Multimodal and Multiscale Deep Neural Networks for the Early Diagnosis of Alzheimer's Disease using structural MR and FDG-PET images

Differential Expression of mRNAs in Peripheral Blood Related to Prodrome and Progression of Alzheimer's Disease

Alzheimer's disease (AD) is a chronic progressive neurodegenerative disease that affects the quality of life of elderly individuals, while the pathogenesis of AD is still unclear. Based on the bioinformatics analysis of differentially expressed genes (DEGs) in peripheral blood samples, we investigated genes related to mild cognitive impairment (MCI), AD, and late-stage AD that might be used for predicting the conversions. Methods. We obtained the DEGs in MCI, AD, and advanced AD patients from the Gene Expression Omnibus (GEO) database. A Venn diagram was used to identify the intersecting genes. Gene Ontology (GO) and Kyoto Gene and Genomic Encyclopedia (KEGG) were used to analyze the functions and pathways of the intersecting genes. Protein-protein interaction (PPI) networks were constructed to visualize the network of the proteins coded by the related genes. Hub genes were selected based on the PPI network. Results. Bioinformatics analysis indicated that there were 61 DEGs in both the MCI and AD groups and 27 the same DEGs among the three groups. Using GO and KEGG analyses, we found that these genes were related to the function of mitochondria and ribosome. Hub genes were determined by bioinformatics software based on the PPI network. Conclusions. Mitochondrial and ribosomal dysfunction in peripheral blood may be early signs in AD patients and related to the disease progression. The identified hub genes may provide the possibility for predicting AD progression or be the possible targets for treatments.

Differential Diagnosis of Frontotemporal Dementia, Alzheimer's Disease, and Normal Aging Using a Multi-Scale Multi-Type Feature Generative Adversarial Deep Neural Network on Structural Magnetic Resonance Images

Methods: Alzheimer's disease and Frontotemporal dementia are the first and third most common forms of dementia. Due to their similar clinical symptoms, they are easily misdiagnosed as each other even with sophisticated clinical guidelines. For disease-specific intervention and treatment, it is essential to develop a computer-aided system to improve the accuracy of their differential diagnosis. Recent advances in deep learning have delivered some of the best performance for medical image recognition tasks. However, its application to the differential diagnosis of AD and FTD pathology has not been explored. Approach: In this study, we proposed a novel deep learning based framework to distinguish between brain images of normal aging individuals and subjects with AD and FTD. Specifically, we combined the multi-scale and multi-type MRI-base image features with Generative Adversarial Network data augmentation technique to improve the differential diagnosis accuracy. Results: Each of the multi-scale, multitype, and data augmentation methods improved the ability for differential diagnosis for both AD and FTD. A 10-fold cross validation experiment performed on a large sample of 1,954 images using the proposed framework achieved a high overall accuracy of 88.28%. Conclusions: The salient contributions of this study are three-fold: (1) our experiments demonstrate that the combination of multiple structural features extracted at different scales with our proposed deep neural network yields superior performance than individual features; (2) we show that the use of Generative Adversarial Network for data augmentation could further improve the discriminant ability of the network regarding challenging tasks such as differentiating dementia sub-types; (3) and finally, we show that ensemble classifier strategy could make the network more robust and stable.

Fueling Clinical and Translational Research in Appalachia: Informatics Platform Approach

BACKGROUND

The Appalachian population is distinct, not just culturally and geographically but also in its health care needs, facing the most health care disparities in the United States. To meet these unique demands, Appalachian medical centers need an arsenal of analytics and data science tools with the foundation of a centralized data warehouse to transform health care data into actionable clinical interventions. However, this is an especially challenging task given the fragmented state of medical data within Appalachia and the need for integration of other types of data such as environmental, social, and economic with medical data.

OBJECTIVE

This paper aims to present the structure and process of the development of an integrated platform at a midlevel Appalachian academic medical center along with its initial uses.

METHODS

The Appalachian Informatics Platform was developed by the Appalachian Clinical and Translational Science Institute's Division of Clinical Informatics and consists of 4 major components: a centralized clinical data warehouse, modeling (statistical and machine learning), visualization, and model evaluation. Data from different clinical systems, billing systems, and state- or national-level data sets were integrated into a centralized data warehouse. The platform supports research efforts by enabling curation and analysis of data using the different components, as appropriate.

RESULTS

The Appalachian Informatics Platform is functional and has supported several research efforts since its implementation for a variety of purposes, such as increasing knowledge of the pathophysiology of diseases, risk identification, risk prediction, and health care resource utilization research and estimation of the economic impact of diseases.

CONCLUSIONS

The platform provides an inexpensive yet seamless way to translate clinical and translational research ideas into clinical applications for regions similar to Appalachia that have limited resources and a largely rural population.

Using machine learning to quantify structural MRI neurodegeneration patterns of Alzheimer's disease into dementia score: Independent validation on 8,834 images from ADNI, AIBL, OASIS, and MIRIAD databases

Biomarkers for dementia of Alzheimer's type (DAT) are sought to facilitate accurate prediction of the disease onset, ideally predating the onset of cognitive deterioration. T1-weighted magnetic resonance imaging (MRI) is a commonly used neuroimaging modality for measuring brain structure in vivo, potentially providing information enabling the design of biomarkers for DAT. We propose a novel biomarker using structural MRI volume-based features to compute a similarity score for the individual's structural patterns relative to those observed in the DAT group. We employed ensemble-learning framework that combines structural features in most discriminative ROIs to create an aggregate measure of neurodegeneration in the brain. This classifier is trained on 423 stable normal control (NC) and 330 DAT subjects, where clinical diagnosis is likely to have the highest certainty. Independent validation on 8,834 unseen images from ADNI, AIBL, OASIS, and MIRIAD Alzheimer's disease (AD) databases showed promising potential to predict the development of DAT depending on the time-to-conversion (TTC). Classification performance on stable versus progressive mild cognitive impairment (MCI) groups achieved an AUC of 0.81 for TTC of 6 months and 0.73 for TTC of up to 7 years, achieving state-of-the-art results. The output score, indicating similarity to patterns seen in DAT, provides an intuitive measure of how closely the individual's brain features resemble the DAT group. This score can be used for assessing the presence of AD structural atrophy patterns in normal aging and MCI stages, as well as monitoring the progression of the individual's brain along with the disease course.

[Artificial intelligence in hybrid imaging]

CLINICAL ISSUE

Hybrid imaging enables the precise visualization of cellular metabolism by combining anatomical and metabolic information. Advances in artificial intelligence (AI) offer new methods for processing and evaluating this data.

METHODOLOGICAL INNOVATIONS

This review summarizes current developments and applications of AI methods in hybrid imaging. Applications in image processing as well as methods for disease-related evaluation are presented and discussed.

MATERIALS AND METHODS

This article is based on a selective literature search with the search engines PubMed and arXiv.

ASSESSMENT

Currently, there are only a few AI applications using hybrid imaging data and no applications are established in clinical routine yet. Although the first promising approaches are emerging, they still need to be evaluated prospectively. In the future, AI applications will support radiologists and nuclear medicine radiologists in diagnosis and therapy.

A Novel Metabolic Connectome Method to Predict Progression to Mild Cognitive Impairment

OBJECTIVE

Glucose-based positron emission tomography (PET) imaging has been widely used to predict the progression of mild cognitive impairment (MCI) into Alzheimer's disease (AD) clinically. However, existing discriminant methods are unsubtle to reveal pathophysiological changes. Therefore, we present a novel metabolic connectome-based predictive modeling to predict progression from MCI to AD accurately.

