBM) from healthy control (HC).

METHOD

Two independent cohorts of patients were used. The first cohort was composed of 262 GBM patients of the Cancer Genome Atlas Glioblastoma Multiform Collection (TCGA-GBM) dataset from the cancer imaging archive (TCIA) database. The second cohort was composed of 60 GBM patients collected from a local hospital. Also, a group of 60 participants with no known brain disease were collected. All the H&E slides were collected. Thirty-three image features (22 GLCM and 11 GLRLM) were retrieved from the tumor volume delineated by medical technologist on H&E slides. Five machine-learning algorithms including decision-tree (DT), extreme-boost (EB), support vector machine (SVM), random forest (RF), and linear model (LM) were used to build five models using the image features extracted from the first cohort of patients. Models built were deployed using the selected key image features for GBM diagnosis from the second cohort (local patients) as model testing, to identify and verify key image features for GBM diagnosis.

RESULTS

All five machine learning algorithms demonstrated excellent performance in GBM diagnosis and achieved an overall accuracy of 100% in the training and validation stage. A total of 12 GLCM and 3 GLRLM image features were identified and they showed a significant difference between the normal and the GBM image. However, only the SVM model maintained its excellent performance in the deployment of the models using the independent local cohort, with an accuracy of 93.5%, sensitivity of 86.95%, and specificity of 99.73%.

CONCLUSION

In this study, we have identified 12 GLCM and 3 GLRLM image features which can aid the GBM diagnosis. Among the five models built, the SVM model proposed in this study demonstrated excellent accuracy with very good sensitivity and specificity. It could potentially be used for GBM diagnosis and future clinical application.

Level of Amyloid-β (Aβ) Binding Leading to Differential Effects on Resting State Functional Connectivity in Major Brain Networks

Introduction: Amyloid-β protein (Aβ) is one of the biomarkers for Alzheimer’s disease (AD). The recent application of interhemispheric functional connectivity (IFC) in resting-state fMRI has been used as a non-invasive diagnostic tool for early dementia. In this study, we focused on the level of Aβ accumulated and its effects on the major functional networks, including default mode network (DMN), central executive network (CEN), salience network (SN), self-referential network (SRN) and sensory motor network (SMN). Methods: 58 participants (27 Hi Aβ (HiAmy) and 31 low Aβ (LowAmy)) and 25 healthy controls (HC) were recruited. [¹⁸F]flutemetamol PET/CT was performed for diseased groups, and MRI scanning was done for all participants. Voxel-by-voxel correlation analysis was done for both groups in all networks. Results: In HiAmy, IFC was reduced in all networks except SN. A negative correlation in DMN, CEN, SRN and SMN suggests high Aβ related to IFC reduction; However, a positive correlation in SN suggests high Aβ related to an increase in IFC. In LowAmy, IFC increased in CEN, SMN, SN and SRN. Positive correlation in all major brain networks. Conclusion: The level of Aβ accumulated demonstrated differential effects on IFC in various brain networks. As the treatment to reduce Aβ plaque deposition is available in the market, it may be an option for the HiAmy group to improve their IFC in major brain networks.

A scoping review of resting-state brain functional alterations in Type 2 diabetes

Resting-state functional magnetic resonance imaging (rs-fMRI) has been actively used in the last decade to investigate brain functional connectivity alterations in Type 2 Diabetes Mellitus (T2DM) to understand the neuropathophysiology of T2DM in cognitive degeneration. Given the emergence of new analysis techniques, this scoping review aims to map the rs-fMRI analysis techniques that have been applied in the literature and reports the latest rs-fMRI findings that have not been covered in previous reviews. Graph theory, the contemporary rs-fMRI analysis, has been used to demonstrate altered brain topological organisations in people with T2DM, which included altered degree centrality, functional connectivity strength, the small-world architecture and network-based statistics. These alterations were correlated with T2DM patients' cognitive performances. Graph theory also contributes to identify unbiased seeds for seed-based analysis. The expanding rs-fMRI analytical approaches continue to provide new evidence that helps to understand the mechanisms of T2DM-related cognitive degeneration.

Impaired cerebral blood flow in type 2 diabetes mellitus - A comparative study with subjective cognitive decline, vascular dementia and Alzheimer's disease subjects

•

CBF impairment is found in T2DM and SCD individuals, which might suggest a preclinical stage of dementia.

•

Comparing to HC, lower CBF in T2DM was due to higher rate of multiple cerebrovascular risk factors.

•

Unlike T2DM, CBF reduction in AD and VD was due to amyloid deposition and microangiopathy respectively.

•

Significant negative correlation between adjusted CBF and HbA1c in all cortical regions in healthy control and T2DM.

