Alzheimer's disease diagnosis based on the Hippocampal Unified Multi-Atlas Network (HUMAN) algorithm

Artificial Intelligence-Assisted Hippocampal Segmentation and Its Diagnostic Value for Alzheimer's Disease: A Meta-analysis

BACKGROUND

Hippocampal atrophy is a key marker of Alzheimer's disease (AD) and mild cognitive impairment (MCI). Diverse artificial intelligence (AI) architectures for automated hippocampal segmentation have been increasingly reported in neuroimaging research. Different hippocampal automated segmentation methods can be of added value for the AD diagnostic work-up and treatment planning. This study aims to conduct a thorough meta-analysis to evaluate the segmentation accuracy and diagnostic performance of AI-assisted hippocampal segmentation in AD and MCI.

METHODS

We searched PubMed, Embase, Web of Science, and the Cochrane Library up to December 2024. Studies using neuroimaging data to assess AI algorithms for hippocampal segmentation and diagnosis in AD or MCI populations were included. Pooled segmentation accuracy was estimated using the Dice similarity coefficient (DSC) through a random-effects model, while diagnostic performance (sensitivity, specificity, and area under the curve [AUC]) was evaluated using a bivariate mixed-effects model.

RESULTS

A total of 27 studies were included. For segmentation accuracy, pooled DSC values were 0.82 (95% CI: 0.80-0.85) for AD, 0.85 (0.83-0.88) for MCI, and 0.86 (0.84-0.88) for normal controls (NC). Subgroup analyses indicated comparable performance between left and right hippocampi (both DSC: 0.87). Diagnostic meta-analysis demonstrated the highest accuracy for AD vs. NC (sensitivity: 0.87, specificity: 0.91, AUC: 0.95), but lower performance for AD vs. MCI (AUC: 0.80) and MCI vs. NC (AUC: 0.83).

CONCLUSION

AI-assisted hippocampal segmentation achieves good accuracy and demonstrates promising diagnostic capabilities for distinguishing AD from NC, though differentiation between AD and MCI remains challenging. Future high-quality research that applied standardized protocols, external validation, and clinical integration is needed to improve reliability in clinical practice.

Classification method based on surf and sift features for alzheimer diagnosis using diffusion tensor magnetic resonance imaging

Alzheimer's disease (AD), the most common dementia in the elderly, poses a challenge for early diagnosis due to its progressive nature and hidden microstructural changes. While traditional T1 and T2 weighted MRI can assess macro-structural brain atrophy, diffusion tensor imaging (DTI) unveils these hidden microstructural alterations. This study explores the use of DTI data, specifically visual patterns in Fractional Anisotropy (FA), Mean Diffusivity (MD), and Radial Diffusivity (RD) maps, to characterize AD progression. This paper proposes a computer-aided diagnosis (CAD) framework employing SIFT and SURF descriptors and a bag-of-words approach to build AD-specific signatures for the hippocampus region, known to be heavily affected by the disease. These signatures are extracted from MD, FA, and RD maps and used to differentiate between AD, mild cognitive impairment (MCI), and normal controls (NC) in both multiclass and binary classification scenarios. Additionally, we investigate late fusion of visual map features for enhanced decision-making. The experiments were accomplished with a subset of participants from the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset formed of AD patients (n = 35), Early Mild Cognitive Impairment (EMCI) (n = 6), Late Mild Cognitive Impairment (LMCI) (n = 24) and cognitively healthy elderly Normal Controls (NC) (n = 31). Promising preliminary results demonstrate the potential of the proposed system as a useful tool to capture the AD leanness with achieving accuracies of 87.5%, 87.4%, 89%, and 95.2% for MD, FA, RD, and fusion of features respectively for the multiclass system using SIFT features. Using FA features for binary discrimination achieves 97.5%. Moreover, the fusion based on the decision level model reached an accuracy of 93.3% AD/MCI, 95.7% AD/NC, and 93.3% MCI/NC (96.2 ± 3.6 MCI vs. NC, 97.5 ± 5 AD vs. NC). Furthermore, fusion of features led to a noteworthy precision boost of 96%. These findings suggest that our DTI-based CAD framework holds promise as a reliable and accurate tool for capturing AD progression, paving the way for earlier diagnosis and potentially improved patient outcomes.