METHODS

In this study, we acquired fluorodeoxyglucose PET images and clinical assessments from 420 MCI patients with 36 months follow-up. Individual metabolic network based on connectome analysis was constructed, and the metabolic connectivity in this network was extracted as predictive features. Three different classification strategies were implemented to interrogate the predictive performance. To verify the effectivity of selected features, specific brain regions associated with MCI conversion were identified based on these features and compared with prior knowledge.

RESULTS

As a result, 4005 connectome features were obtained, and 153 in which were selected as efficient features. Our proposed feature extraction method had achieved 85.2% accuracy for MCI conversion prediction (sensitivity: 88.1%; specificity: 81.2%; and AUC: 0.933). The discriminative brain regions associated with MCI conversion were mainly located in the precentral gyrus, precuneus, lingual, and inferior frontal gyrus.

CONCLUSION

Overall, the results suggest that our proposed individual metabolic connectome method has great potential to predict whether MCI patients will progress to AD. The metabolic connectome may help to identify brain metabolic dysfunction and build a clinically applicable biomarker to predict the MCI progression.

β-amyloid and tau drive early Alzheimer's disease decline while glucose hypometabolism drives late decline

Clinical trials focusing on therapeutic candidates that modify β-amyloid (Aβ) have repeatedly failed to treat Alzheimer’s disease (AD), suggesting that Aβ may not be the optimal target for treating AD. The evaluation of Aβ, tau, and neurodegenerative (A/T/N) biomarkers has been proposed for classifying AD. However, it remains unclear whether disturbances in each arm of the A/T/N framework contribute equally throughout the progression of AD. Here, using the random forest machine learning method to analyze participants in the Alzheimer’s Disease Neuroimaging Initiative dataset, we show that A/T/N biomarkers show varying importance in predicting AD development, with elevated biomarkers of Aβ and tau better predicting early dementia status, and biomarkers of neurodegeneration, especially glucose hypometabolism, better predicting later dementia status. Our results suggest that AD treatments may also need to be disease stage-oriented with Aβ and tau as targets in early AD and glucose metabolism as a target in later AD.

Here the authors analyze the Alzheimer’s Disease Neuroimaging Initiative dataset using random forest machine learning methods and determine that Aβ and tau biomarkers are better predictors of early dementia status, while glucose hypometabolism is a better predictor of later dementia status. These results suggest the need for stage-oriented Alzheimer’s disease treatments.

Predicting rate of cognitive decline at baseline using a deep neural network with multidata analysis

Purpose: Our study investigates whether a machine-learning-based system can predict the rate of cognitive decline in mildly cognitively impaired patients by processing only the clinical and imaging data collected at the initial visit. Approach: We built a predictive model based on a supervised hybrid neural network utilizing a three-dimensional convolutional neural network to perform volume analysis of magnetic resonance imaging (MRI) and integration of nonimaging clinical data at the fully connected layer of the architecture. The experiments are conducted on the Alzheimer's Disease Neuroimaging Initiative dataset. Results: Experimental results confirm that there is a correlation between cognitive decline and the data obtained at the first visit. The system achieved an area under the receiver operator curve of 0.70 for cognitive decline class prediction. Conclusion: To our knowledge, this is the first study that predicts "slowly deteriorating/stable" or "rapidly deteriorating" classes by processing routinely collected baseline clinical and demographic data [baseline MRI, baseline mini-mental state examination (MMSE), scalar volumetric data, age, gender, education, ethnicity, and race]. The training data are built based on MMSE-rate values. Unlike the studies in the literature that focus on predicting mild cognitive impairment (MCI)-to-Alzheimer's disease conversion and disease classification, we approach the problem as an early prediction of cognitive decline rate in MCI patients.

Predicting Alzheimer Disease From Mild Cognitive Impairment With a Deep Belief Network Based on 18F-FDG-PET Images

Objective:

Accurate diagnosis of early Alzheimer disease (AD) plays a critical role in preventing the progression of memory impairment. We aimed to develop a new deep belief network (DBN) framework using 18F-fluorodeoxyglucose (FDG) positron emission tomography (PET) metabolic imaging to identify patients at the mild cognitive impairment (MCI) stage with presymptomatic AD and to discriminate them from other patients with MCI.

Methods:

18F-fluorodeoxyglucose-PET images of 109 patients recruited in the ongoing longitudinal Alzheimer’s Disease Neuroimaging Initiative study were included in this analysis. Patients were grouped into 2 classes: (1) stable mild cognitive impairment (n = 62) or (2) progressive mild cognitive impairment (n = 47). Our framework is composed of 4 steps: (1) image preprocessing: normalization and smoothing; (2) identification of regions of interest (ROIs); (3) feature learning using deep neural networks; and (4) classification by support vector machine with 3 kernels. All classification experiments were performed with a 5-fold cross-validation. Accuracy, sensitivity, and specificity were used to validate the results.

Result:

A total of 1103 ROIs were obtained. One hundred features were learned from ROIs using the DBN. The classification accuracy using linear, polynomial, and RBF kernels was 83.9%, 79.2%, and 86.6%, respectively. This method may be a powerful tool for personalized precision medicine in the population with prediction of early AD progression.

Improving PET Imaging Acquisition and Analysis With Machine Learning: A Narrative Review With Focus on Alzheimer's Disease and Oncology

Machine learning (ML) algorithms have found increasing utility in the medical imaging field and numerous applications in the analysis of digital biomarkers within positron emission tomography (PET) imaging have emerged. Interest in the use of artificial intelligence in PET imaging for the study of neurodegenerative diseases and oncology stems from the potential for such techniques to streamline decision support for physicians providing early and accurate diagnosis and allowing personalized treatment regimens. In this review, the use of ML to improve PET image acquisition and reconstruction is presented, along with an overview of its applications in the analysis of PET images for the study of Alzheimer's disease and oncology.

Cohort discovery and risk stratification for Alzheimer's disease: an electronic health record-based approach

BACKGROUND

We sought to leverage data routinely collected in electronic health records (EHRs), with the goal of developing patient risk stratification tools for predicting risk of developing Alzheimer's disease (AD).

METHOD

Using EHR data from the University of Michigan (UM) hospitals and consensus-based diagnoses from the Michigan Alzheimer's Disease Research Center, we developed and validated a cohort discovery tool for identifying patients with AD. Applied to all UM patients, these labels were used to train an EHR-based machine learning model for predicting AD onset within 10 years.

RESULTS

Applied to a test cohort of 1697 UM patients, the model achieved an area under the receiver operating characteristics curve of 0.70 (95% confidence interval = 0.63-0.77). Important predictive factors included cardiovascular factors and laboratory blood testing.

CONCLUSION

Routinely collected EHR data can be used to predict AD onset with modest accuracy. Mining routinely collected data could shed light on early indicators of AD appearance and progression.

