CMC: A consensus multi-view clustering model for predicting Alzheimer's disease progression

A multi-view learning approach with diffusion model to synthesize FDG PET from MRI T1WI for diagnosis of Alzheimer's disease

INTRODUCTION

This study presents a novel multi-view learning approach for machine learning (ML)-based Alzheimer's disease (AD) diagnosis.

METHODS

A diffusion model is proposed to synthesize the fluorodeoxyglucose positron emission tomography (FDG PET) view from the magnetic resonance imaging T1 weighted imaging (MRI T1WI) view and incorporate two synthesis strategies: one-way synthesis and two-way synthesis. To assess the utility of the synthesized views, we use multilayer perceptron (MLP)-based classifiers with various combinations of the views.

RESULTS

The two-way synthesis achieves state-of-the-art performance with a structural similarity index measure (SSIM) at 0.9380 and a peak-signal-to-noise ratio (PSNR) at 26.47. The one-way synthesis achieves an SSIM at 0.9282 and a PSNR at 23.83. Both synthesized FDG PET views have shown their effectiveness in improving diagnostic accuracy.

DISCUSSION

This work supports the notion that ML-based cross-domain data synthesis can be a useful approach to improve AD diagnosis by providing additional synthesized disease-related views for multi-view learning.

HIGHLIGHTS

We propose a diffusion model with two strategies to synthesize fluorodeoxyglucose positron emission tomography (FDG PET) from magnetic resonance imaging T1 weighted imaging (MRI T1WI). We raise multi-view learning with MRl T1Wl and synthesized FDG PET for Alzheimer's disease (AD) diagnosis. We provide a comprehensive experimental comparison for the synthesized FDG PET view. The feasibility of synthesized FDG PET view in AD diagnosis is validated with various experiments. We demonstrate the ability of synthesized FDG PET to enhance the performance of machine learning-based AD diagnosis.

Tackling the small data problem in medical image classification with artificial intelligence: a systematic review

Though medical imaging has seen a growing interest in AI research, training models require a large amount of data. In this domain, there are limited sets of data available as collecting new data is either not feasible or requires burdensome resources. Researchers are facing with the problem of small datasets and have to apply tricks to fight overfitting. 147 peer-reviewed articles were retrieved from PubMed, published in English, up until 31 July 2022 and articles were assessed by two independent reviewers. We followed the Preferred Reporting Items for Systematic reviews and Meta-Analyse (PRISMA) guidelines for the paper selection and 77 studies were regarded as eligible for the scope of this review. Adherence to reporting standards was assessed by using TRIPOD statement (transparent reporting of a multivariable prediction model for individual prognosis or diagnosis). To solve the small data issue transfer learning technique, basic data augmentation and generative adversarial network were applied in 75%, 69% and 14% of cases, respectively. More than 60% of the authors performed a binary classification given the data scarcity and the difficulty of the tasks. Concerning generalizability, only four studies explicitly stated an external validation of the developed model was carried out. Full access to all datasets and code was severely limited (unavailable in more than 80% of studies). Adherence to reporting standards was suboptimal (<50% adherence for 13 of 37 TRIPOD items). The goal of this review is to provide a comprehensive survey of recent advancements in dealing with small medical images samples size. Transparency and improve quality in publications as well as follow existing reporting standards are also supported.

A prognostic prediction model for ovarian cancer using a cross-modal view correlation discovery network

Ovarian cancer is a tumor with different clinicopathological and molecular features, and the vast majority of patients have local or extensive spread at the time of diagnosis. Early diagnosis and prognostic prediction of patients can contribute to the understanding of the underlying pathogenesis of ovarian cancer and the improvement of therapeutic outcomes. The occurrence of ovarian cancer is influenced by multiple complex mechanisms, including the genome, transcriptome and proteome. Different types of omics analysis help predict the survival rate of ovarian cancer patients. Multi-omics data of ovarian cancer exhibit high-dimensional heterogeneity, and existing methods for integrating multi-omics data have not taken into account the variability and inter-correlation between different omics data. In this paper, we propose a deep learning model, MDCADON, which utilizes multi-omics data and cross-modal view correlation discovery network. We introduce random forest into LASSO regression for feature selection on mRNA expression, DNA methylation, miRNA expression and copy number variation (CNV), aiming to select important features highly correlated with ovarian cancer prognosis. A multi-modal deep neural network is used to comprehensively learn feature representations of each omics data and clinical data, and cross-modal view correlation discovery network is employed to construct the multi-omics discovery tensor, exploring the inter-relationships between different omics data. The experimental results demonstrate that MDCADON is superior to the existing methods in predicting ovarian cancer prognosis, which enables survival analysis for patients and facilitates the determination of follow-up treatment plans. Finally, we perform Gene Ontology (GO) term analysis and biological pathway analysis on the genes identified by MDCADON, revealing the underlying mechanisms of ovarian cancer and providing certain support for guiding ovarian cancer treatments.