Hierarchical based classification method based on fusion of Gaussian map descriptors for Alzheimer diagnosis using T1-weighted magnetic resonance imaging

Alzheimer's disease (AD) is considered one of the most spouting elderly diseases. In 2015, AD is reported the US's sixth cause of death. Substantially, non-invasive imaging is widely employed to provide biomarkers supporting AD screening, diagnosis, and progression. In this study, Gaussian descriptors-based features are proposed to be efficient new biomarkers using Magnetic Resonance Imaging (MRI) T1-weighted images to differentiate between Alzheimer's disease (AD), Mild Cognitive Impairment (MCI), and Normal controls (NC). Several Gaussian map-based features are extracted such as Gaussian shape operator, Gaussian curvature, and mean curvature. The aforementioned features are then introduced to the Support Vector Machine (SVM). They were, first, calculated separately for the Hippocampus and Amygdala. Followed by the fusion of the features. Moreover, Fusion of the regions before feature extraction was also employed. Alzheimer's disease Neuroimaging Initiative (ADNI) dataset, formed of 45, 55, and 65 cases for AD, MCI, and NC respectively, is appointed in this study. The shape operator feature outperformed the other features, with 74.6%, and 98.9% accuracy in the case of normal vs. abnormal, and AD vs. MCI classification respectively.

Explainable machine learning with pairwise interactions for the classification of Parkinson's disease and SWEDD from clinical and imaging features

Scans without evidence of dopaminergic deficit (SWEDD) refers to patients who mimics motor and non-motor symptoms of Parkinson's disease (PD) but showing integrity of dopaminergic system. For this reason, the differential diagnosis between SWEDD and PD patients is often not possible in absence of dopamine imaging. Machine Learning (ML) showed optimal performance in automatically distinguishing these two diseases from clinical and imaging data. However, the most common applied ML algorithms provide high accuracy at expense of findings intelligibility. In this work, a novel ML glass-box model, the Explainable Boosting Machine (EBM), based on Generalized Additive Models plus interactions (GA2Ms), was employed to obtain interpretability in classifying PD and SWEDD while still providing optimal performance. Dataset (168 healthy controls, HC; 396 PD; 58 SWEDD) was obtained from PPMI database and consisted of 178 among clinical and imaging features. Six binary EBM classifiers were trained on feature space with (SBR) and without (noSBR) dopaminergic striatal specific binding ratio: HC-PDSBR, HC-SWEDDSBR, PD-SWEDDSBR and HC-PDnoSBR, HC-SWEDDnoSBR, PD-SWEDDnoSBR. Excellent AUC-ROC (1) was reached in classifying HC from PD and SWEDD, both with and without SBR, and by PD-SWEDDSBR (0.986), while PD-SWEDDnoSBR showed lower AUC-ROC (0.882). Apart from optimal accuracies, EBM algorithm was able to provide global and local explanations, revealing that the presence of pairwise interactions between UPSIT Booklet #1 and Epworth Sleepiness Scale item 3 (ESS3), MDS-UPDRS-III pronation-supination movements right hand (NP3PRSPR) and MDS-UPDRS-III rigidity left upper limb (NP3RIGLU) could provide good performance in predicting PD and SWEDD also without imaging features.

Microcalcification Discrimination in Mammography Using Deep Convolutional Neural Network: Towards Rapid and Early Breast Cancer Diagnosis

Breast cancer is among the most common types of cancer in women and under the cases of misdiagnosed, or delayed in treatment, the mortality risk is high. The existence of breast microcalcifications is common in breast cancer patients and they are an effective indicator for early sign of breast cancer. However, microcalcifications are often missed and wrongly classified during screening due to their small sizes and indirect scattering in mammogram images. Motivated by this issue, this project proposes an adaptive transfer learning deep convolutional neural network in segmenting breast mammogram images with calcifications cases for early breast cancer diagnosis and intervention. Mammogram images of breast microcalcifications are utilized to train several deep neural network models and their performance is compared. Image filtering of the region of interest images was conducted to remove possible artifacts and noises to enhance the quality of the images before the training. Different hyperparameters such as epoch, batch size, etc were tuned to obtain the best possible result. In addition, the performance of the proposed fine-tuned hyperparameter of ResNet50 is compared with another state-of-the-art machine learning network such as ResNet34, VGG16, and AlexNet. Confusion matrices were utilized for comparison. The result from this study shows that the proposed ResNet50 achieves the highest accuracy with a value of 97.58%, followed by ResNet34 of 97.35%, VGG16 96.97%, and finally AlexNet of 83.06%.