Development and validation of an interpretable deep learning framework for Alzheimer's disease classification

Alzheimer's disease is the primary cause of dementia worldwide, with an increasing morbidity burden that may outstrip diagnosis and management capacity as the population ages. Current methods integrate patient history, neuropsychological testing and MRI to identify likely cases, yet effective practices remain variably applied and lacking in sensitivity and specificity. Here we report an interpretable deep learning strategy that delineates unique Alzheimer's disease signatures from multimodal inputs of MRI, age, gender, and Mini-Mental State Examination score. Our framework linked a fully convolutional network, which constructs high resolution maps of disease probability from local brain structure to a multilayer perceptron and generates precise, intuitive visualization of individual Alzheimer's disease risk en route to accurate diagnosis. The model was trained using clinically diagnosed Alzheimer's disease and cognitively normal subjects from the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset (n = 417) and validated on three independent cohorts: the Australian Imaging, Biomarker and Lifestyle Flagship Study of Ageing (AIBL) (n = 382), the Framingham Heart Study (n = 102), and the National Alzheimer's Coordinating Center (NACC) (n = 582). Performance of the model that used the multimodal inputs was consistent across datasets, with mean area under curve values of 0.996, 0.974, 0.876 and 0.954 for the ADNI study, AIBL, Framingham Heart Study and NACC datasets, respectively. Moreover, our approach exceeded the diagnostic performance of a multi-institutional team of practicing neurologists (n = 11), and high-risk cerebral regions predicted by the model closely tracked post-mortem histopathological findings. This framework provides a clinically adaptable strategy for using routinely available imaging techniques such as MRI to generate nuanced neuroimaging signatures for Alzheimer's disease diagnosis, as well as a generalizable approach for linking deep learning to pathophysiological processes in human disease.

Deep residual learning for neuroimaging: An application to predict progression to Alzheimer's disease

BACKGROUND

The unparalleled performance of deep learning approaches in generic image processing has motivated its extension to neuroimaging data. These approaches learn abstract neuroanatomical and functional brain alterations that could enable exceptional performance in classification of brain disorders, predicting disease progression, and localizing brain abnormalities.

NEW METHOD

This work investigates the suitability of a modified form of deep residual neural networks (ResNet) for studying neuroimaging data in the specific application of predicting progression from mild cognitive impairment (MCI) to Alzheimer's disease (AD). Prediction was conducted first by training the deep models using MCI individuals only, followed by a domain transfer learning version that additionally trained on AD and controls. We also demonstrate a network occlusion based method to localize abnormalities.

RESULTS

The implemented framework captured non-linear features that successfully predicted AD progression and also conformed to the spectrum of various clinical scores. In a repeated cross-validated setup, the learnt predictive models showed highly similar peak activations that corresponded to previous AD reports.

COMPARISON WITH EXISTING METHODS

The implemented architecture achieved a significant performance improvement over the classical support vector machine and the stacked autoencoder frameworks (p <  0.005), numerically better than state-of-the-art performance using sMRI data alone (> 7% than the second-best performing method) and within 1% of the state-of-the-art performance considering learning using multiple neuroimaging modalities as well.

CONCLUSIONS

The explored frameworks reflected the high potential of deep learning architectures in learning subtle predictive features and utility in critical applications such as predicting and understanding disease progression.

Diagnostic and Prognostic Classification of Brain Disorders Using Residual Learning on Structural MRI Data

In this work, we study the potential of the deep residual neural network (ResNet) architecture to learn abstract neuroanatomical alterations in the structural MRI data by evaluating its diagnostic and prognostic classification performance on two large, independent multi-group (ADNI and BSNIP) neuroimaging datasets. We conduct several binary classification tasks to assess the diagnostic/prognostic performance of the ResNet architecture through a rigorous, repeated and stratified k-fold cross-validation procedure for each of the classification tasks independently. We obtained better than state of the art performance for the clinically most important task in the ADNI dataset analysis, and significantly higher classification accuracies over a standard machine learning method (linear SVM) in each of the ADNI and BSNIP classification tasks. Overall, our results indicate the high potential of this architecture to learn effectual feature representations from structural brain imaging data.

Multi-modal neuroimaging feature fusion for diagnosis of Alzheimer's disease

BACKGROUND

Compared with single-modal neuroimages classification of AD, multi-modal classification can achieve better performance by fusing different information. Exploring synergy among various multi-modal neuroimages is contributed to identifying the pathological process of neurological disorders. However, it is still problematic to effectively exploit multi-modal information since the lack of an effective fusion method.

NEW METHOD

In this paper, we propose a deep multi-modal fusion network based on the attention mechanism, which can selectively extract features from MRI and PET branches and suppress irrelevant information. In the attention model, the fusion ratio of each modality is assigned automatically according to the importance of the data. A hierarchical fusion method is adopted to ensure the effectiveness of Multi-modal Fusion.

RESULTS

Evaluating the model on the ADNI dataset, the experimental results show that it outperforms the state-of-the-art methods. In particular, the final classification results of the NC/AD, SMCI/PMCI and Four-Class are 95.21 %, 89.79 %, and 86.15 %, respectively.

COMPARISON WITH EXISTING METHODS

Different from the early fusion and the late fusion, the hierarchical fusion method contributes to learning the synergy between the multi-modal data. Compared with some other prominent algorithms, the attention model enables our network to focus on the regions of interest and effectively fuse the multi-modal data.

CONCLUSION

Benefit from the hierarchical structure with attention model, the proposed network is capable of exploiting low-level and high-level features extracted from the multi-modal data and improving the accuracy of AD diagnosis. Results show its promising performance.

Convolutional neural networks for classification of Alzheimer's disease: Overview and reproducible evaluation

Numerous machine learning (ML) approaches have been proposed for automatic classification of Alzheimer's disease (AD) from brain imaging data. In particular, over 30 papers have proposed to use convolutional neural networks (CNN) for AD classification from anatomical MRI. However, the classification performance is difficult to compare across studies due to variations in components such as participant selection, image preprocessing or validation procedure. Moreover, these studies are hardly reproducible because their frameworks are not publicly accessible and because implementation details are lacking. Lastly, some of these papers may report a biased performance due to inadequate or unclear validation or model selection procedures. In the present work, we aim to address these limitations through three main contributions. First, we performed a systematic literature review. We identified four main types of approaches: i) 2D slice-level, ii) 3D patch-level, iii) ROI-based and iv) 3D subject-level CNN. Moreover, we found that more than half of the surveyed papers may have suffered from data leakage and thus reported biased performance. Our second contribution is the extension of our open-source framework for classification of AD using CNN and T1-weighted MRI. The framework comprises previously developed tools to automatically convert ADNI, AIBL and OASIS data into the BIDS standard, and a modular set of image preprocessing procedures, classification architectures and evaluation procedures dedicated to deep learning. Finally, we used this framework to rigorously compare different CNN architectures. The data was split into training/validation/test sets at the very beginning and only the training/validation sets were used for model selection. To avoid any overfitting, the test sets were left untouched until the end of the peer-review process. Overall, the different 3D approaches (3D-subject, 3D-ROI, 3D-patch) achieved similar performances while that of the 2D slice approach was lower. Of note, the different CNN approaches did not perform better than a SVM with voxel-based features. The different approaches generalized well to similar populations but not to datasets with different inclusion criteria or demographical characteristics. All the code of the framework and the experiments is publicly available: general-purpose tools have been integrated into the Clinica software (www.clinica.run) and the paper-specific code is available at: https://github.com/aramis-lab/AD-DL.

Imaging biomarkers in neurodegeneration: current and future practices

There is an increasing role for biological markers (biomarkers) in the understanding and diagnosis of neurodegenerative disorders. The application of imaging biomarkers specifically for the in vivo investigation of neurodegenerative disorders has increased substantially over the past decades and continues to provide further benefits both to the diagnosis and understanding of these diseases. This review forms part of a series of articles which stem from the University College London/University of Gothenburg course "Biomarkers in neurodegenerative diseases". In this review, we focus on neuroimaging, specifically positron emission tomography (PET) and magnetic resonance imaging (MRI), giving an overview of the current established practices clinically and in research as well as new techniques being developed. We will also discuss the use of machine learning (ML) techniques within these fields to provide additional insights to early diagnosis and multimodal analysis.