Multi-omics data fusion using adaptive GTO guided Non-negative matrix factorization for cancer subtype discovery

BACKGROUND AND OBJECTIVE

Cancer subtype discovery is essential for personalized clinical treatment. With the onset of progressive profile techniques for cancer, a large amount of heterogeneous and high-dimensional transcriptomic, proteomic and genomic datasets are easily accumulated. Integrative clustering of such multi-omics data is crucial to recognize their latent structure and to acknowledge the correlation within and across them. Although the integrative analysis of diversified multi-omics data is informative, it is challenging when multiplicity in data inflicts poor accordance w.r.t. clustering structure. The objective of this work is to develop an effective integrative analysis framework that encapsulates the heterogeneity of various biological mechanisms and predicts homogeneous subgroups of cancer patients.

METHOD

In this paper, improved sparse-joint non-negative matrix factorization (sparse-jNMF) has been devised for the problem of cancer-subtype discovery. The initial points of sparse-jNMF have improved using a novel meta-heuristic algorithm adaptive gorilla troops optimizer (Ada-GTO). Improving the initialization of sparse-jNMF enhances its convergence behavior and further strengthens the clustering performance. In addition, the consensus clustering approach has been adopted to construct a patient-patient similarity matrix for obtaining stable clusters of patient samples.

RESULT

The proposed framework has been applied to 4 different real-life multi-omics cancer datasets, namely colon adenocarcinoma, breast-invasive carcinoma, kidney-renal clear-cell carcinoma, and lung adenocarcinoma. The proposed method results in patient clusters with better silhouette scores and cluster purity than classical initialization and similar meta-heuristics for initial point estimation approaches. Survival probabilities estimated using Kaplan-Meier (KM) curve show statistically significant difference (p < 0.05) for the homogenous cancer patient clusters obtained using the proposed method as compared to iCluster. The presented approach further identified the somatic mutations for the classified subgroups, which is beneficial to provide targeted treatments.

CONCLUSION

This paper proposes Ada-GTO guided sparse-jNMF framework for cancer subtype discovery, considering the information from multiple omic features that provide comprehension. The proposed meta-guided framework outperforms all other state-of-the-art counterparts. It also has great potential for obtaining the homogeneous subgroups of other diseases.

Multi-modal intermediate integrative methods in neuropsychiatric disorders: A review

The etiology of neuropsychiatric disorders involves complex biological processes at different omics layers, such as genomics, transcriptomics, epigenetics, proteomics, and metabolomics. The advent of high-throughput technology, as well as the availability of large open-source datasets, has ushered in a new era in system biology, necessitating the integration of various types of omics data. The complexity of biological mechanisms, the limitations of integrative strategies, and the heterogeneity of multi-omics data have all presented significant challenges to computational scientists. In comparison to early and late integration, intermediate integration may transform each data type into appropriate intermediate representations using various data transformation techniques, allowing it to capture more complementary information contained in each omics and highlight new interactions across omics layers. Here, we reviewed multi-modal intermediate integrative techniques based on component analysis, matrix factorization, similarity network, multiple kernel learning, Bayesian network, artificial neural networks, and graph transformation, as well as their applications in neuropsychiatric domains. We depicted advancements in these approaches and compared the strengths and weaknesses of each method examined. We believe that our findings will aid researchers in their understanding of the transformation and integration of multi-omics data in neuropsychiatric disorders.