Applications of artificial intelligence to aid early detection of dementia: A scoping review on current capabilities and future directions

BACKGROUND & OBJECTIVE

With populations aging, the number of people with dementia worldwide is expected to triple to 152 million by 2050. Seventy percent of cases are due to Alzheimer's disease (AD) pathology and there is a 10-20 year 'pre-clinical' period before significant cognitive decline occurs. We urgently need, cost effective, objective biomarkers to detect AD, and other dementias, at an early stage. Risk factor modification could prevent 40% of cases and drug trials would have greater chances of success if participants are recruited at an earlier stage. Currently, detection of dementia is largely by pen and paper cognitive tests but these are time consuming and insensitive to the pre-clinical phase. Specialist brain scans and body fluid biomarkers can detect the earliest stages of dementia but are too invasive or expensive for widespread use. With the advancement of technology, Artificial Intelligence (AI) shows promising results in assisting with detection of early-stage dementia. This scoping review aims to summarise the current capabilities of AI-aided digital biomarkers to aid in early detection of dementia, and also discusses potential future research directions.

METHODS & MATERIALS

In this scoping review, we used PubMed and IEEE Xplore to identify relevant papers. The resulting records were further filtered to retrieve articles published within five years and written in English. Duplicates were removed, titles and abstracts were screened and full texts were reviewed.

RESULTS

After an initial yield of 1,463 records, 1,444 records were screened after removal of duplication. A further 771 records were excluded after screening titles and abstracts, and 496 were excluded after full text review. The final yield was 177 studies. Records were grouped into different artificial intelligence based tests: (a) computerized cognitive tests (b) movement tests (c) speech, conversion, and language tests and (d) computer-assisted interpretation of brain scans.

CONCLUSIONS

In general, AI techniques enhance the performance of dementia screening tests because more features can be retrieved from a single test, there are less errors due to subjective judgements and AI shifts the automation of dementia screening to a higher level. Compared with traditional cognitive tests, AI-based computerized cognitive tests improve the discrimination sensitivity by around 4% and specificity by around 3%. In terms of speech, conversation and language tests, combining both acoustic features and linguistic features achieve the best result with accuracy around 94%. Deep learning techniques applied in brain scan analysis achieves around 92% accuracy. Movement tests and setting smart environments to capture daily life behaviours are two potential future directions that may help discriminate dementia from normal aging. AI-based smart environments and multi-modal tests are promising future directions to improve detection of dementia in the earliest stages.

Classification of Alzheimer’s Disease Based on Abnormal Hippocampal Functional Connectivity and Machine Learning

Objective

Alzheimer’s disease (AD) is a neurodegenerative disease characterized by progressive deterioration of memory and cognition. Mild cognitive impairment (MCI) has been implicated as a prodromal phase of AD. Although abnormal functional connectivity (FC) has been demonstrated in AD and MCI, the clinical differentiation of AD, MCI, and normal aging remains difficult, and the distinction between MCI and normal aging is especially problematic. We hypothesized that FC between the hippocampus and other brain structures is altered in AD and MCI, and that measurement of abnormal FC could have diagnostic utility for the classification of different AD stages.

Methods

Elderly adults aged 60–85 years were assigned to AD, MCI, or normal control (NC) groups based on clinical criteria. Functional magnetic resonance scanning was completed by 119 subjects. Five dimension reduction/classification methods were applied, using hippocampus-derived FC strengths as input features. Classification performance of the five dimensionality reduction methods was compared between AD, MCI, and NC groups.

Results

FCs between the hippocampus and left insula, left thalamus, cerebellum, right lingual gyrus, posterior cingulate cortex, and precuneus were significantly reduced in AD and MCI. Support vector machine learning coupled with sparse principal component analysis demonstrated the best discriminative performance, yielding classification accuracies of 82.02% (AD vs. NC), 81.33% (MCI vs. NC), and 81.08% (AD vs. MCI).

Conclusion

Hippocampus-seed-based FCs were significantly different between AD, MCI, and NC groups. FC assessment combined with widely used machine learning methods can improve AD differential diagnosis, and may be especially useful to distinguish MCI from normal aging.