Deep learning to detect Alzheimer's disease from neuroimaging: A systematic literature review

Alzheimer's Disease (AD) is one of the leading causes of death in developed countries. From a research point of view, impressive results have been reported using computer-aided algorithms, but clinically no practical diagnostic method is available. In recent years, deep models have become popular, especially in dealing with images. Since 2013, deep learning has begun to gain considerable attention in AD detection research, with the number of published papers in this area increasing drastically since 2017. Deep models have been reported to be more accurate for AD detection compared to general machine learning techniques. Nevertheless, AD detection is still challenging, and for classification, it requires a highly discriminative feature representation to separate similar brain patterns. This paper reviews the current state of AD detection using deep learning. Through a systematic literature review of over 100 articles, we set out the most recent findings and trends. Specifically, we review useful biomarkers and features (personal information, genetic data, and brain scans), the necessary pre-processing steps, and different ways of dealing with neuroimaging data originating from single-modality and multi-modality studies. Deep models and their performance are described in detail. Although deep learning has achieved notable performance in detecting AD, there are several limitations, especially regarding the availability of datasets and training procedures.

Semi-supervised Learning Approach to Generate Neuroimaging Modalities with Adversarial Training

Magnetic Resonance Imaging (MRI) of the brain can come in the form of different modalities such as T1-weighted and Fluid Attenuated Inversion Recovery (FLAIR) which has been used to investigate a wide range of neurological disorders. Current state-of-the-art models for brain tissue segmentation and disease classification require multiple modalities for training and inference. However, the acquisition of all of these modalities are expensive, time-consuming, inconvenient and the required modalities are often not available. As a result, these datasets contain large amounts of unpaired data, where examples in the dataset do not contain all modalities. On the other hand, there is smaller fraction of examples that contain all modalities (paired data) and furthermore each modality is high dimensional when compared to number of datapoints. In this work, we develop a method to address these issues with semi-supervised learning in translating between two neuroimaging modalities. Our proposed model, Semi-Supervised Adversarial CycleGAN (SSA-CGAN), uses an adversarial loss to learn from unpaired data points, cycle loss to enforce consistent reconstructions of the mappings and another adversarial loss to take advantage of paired data points. Our experiments demonstrate that our proposed framework produces an improvement in reconstruction error and reduced variance for the pairwise translation of multiple modalities and is more robust to thermal noise when compared to existing methods.

DMCNN: A Deep Multiscale Convolutional Neural Network Model for Medical Image Segmentation

Medical image segmentation is one of the hot issues in the related area of image processing. Precise segmentation for medical images is a vital guarantee for follow-up treatment. At present, however, low gray contrast and blurred tissue boundaries are common in medical images, and the segmentation accuracy of medical images cannot be effectively improved. Especially, deep learning methods need more training samples, which lead to time-consuming process. Therefore, we propose a novelty model for medical image segmentation based on deep multiscale convolutional neural network (CNN) in this article. First, we extract the region of interest from the raw medical images. Then, data augmentation is operated to acquire more training datasets. Our proposed method contains three models: encoder, U-net, and decoder. Encoder is mainly responsible for feature extraction of 2D image slice. The U-net cascades the features of each block of the encoder with those obtained by deconvolution in the decoder under different scales. The decoding is mainly responsible for the upsampling of the feature graph after feature extraction of each group. Simulation results show that the new method can boost the segmentation accuracy. And, it has strong robustness compared with other segmentation methods.

Analyze Informant-Based Questionnaire for The Early Diagnosis of Senile Dementia Using Deep Learning

OBJECTIVE

This paper proposes a multiclass deep learning method for the classification of dementia using an informant-based questionnaire.

METHODS

A deep neural network classification model based on Keras framework is proposed in this paper. To evaluate the advantages of our proposed method, we compared the performance of our model with industry-standard machine learning approaches. We enrolled 6,701 individuals, which were randomly divided into training data sets (6030 participants) and test data sets (671 participants). We evaluated each diagnostic model in the test set using accuracy, precision, recall, and F1-Score.

RESULTS

Compared with the seven conventional machine learning algorithms, the DNN showed higher stability and achieved the best accuracy with 0.88, which also showed good results for identifying normal (F1-score = 0.88), mild cognitive impairment (MCI) (F1-score = 0.87), very mild dementia (VMD) (F1-score = 0.77) and Severe dementia (F1-score = 0.94).

CONCLUSION

The deep neural network (DNN) classification model can effectively help doctors accurately screen patients who have normal cognitive function, mild cognitive impairment (MCI), very mild dementia (VMD), mild dementia (Mild), moderate dementia (Moderate), and severe dementia (Severe).

Artificial intelligence for detecting small FDG-positive lung nodules in digital PET/CT: impact of image reconstructions on diagnostic performance

OBJECTIVES

To evaluate the diagnostic performance of a deep learning algorithm for automated detection of small ¹⁸F-FDG-avid pulmonary nodules in PET scans, and to assess whether novel block sequential regularized expectation maximization (BSREM) reconstruction affects detection accuracy as compared to ordered subset expectation maximization (OSEM) reconstruction.

METHODS

Fifty-seven patients with 92 ¹⁸F-FDG-avid pulmonary nodules (all ≤ 2 cm) undergoing PET/CT for oncological (re-)staging were retrospectively included and a total of 8824 PET images of the lungs were extracted using OSEM and BSREM reconstruction. Per-slice and per-nodule sensitivity of a deep learning algorithm was assessed, with an expert readout by a radiologist/nuclear medicine physician serving as standard of reference. Receiver-operator characteristic (ROC) curve of OSEM and BSREM were assessed and the areas under the ROC curve (AUC) were compared. A maximum standardized uptake value (SUV_max)-based sensitivity analysis and a size-based sensitivity analysis with subgroups defined by nodule size was performed.

RESULTS

The AUC of the deep learning algorithm for nodule detection using OSEM reconstruction was 0.796 (CI 95%; 0.772-0.869), and 0.848 (CI 95%; 0.828-0.869) using BSREM reconstruction. The AUC was significantly higher for BSREM compared to OSEM (p = 0.001). On a per-slice analysis, sensitivity and specificity were 66.7% and 79.0% for OSEM, and 69.2% and 84.5% for BSREM. On a per-nodule analysis, the overall sensitivity of OSEM was 81.5% compared to 87.0% for BSREM.

CONCLUSIONS

Our results suggest that machine learning algorithms may aid detection of small ¹⁸F-FDG-avid pulmonary nodules in clinical PET/CT. AI performed significantly better on images with BSREM than OSEM.

KEY POINTS

• The diagnostic value of deep learning for detecting small lung nodules (≤ 2 cm) in PET images using BSREM and OSEM reconstruction was assessed. • BSREM yields higher SUV_maxof small pulmonary nodules as compared to OSEM reconstruction. • The use of BSREM translates into a higher detectability of small pulmonary nodules in PET images as assessed with artificial intelligence.

Deep neural networks in psychiatry

Machine and deep learning methods, today's core of artificial intelligence, have been applied with increasing success and impact in many commercial and research settings. They are powerful tools for large scale data analysis, prediction and classification, especially in very data-rich environments ("big data"), and have started to find their way into medical applications. Here we will first give an overview of machine learning methods, with a focus on deep and recurrent neural networks, their relation to statistics, and the core principles behind them. We will then discuss and review directions along which (deep) neural networks can be, or already have been, applied in the context of psychiatry, and will try to delineate their future potential in this area. We will also comment on an emerging area that so far has been much less well explored: by embedding semantically interpretable computational models of brain dynamics or behavior into a statistical machine learning context, insights into dysfunction beyond mere prediction and classification may be gained. Especially this marriage of computational models with statistical inference may offer insights into neural and behavioral mechanisms that could open completely novel avenues for psychiatric treatment.