Physical activity and risk of multiple sclerosis: A Mendelian randomization study

Multiple evidence from epidemiological studies has suggested association between physical activity and risk of multiple sclerosis (MS). However, the conclusion was still controversial between studies, and whether the association was causal or confounded is elusive. To evaluate the role of physical activity with different intensities in the risk of MS, we first estimated their genetic correlation, and then conducted two-sample and multivariable Mendelian randomization analyses based on summary statistics from previous large genome-wide association studies. A significant genetic correlation was identified between moderate physical activity and the risk of MS (genetic correlation: -0.15, SE=0.05, P=2.9E-03). Meanwhile, higher moderate physical activity was significantly associated with a reduced risk of MS (OR:0.87, 95% CI:0.80-0.96, P=3.45E-03). Such association was further verified using summary statistics from another study on overall physical activity (OR:0.36, 95% CI:0.17-0.76, P=6.82E-03). The results were robust under all sensitivity analyses. Current results suggested moderate physical activity could reduce the risk of MS. These findings help better understand the role of physical activity in MS, and provide some lifestyle recommendations for individuals susceptible to MS.

Logarithmic Negation of Basic Probability Assignment and Its Application in Target Recognition

The negation of probability distribution is a new perspective from which to obtain information. Dempster–Shafer (D–S) evidence theory, as an extension of possibility theory, is widely used in decision-making-level fusion. However, how to reasonably construct the negation of basic probability assignment (BPA) in D–S evidence theory is an open issue. This paper proposes a new negation of BPA, logarithmic negation. It solves the shortcoming of Yin’s negation that maximal entropy cannot be obtained when there are only two focal elements in the BPA. At the same time, the logarithmic negation of BPA inherits the good properties of the negation of probability, such as order reversal, involution, convergence, degeneration, and maximal entropy. Logarithmic negation degenerates into Gao’s negation when the values of the elements all approach 0. In addition, the data fusion method based on logarithmic negation has a higher belief value of the correct target in target recognition application.

Rethinking modeling Alzheimer's disease progression from a multi-task learning perspective with deep recurrent neural network

Alzheimer's disease (AD) is a severe neurodegenerative disorder that usually starts slowly and progressively worsens. Predicting the progression of Alzheimer's disease with longitudinal analysis on the time series data has recently received increasing attention. However, training an accurate progression model for brain network faces two major challenges: missing features, and the small sample size during the follow-up study. According to our analysis on the AD progression task, we thoroughly analyze the correlation among the multiple predictive tasks of AD progression at multiple time points. Thus, we propose a multi-task learning framework that can adaptively impute missing values and predict future progression over time from a subject's historical measurements. Progression is measured in terms of MRI volumetric measurements, trajectories of a cognitive score and clinical status. To this end, we propose a new perspective of predicting the AD progression with a multi-task learning paradigm. In our multi-task learning paradigm, we hypothesize that the inherent correlations exist among: (i). the prediction tasks of clinical diagnosis, cognition and ventricular volume at each time point; (ii). the tasks of imputation and prediction; and (iii). the prediction tasks at multiple future time points. According to our findings of the task correlation, we develop an end-to-end deep multi-task learning method to jointly improve the performance of assigning missing value and prediction. We design a balanced multi-task dynamic weight optimization. With in-depth analysis and empirical evidence on Alzheimer's Disease Neuroimaging Initiative (ADNI), we show the benefits and flexibility of the proposed multi-task learning model, especially for the prediction at the M60 time point. The proposed approach achieves 5.6%, 5.7%, 4.0% and 11.8% improvement with respect to mAUC, BCA and MAE (ADAS-Cog13 and Ventricles), respectively.