Multiplex Networks to Characterize Seizure Development in Traumatic Brain Injury Patients

Traumatic brain injury (TBI) may cause secondary debilitating problems, such as post-traumatic epilepsy (PTE), which occurs with unprovoked recurrent seizures, months or even years after TBI. Currently, the Epilepsy Bioinformatics Study for Antiepileptogenic Therapy (EpiBioS4Rx) has been enrolling moderate-severe TBI patients with the goal to identify biomarkers of epileptogenesis that may help to prevent seizure occurrence and better understand the mechanism underlying PTE. In this work, we used a novel complex network approach based on segmenting T1-weighted Magnetic Resonance Imaging (MRI) scans in patches of the same dimension (network nodes) and measured pairwise patch similarities using Pearson's correlation (network connections). This network model allowed us to obtain a series of single and multiplex network metrics to comprehensively analyze the different interactions between brain components and capture structural MRI alterations related to seizure development. We used these complex network features to train a Random Forest (RF) classifier and predict, with an accuracy of 70 and a 95% confidence interval of [67, 73%], which subjects from EpiBioS4Rx have had at least one seizure after a TBI. This complex network approach also allowed the identification of the most informative scales and brain areas for the discrimination between the two clinical groups: seizure-free and seizure-affected subjects, demonstrating to be a promising pilot study which, in the future, may serve to identify and validate biomarkers of PTE.

Impairment of Everyday Spatial Navigation Abilities in Mild Cognitive Impairment Is Weakly Associated with Reduced Grey Matter Volume in the Medial Part of the Entorhinal Cortex

BACKGROUND

Research in rodents identified specific neuron populations encoding information for spatial navigation with particularly high density in the medial part of the entorhinal cortex (ERC), which may be homologous with Brodmann area 34 (BA34) in the human brain.

OBJECTIVE

The aim of this study was to test whether impaired spatial navigation frequently occurring in mild cognitive impairment (MCI) is specifically associated with neurodegeneration in BA34.

METHODS

The study included baseline data of MCI patients enrolled in the Alzheimer's Disease Neuroimaging Initiative with high-resolution structural MRI, brain FDG PET, and complete visuospatial ability scores of the Everyday Cognition test (VS-ECog) within 30 days of PET. A standard mask of BA34 predefined in MNI space was mapped to individual native space to determine grey matter volume and metabolic activity in BA34 on MRI and on (partial volume corrected) FDG PET, respectively. The association of the VS-ECog sum score with grey matter volume and metabolic activity in BA34, APOE4 carrier status, age, education, and global cognition (ADAS-cog-13 score) was tested by linear regression. BA28, which constitutes the lateral part of the ERC, was used as control region.

RESULTS

The eligibility criteria led to inclusion of 379 MCI subjects. The VS-ECog sum score was negatively correlated with grey matter volume in BA34 (β= -0.229, p = 0.022) and age (β= -0.124, p = 0.036), and was positively correlated with ADAS-cog-13 (β= 0.175, p = 0.003). None of the other predictor variables contributed significantly.

CONCLUSION

Impairment of spatial navigation in MCI is weakly associated with BA34 atrophy.

A multi-model deep convolutional neural network for automatic hippocampus segmentation and classification in Alzheimer's disease

Alzheimer's disease (AD) is a progressive and irreversible brain degenerative disorder. Mild cognitive impairment (MCI) is a clinical precursor of AD. Although some treatments can delay its progression, no effective cures are available for AD. Accurate early-stage diagnosis of AD is vital for the prevention and intervention of the disease progression. Hippocampus is one of the first affected brain regions in AD. To help AD diagnosis, the shape and volume of the hippocampus are often measured using structural magnetic resonance imaging (MRI). However, these features encode limited information and may suffer from segmentation errors. Additionally, the extraction of these features is independent of the classification model, which could result in sub-optimal performance. In this study, we propose a multi-model deep learning framework based on convolutional neural network (CNN) for joint automatic hippocampal segmentation and AD classification using structural MRI data. Firstly, a multi-task deep CNN model is constructed for jointly learning hippocampal segmentation and disease classification. Then, we construct a 3D Densely Connected Convolutional Networks (3D DenseNet) to learn features of the 3D patches extracted based on the hippocampal segmentation results for the classification task. Finally, the learned features from the multi-task CNN and DenseNet models are combined to classify disease status. Our method is evaluated on the baseline T1-weighted structural MRI data collected from 97 AD, 233 MCI, 119 Normal Control (NC) subjects in the Alzheimer's Disease Neuroimaging Initiative (ADNI) database. The proposed method achieves a dice similarity coefficient of 87.0% for hippocampal segmentation. In addition, the proposed method achieves an accuracy of 88.9% and an AUC (area under the ROC curve) of 92.5% for classifying AD vs. NC subjects, and an accuracy of 76.2% and an AUC of 77.5% for classifying MCI vs. NC subjects. Our empirical study also demonstrates that the proposed multi-model method outperforms the single-model methods and several other competing methods.