Narrative impairment, white matter damage and CSF biomarkers in the Alzheimer's disease spectrum

BACKGROUND

Narrative discourse (ND) refers to one's ability to verbally reproduce a sequence of temporally and logically-linked events. Impairments in ND may occur in subjects with Amnestic Mild Cognitive Impairment (aMCI) and Alzheimer's Disease (AD), but correlates across this function, neuroimaging and cerebrospinal fluid (CSF) AD biomarkers remain understudied.

OBJECTIVES

We sought to measure correlates among ND, Diffusion Tensor Imaging (DTI) indexes and AD CSF biomarkers in patients within the AD spectrum.

RESULTS

Groups differed in narrative production (NProd) and comprehension. aMCI and AD presented poorer inference abilities than controls. AD subjects were more impaired than controls and aMCI regarding WB (p<0.01). ROIs DTI assessment distinguished the three groups. Mean Diffusivity (MD) in the uncinate, bilateral parahippocampal cingulate and left inferior occipitofrontal fasciculi negatively correlated with NProd. Changes in specific tracts correlated with T-tau/Aβ1-42 ratio in CSF.

CONCLUSIONS

AD and aMCI patients presented more ND impairments than controls. Those findings were associated with changes in ventral language-associated and in the inferior parahippocampal pathways. The latest were correlated with biomarkers' levels in the CSF.

METHODS

AD (N=14), aMCI (N=31) and Control (N=39) groups were compared for whole brain (WB) and regions of interest (ROI) DTI parameters, ND and AD CSF biomarkers.

Early diagnosis of Alzheimer's disease using combined features from voxel-based morphometry and cortical, subcortical, and hippocampus regions of MRI T1 brain images

In recent years, several high-dimensional, accurate, and effective classification methods have been proposed for the automatic discrimination of the subject between Alzheimer's disease (AD) or its prodromal phase {i.e., mild cognitive impairment (MCI)} and healthy control (HC) persons based on T1-weighted structural magnetic resonance imaging (sMRI). These methods emphasis only on using the individual feature from sMRI images for the classification of AD, MCI, and HC subjects and their achieved classification accuracy is low. However, latest multimodal studies have shown that combining multiple features from different sMRI analysis techniques can improve the classification accuracy for these types of subjects. In this paper, we propose a novel classification technique that precisely distinguishes individuals with AD, aAD (stable MCI, who had not converted to AD within a 36-month time period), and mAD (MCI caused by AD, who had converted to AD within a 36-month time period) from HC individuals. The proposed method combines three different features extracted from structural MR (sMR) images using voxel-based morphometry (VBM), hippocampal volume (HV), and cortical and subcortical segmented region techniques. Three classification experiments were performed (AD vs. HC, aAD vs. mAD, and HC vs. mAD) with 326 subjects (171 elderly controls and 81 AD, 35 aAD, and 39 mAD patients). For the development and validation of the proposed classification method, we acquired the sMR images from the dataset of the National Research Center for Dementia (NRCD). A five-fold cross-validation technique was applied to find the optimal hyperparameters for the classifier, and the classification performance was compared by using three well-known classifiers: K-nearest neighbor, support vector machine, and random forest. Overall, the proposed model with the SVM classifier achieved the best performance on the NRCD dataset. For the individual feature, the VBM technique provided the best results followed by the HV technique. However, the use of combined features improved the classification accuracy and predictive power for the early classification of AD compared to the use of individual features. The most stable and reliable classification results were achieved when combining all extracted features. Additionally, to analyze the efficiency of the proposed model, we used the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset to compare the classification performance of the proposed model with those of several state-of-the-art methods.

Deep Learning in Alzheimer's Disease: Diagnostic Classification and Prognostic Prediction Using Neuroimaging Data

Deep learning, a state-of-the-art machine learning approach, has shown outstanding performance over traditional machine learning in identifying intricate structures in complex high-dimensional data, especially in the domain of computer vision. The application of deep learning to early detection and automated classification of Alzheimer's disease (AD) has recently gained considerable attention, as rapid progress in neuroimaging techniques has generated large-scale multimodal neuroimaging data. A systematic review of publications using deep learning approaches and neuroimaging data for diagnostic classification of AD was performed. A PubMed and Google Scholar search was used to identify deep learning papers on AD published between January 2013 and July 2018. These papers were reviewed, evaluated, and classified by algorithm and neuroimaging type, and the findings were summarized. Of 16 studies meeting full inclusion criteria, 4 used a combination of deep learning and traditional machine learning approaches, and 12 used only deep learning approaches. The combination of traditional machine learning for classification and stacked auto-encoder (SAE) for feature selection produced accuracies of up to 98.8% for AD classification and 83.7% for prediction of conversion from mild cognitive impairment (MCI), a prodromal stage of AD, to AD. Deep learning approaches, such as convolutional neural network (CNN) or recurrent neural network (RNN), that use neuroimaging data without pre-processing for feature selection have yielded accuracies of up to 96.0% for AD classification and 84.2% for MCI conversion prediction. The best classification performance was obtained when multimodal neuroimaging and fluid biomarkers were combined. Deep learning approaches continue to improve in performance and appear to hold promise for diagnostic classification of AD using multimodal neuroimaging data. AD research that uses deep learning is still evolving, improving performance by incorporating additional hybrid data types, such as-omics data, increasing transparency with explainable approaches that add knowledge of specific disease-related features and mechanisms.

Bimodal learning via trilogy of skip-connection deep networks for diabetic retinopathy risk progression identification

BACKGROUND

Diabetic Retinopathy (DR) is considered a pathology of retinal vascular complications, which stays in the top causes of vision impairment and blindness. Therefore, precisely inspecting its progression enables the ophthalmologists to set up appropriate next-visit schedule and cost-effective treatment plans. In the literature, existing work only makes use of numerical attributes in Electronic Medical Records (EMR) for acquiring such kind of DR-oriented knowledge through conventional machine learning techniques, which require an exhaustive job of engineering most impactful risk factors.

OBJECTIVE

In this paper, an approach of deep bimodal learning is introduced to leverage the performance of DR risk progression identification.

METHODS

In particular, we further involve valuable clinical information of fundus photography in addition to the aforementioned systemic attributes. Accordingly, a Trilogy of Skip-connection Deep Networks, namely Tri-SDN, is proposed to exhaustively exploit underlying relationships between the baseline and follow-up information of the fundus images and EMR-based attributes. Besides that, we adopt Skip-Connection Blocks as basis components of the Tri-SDN for making the end-to-end flow of signals more efficient during feedforward and backpropagation processes.

RESULTS

Through a 10-fold cross validation strategy on a private dataset of 96 diabetic mellitus patients, the proposed method attains superior performance over the conventional EMR-modality learning approach in terms of Accuracy (90.6%), Sensitivity (96.5%), Precision (88.7%), Specificity (82.1%), and Area Under Receiver Operating Characteristics (88.8%).

CONCLUSIONS

The experimental results show that the proposed Tri-SDN can combine features of different modalities (i.e., fundus images and EMR-based numerical risk factors) smoothly and effectively during training and testing processes, respectively. As a consequence, with impressive performance of DR risk progression recognition, the proposed approach is able to help the ophthalmologists properly decide follow-up schedule and subsequent treatment plans.