Disease Modelling of Cognitive Outcomes and Biomarkers in the European Prevention of Alzheimer's Dementia Longitudinal Cohort

A key challenge for the secondary prevention of Alzheimer’s dementia is the need to identify individuals early on in the disease process through sensitive cognitive tests and biomarkers. The European Prevention of Alzheimer’s Dementia (EPAD) consortium recruited participants into a longitudinal cohort study with the aim of building a readiness cohort for a proof-of-concept clinical trial and also to generate a rich longitudinal data-set for disease modelling. Data have been collected on a wide range of measurements including cognitive outcomes, neuroimaging, cerebrospinal fluid biomarkers, genetics and other clinical and environmental risk factors, and are available for 1,828 eligible participants at baseline, 1,567 at 6 months, 1,188 at one-year follow-up, 383 at 2 years, and 89 participants at three-year follow-up visit. We novelly apply state-of-the-art longitudinal modelling and risk stratification approaches to these data in order to characterise disease progression and biological heterogeneity within the cohort. Specifically, we use longitudinal class-specific mixed effects models to characterise the different clinical disease trajectories and a semi-supervised Bayesian clustering approach to explore whether participants can be stratified into homogeneous subgroups that have different patterns of cognitive functioning evolution, while also having subgroup-specific profiles in terms of baseline biomarkers and longitudinal rate of change in biomarkers.

Early Alzheimer's disease diagnosis with the contrastive loss using paired structural MRIs

BACKGROUND AND OBJECTIVE

Alzheimer's Disease (AD) is a chronic and fatal neurodegenerative disease with progressive impairment of memory. Brain structural magnetic resonance imaging (sMRI) has been widely applied as important biomarkers of AD. Various machine learning approaches, especially deep learning-based models, have been proposed for the early diagnosis of AD and monitoring the disease progression on sMRI data. However, the requirement for a large number of training images still hinders the extensive usage of AD diagnosis. In addition, due to the similarities in human whole-brain structure, finding the subtle brain changes is essential to extract discriminative features from limited sMRI data effectively.

METHODS

In this work, we proposed two types of contrastive losses with paired sMRIs to promote the diagnostic performance using group categories (G-CAT) and varying subject mini-mental state examination (S-MMSE) information, respectively. Specifically, G-CAT contrastive loss layer was used to learn the closer feature representation from sMRIs with the same categories, while ranking information from S-MMSE assists the model to explore subtle changes between individuals.

RESULTS

The model was trained on ADNI-1. Comparison with baseline methods was performed on MIRIAD and ADNI-2. For the classification task on MIRIAD, S-MMSE achieves 93.5% of accuracy, 96.6% of sensitivity, and 94.9% of specificity, respectively. G-CAT and S-MMSE both reach remarkable performance in terms of classification sensitivity and specificity respectively. Comparing with state-of-the-art methods, we found this proposed method could achieve comparable results with other approaches.

CONCLUSION

The proposed model could extract discriminative features under whole-brain similarity. Extensive experiments also support the accuracy of this model, i.e., it provides better ability to identify uncertain samples, especially for the classification task of subjects with MMSE in 22-27. Source code is freely available at https://github.com/fengduqianhe/ADComparative.

Deep sequence modelling for Alzheimer's disease detection using MRI

BACKGROUND

Alzheimer's disease (AD) is one of the deadliest diseases in developed countries. Treatments following early AD detection can significantly delay institutionalisation and extend patients' independence. There has been a growing focus on early AD detection using artificial intelligence. Convolutional neural networks (CNNs) have proven revolutionary for image-based applications and have been applied to brain scans. In recent years, studies have utilised two-dimensional (2D) CNNs on magnetic resonance imaging (MRI) scans for AD detection. To apply a 2D CNN on three-dimensional (3D) MRI volumes, each MRI scan is split into 2D image slices. A CNN is trained over the image slices by calculating a loss function between each subject's label and each image slice's predicted output. Although 2D CNNs can discover spatial dependencies in an image slice, they cannot understand the temporal dependencies among 2D image slices in a 3D MRI volume. This study aims to resolve this issue by modelling the sequence of MRI features produced by a CNN with deep sequence-based networks for AD detection.

METHOD

The CNN utilised in this paper was ResNet-18 pre-trained on an ImageNet dataset. The employed sequence-based models were the temporal convolutional network (TCN) and different types of recurrent neural networks. Several deep sequence-based models and configurations were implemented and compared for AD detection.

RESULTS

Our proposed TCN model achieved the best classification performance with 91.78% accuracy, 91.56% sensitivity and 92% specificity.

CONCLUSION

Our results show that applying sequence-based models can improve the classification accuracy of 2D and 3D CNNs for AD detection by up to 10%.