The link between non-demented type 2 diabetes mellitus (T2DM) and different types of cognitive impairment is controversial. By controlling for co-morbidities such as cerebral macrovascular and microvascular changes, cerebral atrophy, amyloid burden, hypertension or hyperlipidemia, the current study investigated the cerebral blood flow of T2DM individuals as compared to cognitively impaired subjects recruited from a memory clinic.

15 healthy control (71.8 ± 6.1 years), 18 T2DM (62.5 ± 3.7 years), as well as 8 Subjective Cognitive Decline (69.5 ± 7.5 years), 12 Vascular Dementia (79.3 ± 4.2 years) and 17 Alzheimer’s Disease (75.1 ± 8.2 years) underwent multi-parametric MRI brain scanning. Subjects with T2DM and from the memory clinic also had 18-F Flutametamol PET-CT scanning to look for any amyloid burden. Pseudocontinuous Arterial Spin Labeling (PCASL), MR Angiography Head, 3D FLAIR and 3D T1-weighted sequences were used to quantify cerebral blood flow, cerebrovascular changes, white matter hyperintensities and brain atrophy respectively. Vascular risk factors were retrieved from the medical records. The 37 subjects from memory clinic were classified into subjective cognitive decline (SCD), vascular dementia (VD) and Alzheimer’s disease (AD) subgroups by a multi-disciplinary panel consisting of a neuroradiologist, and 2 geriatricians.

Absolute cortical CBF in our cohort of T2DM, SCD, VD and AD was significantly decreased (p < 0.01) as compared to healthy controls (HC) in both whole brain and eight paired brain regions, after age, normalized grey matter volume and gender adjustment and Bonferroni correction.

Subgroup analysis between T2DM, SCD, VD, and AD revealed that CBF of T2DM was not significantly different from AD, VD or SCD. By controlling for co-morbidities, impaired cortical CBF in T2DM was not related to microangiopathy or amyloid deposition, but to the interaction of triple risk factors (such as diabetes mellitus, hypertension, and hyperlipidemia).

There was statistically significant negative correlation (p ≤ 0.05) between adjusted CBF and HbA1c in all brain regions of T2DM and HC (with partial correlation ranging from −0.30 to −0.46).

Taken together, altered cerebral blood flow in T2DM might be related to disruption of cerebrovascular autoregulation related to vascular risk factors, and such oligemia occurred before clinical manifestation due to altered glycemic control.

Solar insolation in springtime influences age of onset of bipolar I disorder

OBJECTIVE

To confirm prior findings that the larger the maximum monthly increase in solar insolation in springtime, the younger the age of onset of bipolar disorder.

METHOD

Data were collected from 5536 patients at 50 sites in 32 countries on six continents. Onset occurred at 456 locations in 57 countries. Variables included solar insolation, birth-cohort, family history, polarity of first episode and country physician density.

RESULTS

There was a significant, inverse association between the maximum monthly increase in solar insolation at the onset location, and the age of onset. This effect was reduced in those without a family history of mood disorders and with a first episode of mania rather than depression. The maximum monthly increase occurred in springtime. The youngest birth-cohort had the youngest age of onset. All prior relationships were confirmed using both the entire sample, and only the youngest birth-cohort (all estimated coefficients P < 0.001).

CONCLUSION

A large increase in springtime solar insolation may impact the onset of bipolar disorder, especially with a family history of mood disorders. Recent societal changes that affect light exposure (LED lighting, mobile devices backlit with LEDs) may influence adaptability to a springtime circadian challenge.

Psychiatric morbidity in Chinese women after cervical cancer treatment in a regional gynaecology clinic

OBJECTIVES

To identify the prevalence and factors associated with psychiatric disorders in Chinese cervical cancer survivors.

METHODS

A cross-sectional study was conducted from May 2011 to April 2012 at the specialist gynaecology outpatient clinic at Pamela Youde Nethersole Eastern Hospital, Hong Kong. All cervical cancer patients who had completed treatment were consecutively recruited. They were interviewed using the Chinese-Bilingual Structured Clinical Interview for DSM-IV Axis I Disorders, Patient Research version. Socio-demographic data and clinical information were collected from the patients and their hospital records were reviewed.

RESULTS

A total of 113 patients were recruited into the study. The point prevalence of psychiatric disorders as a group in cervical cancer survivors was 37%. The point prevalence of depressive disorders, anxiety disorders, and schizophrenia were 31%, 16%, and 2%, respectively. Major depressive disorder was the most common mood disorder and generalised anxiety disorder the most common anxiety disorder. Younger age, a history of psychiatric illness, fatigue, menopausal symptoms, and pain were independent predictors of current psychiatric disorders.

CONCLUSION

Psychiatric disorders, predominantly depressive and anxiety disorders, are common in Chinese cervical cancer survivors. Identification of independent predictors can help gynaecologists detect these disorders earlier and arrange appropriate interventions.