Layer-Wise Relevance Propagation for Explaining Deep Neural Network Decisions in MRI-Based Alzheimer's Disease Classification

Deep neural networks have led to state-of-the-art results in many medical imaging tasks including Alzheimer's disease (AD) detection based on structural magnetic resonance imaging (MRI) data. However, the network decisions are often perceived as being highly non-transparent, making it difficult to apply these algorithms in clinical routine. In this study, we propose using layer-wise relevance propagation (LRP) to visualize convolutional neural network decisions for AD based on MRI data. Similarly to other visualization methods, LRP produces a heatmap in the input space indicating the importance/relevance of each voxel contributing to the final classification outcome. In contrast to susceptibility maps produced by guided backpropagation ("Which change in voxels would change the outcome most?"), the LRP method is able to directly highlight positive contributions to the network classification in the input space. In particular, we show that (1) the LRP method is very specific for individuals ("Why does this person have AD?") with high inter-patient variability, (2) there is very little relevance for AD in healthy controls and (3) areas that exhibit a lot of relevance correlate well with what is known from literature. To quantify the latter, we compute size-corrected metrics of the summed relevance per brain area, e.g., relevance density or relevance gain. Although these metrics produce very individual "fingerprints" of relevance patterns for AD patients, a lot of importance is put on areas in the temporal lobe including the hippocampus. After discussing several limitations such as sensitivity toward the underlying model and computation parameters, we conclude that LRP might have a high potential to assist clinicians in explaining neural network decisions for diagnosing AD (and potentially other diseases) based on structural MRI data.

A Systematic Review and Implementation of IoT-Based Pervasive Sensor-Enabled Tracking System for Dementia Patients

In today's world, 46.8 million people suffer from brain related diseases. Dementia is most prevalent of all. In general scenario, a dementia patient lacks proper guidance in searching out the way to return back at his/her home. Thus, increasing the risk of getting damaged at individual-health level. Therefore, it is important to track their movement in more sophisticated manner as possible. With emergence of wearables, GPS sensors and Internet of Things (IoT), such devices have become available in public domain. Smartphone apps support caregiver to locate the dementia patients in real-time. RF, GSM, 3G, Wi-Fi and 4G technology fill the communication gap between patient and caregiver to bring them closer. In this paper, we incorporated 7 most popular wearables for investigation to seek appropriateness for dementia tracking in recent times in systematic manners. We performed an in-depth review of these wearables as per the cost, technology wise and application wise characteristics. A case novel study i.e. IoT-based Force Sensor Resistance enabled System-FSRIoT, has been proposed and implemented to validate the effectiveness of IoT in the domain of smarter dementia patient tracking in wearable form factor. The results show promising aspect of a whole new notion to leverage efficient assistive physio-medical healthcare to the dementia patients and the affected family members to reduce life risks and achieve a better social life.

MildInt: Deep Learning-Based Multimodal Longitudinal Data Integration Framework

As large amounts of heterogeneous biomedical data become available, numerous methods for integrating such datasets have been developed to extract complementary knowledge from multiple domains of sources. Recently, a deep learning approach has shown promising results in a variety of research areas. However, applying the deep learning approach requires expertise for constructing a deep architecture that can take multimodal longitudinal data. Thus, in this paper, a deep learning-based python package for data integration is developed. The python package deep learning-based multimodal longitudinal data integration framework (MildInt) provides the preconstructed deep learning architecture for a classification task. MildInt contains two learning phases: learning feature representation from each modality of data and training a classifier for the final decision. Adopting deep architecture in the first phase leads to learning more task-relevant feature representation than a linear model. In the second phase, linear regression classifier is used for detecting and investigating biomarkers from multimodal data. Thus, by combining the linear model and the deep learning model, higher accuracy and better interpretability can be achieved. We validated the performance of our package using simulation data and real data. For the real data, as a pilot study, we used clinical and multimodal neuroimaging datasets in Alzheimer's disease to predict the disease progression. MildInt is capable of integrating multiple forms of numerical data including time series and non-time series data for extracting complementary features from the multimodal dataset.

Cognitive impairment of workers in a large-scale aluminium factory in China: a cross-sectional study

OBJECTIVES

To investigate the prevalence of mild cognitive impairment and the relationship with plasma aluminium among aluminium workers.

DESIGN

This was a cross-sectional case-control study in the SH Aluminium Factory, China.

SETTING

The university and affiliated hospital cooperated in the study.

PARTICIPANTS

There were 910 aluminium workers on duty, among whom 853 participated in our study. Participants, such as those with cerebral vascular disease, epilepsy, brain trauma, Parkinson's and mental diseases, aluminium-containing drug and mental drug use, and any family history of dementia in first-degree relatives were excluded.

PRIMARY AND SECONDARY OUTCOME MEASURES

Blood samples were collected, and plasma aluminium was measured by inductively coupled plasma-mass spectrometry. For each case, four age-matched controls were evaluated to determine the relationship between aluminium exposure and mild cognitive impairment. Conditional logistic regression was used to explore influential factors in mild cognitive impairment.

RESULTS

Among 910 workers, 93.74% participated in stage 1; 53 cases were finally diagnosed. The crude prevalence of mild cognitive impairment among aluminium workers on duty was 6.21%. There was a significant difference in plasma aluminium concentration between the two groups. In the multivariate analysis, we found that a higher level of plasma aluminium was associated with a high risk of cognitive impairment when compared with a lower aluminium level (AOR=2.24, 95% CI=1.17 to 4.26), and a high education level was a protective factor (AOR=0.36, 95% CI=0.18 to 0.70). No other factor was statistically significant.

CONCLUSIONS

Mild cognitive impairment is no longer a disease specific to elderly people. High plasma aluminium exposure might be associated with an increased risk of cognitive impairment, but a reduced risk was observed with a high education level. The cognitive function of aluminium workers on duty must be considered seriously.

Interpretable classification of Alzheimer's disease pathologies with a convolutional neural network pipeline

Neuropathologists assess vast brain areas to identify diverse and subtly-differentiated morphologies. Standard semi-quantitative scoring approaches, however, are coarse-grained and lack precise neuroanatomic localization. We report a proof-of-concept deep learning pipeline that identifies specific neuropathologies-amyloid plaques and cerebral amyloid angiopathy-in immunohistochemically-stained archival slides. Using automated segmentation of stained objects and a cloud-based interface, we annotate > 70,000 plaque candidates from 43 whole slide images (WSIs) to train and evaluate convolutional neural networks. Networks achieve strong plaque classification on a 10-WSI hold-out set (0.993 and 0.743 areas under the receiver operating characteristic and precision recall curve, respectively). Prediction confidence maps visualize morphology distributions at high resolution. Resulting network-derived amyloid beta (Aβ)-burden scores correlate well with established semi-quantitative scores on a 30-WSI blinded hold-out. Finally, saliency mapping demonstrates that networks learn patterns agreeing with accepted pathologic features. This scalable means to augment a neuropathologist's ability suggests a route to neuropathologic deep phenotyping.

A highly predictive signature of cognition and brain atrophy for progression to Alzheimer's dementia

BACKGROUND

Clinical trials in Alzheimer's disease need to enroll patients whose cognition will decline over time, if left untreated, in order to demonstrate the efficacy of an intervention. Machine learning models used to screen for patients at risk of progression to dementia should therefore favor specificity (detecting only progressors) over sensitivity (detecting all progressors), especially when the prevalence of progressors is low. Here, we explore whether such high-risk patients can be identified using cognitive assessments and structural neuroimaging by training machine learning tools in a high-specificity regime.

RESULTS

A multimodal signature of Alzheimer's dementia was first extracted from the ADNI1 dataset. We then validated the predictive value of this signature on ADNI1 patients with mild cognitive impairment (N = 235). The signature was optimized to predict progression to dementia over 3 years with low sensitivity (55.1%) but high specificity (95.6%), resulting in only moderate accuracy (69.3%) but high positive predictive value (80.4%, adjusted for a "typical" 33% prevalence rate of true progressors). These results were replicated in ADNI2 (N = 235), with 87.8% adjusted positive predictive value (96.7% specificity, 47.3% sensitivity, 85.1% accuracy).

CONCLUSIONS

We found that cognitive measures alone could identify high-risk individuals, with structural measurements providing a slight improvement. The signature had comparable receiver operating characteristics to standard machine learning tools, yet a marked improvement in positive predictive value was achieved over the literature by selecting a high-specificity operating point. The multimodal signature can be readily applied for the enrichment of clinical trials.

A supervised machine learning approach to characterize spinal network function

Spontaneous activity is a common feature of immature neuronal networks throughout the central nervous system and plays an important role in network development and consolidation. In postnatal rodents, spontaneous activity in the spinal cord exhibits complex, stochastic patterns that have historically proven challenging to characterize. We developed a software tool for quickly and automatically characterizing and classifying episodes of spontaneous activity generated from developing spinal networks. We recorded spontaneous activity from in vitro lumbar ventral roots of 16 neonatal [postnatal day (P)0-P3] mice. Recordings were DC coupled and detrended, and episodes were separated for analysis. Amplitude-, duration-, and frequency-related features were extracted from each episode and organized into five classes. Paired classes and features were used to train and test supervised machine learning algorithms. Multilayer perceptrons were used to classify episodes as rhythmic or multiburst. We increased network excitability with potassium chloride and tested the utility of the tool to detect changes in features and episode class. We also demonstrate usability by having a novel experimenter use the program to classify episodes collected at a later time point (P5). Supervised machine learning-based classification of episodes accounted for changes that traditional approaches cannot detect. Our tool, named SpontaneousClassification, advances the detail in which we can study not only developing spinal networks, but also spontaneous networks in other areas of the nervous system. NEW & NOTEWORTHY Spontaneous activity is important for nervous system network development and consolidation. Our software uses machine learning to automatically and quickly characterize and classify episodes of spontaneous activity in the spinal cord of newborn mice. It detected changes in network activity following KCl-enhanced excitation. Using our software to classify spontaneous activity throughout development, in pathological models, or with neuromodulation, may offer insight into the development and organization of spinal circuits.

An interpretable machine learning model for diagnosis of Alzheimer's disease

We present an interpretable machine learning model for medical diagnosis called sparse high-order interaction model with rejection option (SHIMR). A decision tree explains to a patient the diagnosis with a long rule (i.e., conjunction of many intervals), while SHIMR employs a weighted sum of short rules. Using proteomics data of 151 subjects in the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset, SHIMR is shown to be as accurate as other non-interpretable methods (Sensitivity, SN = 0.84 ± 0.1, Specificity, SP = 0.69 ± 0.15 and Area Under the Curve, AUC = 0.86 ± 0.09). For clinical usage, SHIMR has a function to abstain from making any diagnosis when it is not confident enough, so that a medical doctor can choose more accurate but invasive and/or more costly pathologies. The incorporation of a rejection option complements SHIMR in designing a multistage cost-effective diagnosis framework. Using a baseline concentration of cerebrospinal fluid (CSF) and plasma proteins from a common cohort of 141 subjects, SHIMR is shown to be effective in designing a patient-specific cost-effective Alzheimer's disease (AD) pathology. Thus, interpretability, reliability and having the potential to design a patient-specific multistage cost-effective diagnosis framework can make SHIMR serve as an indispensable tool in the era of precision medicine that can cater to the demand of both doctors and patients, and reduce the overwhelming financial burden of medical diagnosis.

Predicting Alzheimer's disease progression using multi-modal deep learning approach

Alzheimer's disease (AD) is a progressive neurodegenerative condition marked by a decline in cognitive functions with no validated disease modifying treatment. It is critical for timely treatment to detect AD in its earlier stage before clinical manifestation. Mild cognitive impairment (MCI) is an intermediate stage between cognitively normal older adults and AD. To predict conversion from MCI to probable AD, we applied a deep learning approach, multimodal recurrent neural network. We developed an integrative framework that combines not only cross-sectional neuroimaging biomarkers at baseline but also longitudinal cerebrospinal fluid (CSF) and cognitive performance biomarkers obtained from the Alzheimer's Disease Neuroimaging Initiative cohort (ADNI). The proposed framework integrated longitudinal multi-domain data. Our results showed that 1) our prediction model for MCI conversion to AD yielded up to 75% accuracy (area under the curve (AUC) = 0.83) when using only single modality of data separately; and 2) our prediction model achieved the best performance with 81% accuracy (AUC = 0.86) when incorporating longitudinal multi-domain data. A multi-modal deep learning approach has potential to identify persons at risk of developing AD who might benefit most from a clinical trial or as a stratification approach within clinical trials.

Multivariate Deep Learning Classification of Alzheimer's Disease Based on Hierarchical Partner Matching Independent Component Analysis

Machine learning and pattern recognition have been widely investigated in order to look for the biomarkers of Alzheimer’s disease (AD). However, most existing methods extract features by seed-based correlation, which not only requires prior information but also ignores the relationship between resting state functional magnetic resonance imaging (rs-fMRI) voxels. In this study, we proposed a deep learning classification framework with multivariate data-driven based feature extraction for automatic diagnosis of AD. Specifically, a three-level hierarchical partner matching independent components analysis (3LHPM-ICA) approach was proposed first in order to address the issues in spatial individual ICA, including the uncertainty of the numbers of components, the randomness of initial values, and the correspondence of ICs of multiple subjects, resulting in stable and reliable ICs which were applied as the intrinsic brain functional connectivity (FC) features. Second, Granger causality (GC) was utilized to infer directional interaction between the ICs that were identified by the 3LHPM-ICA method and extract the effective connectivity features. Finally, a deep learning classification framework was developed to distinguish AD from controls by fusing the functional and effective connectivities. A resting state fMRI dataset containing 34 AD patients and 34 normal controls (NCs) was applied to the multivariate deep learning platform, leading to a classification accuracy of 95.59%, with a sensitivity of 97.06% and a specificity of 94.12% with leave-one-out cross validation (LOOCV). The experimental results demonstrated that the measures of neural connectivities of ICA and GC followed by deep learning classification represented the most powerful methods of distinguishing AD clinical data from NCs, and these aberrant brain connectivities might serve as robust brain biomarkers for AD. This approach also allows for expansion of the methodology to classify other psychiatric disorders.

An overview of deep learning in medical imaging focusing on MRI

What has happened in machine learning lately, and what does it mean for the future of medical image analysis? Machine learning has witnessed a tremendous amount of attention over the last few years. The current boom started around 2009 when so-called deep artificial neural networks began outperforming other established models on a number of important benchmarks. Deep neural networks are now the state-of-the-art machine learning models across a variety of areas, from image analysis to natural language processing, and widely deployed in academia and industry. These developments have a huge potential for medical imaging technology, medical data analysis, medical diagnostics and healthcare in general, slowly being realized. We provide a short overview of recent advances and some associated challenges in machine learning applied to medical image processing and image analysis. As this has become a very broad and fast expanding field we will not survey the entire landscape of applications, but put particular focus on deep learning in MRI. Our aim is threefold: (i) give a brief introduction to deep learning with pointers to core references; (ii) indicate how deep learning has been applied to the entire MRI processing chain, from acquisition to image retrieval, from segmentation to disease prediction; (iii) provide a starting point for people interested in experimenting and perhaps contributing to the field of deep learning for medical imaging by pointing out good educational resources, state-of-the-art open-source code, and interesting sources of data and problems related medical imaging.

Neuroimaging and Machine Learning for Dementia Diagnosis: Recent Advancements and Future Prospects

Dementia, a chronic and progressive cognitive declination of brain function caused by disease or impairment, is becoming more prevalent due to the aging population. A major challenge in dementia is achieving accurate and timely diagnosis. In recent years, neuroimaging with computer-aided algorithms have made remarkable advances in addressing this challenge. The success of these approaches is mostly attributed to the application of machine learning techniques for neuroimaging. In this review paper, we present a comprehensive survey of automated diagnostic approaches for dementia using medical image analysis and machine learning algorithms published in the recent years. Based on the rigorous review of the existing works, we have found that, while most of the studies focused on Alzheimer's disease, recent research has demonstrated reasonable performance in the identification of other types of dementia remains a major challenge. Multimodal imaging analysis deep learning approaches have shown promising results in the diagnosis of these other types of dementia. The main contributions of this review paper are as follows. 1) Based on the detailed analysis of the existing literature, this paper discusses neuroimaging procedures for dementia diagnosis. 2) It systematically explains the most recent machine learning techniques and, in particular, deep learning approaches for early detection of dementia.

Convolutional Neural Networks-Based MRI Image Analysis for the Alzheimer's Disease Prediction From Mild Cognitive Impairment

Mild cognitive impairment (MCI) is the prodromal stage of Alzheimer's disease (AD). Identifying MCI subjects who are at high risk of converting to AD is crucial for effective treatments. In this study, a deep learning approach based on convolutional neural networks (CNN), is designed to accurately predict MCI-to-AD conversion with magnetic resonance imaging (MRI) data. First, MRI images are prepared with age-correction and other processing. Second, local patches, which are assembled into 2.5 dimensions, are extracted from these images. Then, the patches from AD and normal controls (NC) are used to train a CNN to identify deep learning features of MCI subjects. After that, structural brain image features are mined with FreeSurfer to assist CNN. Finally, both types of features are fed into an extreme learning machine classifier to predict the AD conversion. The proposed approach is validated on the standardized MRI datasets from the Alzheimer's Disease Neuroimaging Initiative (ADNI) project. This approach achieves an accuracy of 79.9% and an area under the receiver operating characteristic curve (AUC) of 86.1% in leave-one-out cross validations. Compared with other state-of-the-art methods, the proposed one outperforms others with higher accuracy and AUC, while keeping a good balance between the sensitivity and specificity. Results demonstrate great potentials of the proposed CNN-based approach for the prediction of MCI-to-AD conversion with solely MRI data. Age correction and assisted structural brain image features can boost the prediction performance of CNN.

Automated detection of lung cancer at ultralow dose PET/CT by deep neural networks - Initial results

OBJECTIVES

We evaluated whether machine learning may be helpful for the detection of lung cancer in FDG-PET imaging in the setting of ultralow dose PET scans.

MATERIALS AND METHODS

We studied the performance of an artificial neural network discriminating lung cancer patients (n = 50) from controls (n = 50) without pulmonary malignancies. A total of 3936 PET slices including images in which the lung tumor is visually present and image slices of patients with no lung cancer were exported. The diagnostic performance of the artificial neural network based on clinical standard dose PET images (PET_100%) as well as with a tenfold (PET_10%) and thirtyfold (PET_3.3%) reduced radiation dose (∼0.11 mSv) was assessed.

RESULTS

The area under the curve of the deep learning algorithm for lung cancer detection was 0.989, 0.983 and 0.970 for standard dose images (PET_100%), and reduced dose PET_10%, and PET_3.3% reconstruction, respectively. The artificial neural network achieved a sensitivity of 95.9% and 91.5% and a specificity of 98.1% and 94.2%, at standard dose and ultralow dose PET_3.3%, respectively.

CONCLUSION

Our results suggest that machine learning algorithms may aid fully automated lung cancer detection even at very low effective radiation doses of 0.11 mSv. Further improvement of this technology might improve the specificity of lung cancer screening efforts and could lead to new applications of FDG-PET.

Reproducible evaluation of classification methods in Alzheimer's disease: Framework and application to MRI and PET data

A large number of papers have introduced novel machine learning and feature extraction methods for automatic classification of Alzheimer's disease (AD). However, while the vast majority of these works use the public dataset ADNI for evaluation, they are difficult to reproduce because different key components of the validation are often not readily available. These components include selected participants and input data, image preprocessing and cross-validation procedures. The performance of the different approaches is also difficult to compare objectively. In particular, it is often difficult to assess which part of the method (e.g. preprocessing, feature extraction or classification algorithms) provides a real improvement, if any. In the present paper, we propose a framework for reproducible and objective classification experiments in AD using three publicly available datasets (ADNI, AIBL and OASIS). The framework comprises: i) automatic conversion of the three datasets into a standard format (BIDS); ii) a modular set of preprocessing pipelines, feature extraction and classification methods, together with an evaluation framework, that provide a baseline for benchmarking the different components. We demonstrate the use of the framework for a large-scale evaluation on 1960 participants using T1 MRI and FDG PET data. In this evaluation, we assess the influence of different modalities, preprocessing, feature types (regional or voxel-based features), classifiers, training set sizes and datasets. Performances were in line with the state-of-the-art. FDG PET outperformed T1 MRI for all classification tasks. No difference in performance was found for the use of different atlases, image smoothing, partial volume correction of FDG PET images, or feature type. Linear SVM and L2-logistic regression resulted in similar performance and both outperformed random forests. The classification performance increased along with the number of subjects used for training. Classifiers trained on ADNI generalized well to AIBL and OASIS. All the code of the framework and the experiments is publicly available: general-purpose tools have been integrated into the Clinica software (www.clinica.run) and the paper-specific code is available at: https://gitlab.icm-institute.org/aramislab/AD-ML.

Brain MRI analysis for Alzheimer's disease diagnosis using an ensemble system of deep convolutional neural networks

Alzheimer's disease is an incurable, progressive neurological brain disorder. Earlier detection of Alzheimer's disease can help with proper treatment and prevent brain tissue damage. Several statistical and machine learning models have been exploited by researchers for Alzheimer's disease diagnosis. Analyzing magnetic resonance imaging (MRI) is a common practice for Alzheimer's disease diagnosis in clinical research. Detection of Alzheimer's disease is exacting due to the similarity in Alzheimer's disease MRI data and standard healthy MRI data of older people. Recently, advanced deep learning techniques have successfully demonstrated human-level performance in numerous fields including medical image analysis. We propose a deep convolutional neural network for Alzheimer's disease diagnosis using brain MRI data analysis. While most of the existing approaches perform binary classification, our model can identify different stages of Alzheimer's disease and obtains superior performance for early-stage diagnosis. We conducted ample experiments to demonstrate that our proposed model outperformed comparative baselines on the Open Access Series of Imaging Studies dataset.