Deep Learning Framework For Classification Research Articles

A deep learning classification framework for research methods of marine protected area management

The latest emerging transdisciplinary marine protected area (MPA) research scheme requires efficient approaches for theoretically based and data-driven method integration. However, due to the rapid development and diversification of research methods, it is growingly difficult to locate new methods in methodological dimensions and integrate them to the utmost utility. This study proposes a deep learning-based classification framework for MPA management methods focused particularly on data and theory capabilities using natural language processing (NLP). It extracted keywords from academic sources and performed clustering based on semantic similarity, generating benchmark texts for abstract labeling. By training the deep learning NLP model and analyzing the abstracts of 9049 MPA management empirical research articles from 1986 to 2024, the data and theory scores were attributed to each article, and a total of 19 major method categories and 110 segment branches were identified in qualitative, quantitative, and mixed genres. Combination types of research methods were summarized, yielding the data-theory neutralization principle where the average data and theory scores tend to approximate 0.50. Applying the principle broadens traditional boundaries for method integration and extends method synthesis to higher numbers, generating a practical research 2paradigm for future MPA research. Implications include bridging social and ecological data, theorizing emergent challenges in complex systems and integrating theory construction and data science. The framework is applicable to quantification of other environmental management disciplines and can serve as guidance for multidisciplinary method integration.© 2017 Elsevier Inc. All rights reserved.

Brain magnetic resonance imaging image classification for Alzheimer's disease and its hardware acceleration

<span lang="EN-US">Alzheimer's is a progressive neurodegenerative disorder and is considered the sixth leading cause of death after cancer and heart attack. Early detection and diagnosis provide individuals to go through a wider variety of clinical trials and get multiple medical benefits. Research on the application of deep learning and machine learning to the early detection of Alzheimer's disease has recently gained considerable attention. In this paper, we propose a deep learning classification framework to classify the individual with different progression stages of Alzheimer's disease such as mild cognitive impairment (MCI) and cognitive normal (CN). The dataset from Alzheimer’s disease neuroimaging initiative (ADNI) is considered in this paper which is a multisite having collection of Neuroimaging data for researchers. Structural magnetic resonance imaging (MRI) images are considered from the ADNI data set and feature extraction is done using a 2D discrete wavelet transform. 97% of data reduction is achieved during data pre-processing. The algorithm is trained and validated. The algorithm is accelerated in Nvidia Tx2 graphics processing unit (GPU) to get the better throughput. The result shows our algorithm outperforms the other deep learning algorithms with 91.56% accuracy. </span>

A Deep Learning Framework for Classification of Immature White Blood Cells in Acute Myeloid Leukemia Using Convolutional Autoencoder and Transfer-Learning Ensemble

Acute myeloid leukemia (AML) is a type of acute leukemia that affects myeloid cells. It is characterized by uncontrolled proliferation of immature WBCs causing an accumulation of immature WBCs in the bone marrow and peripheral circulation. AML is an aggressive cancer that can be life-threatening if not treated promptly, therefore, early detection is crucial. Identification of immature WBCs and their subtypes is the first step towards AML diagnosis. Microscopic blood smear image analysis is commonly used to diagnose AML as it is less expensive non-invasive diagnostic tool compared to bone marrow biopsy and immunophenotype. Nonetheless, classifying immature WBCs presents challenges due to their high similarity and minimal interclass variation, especially for intermediate stages of myelopoiesis which make is susceptible to misclassification. The current diagnostic methods of AML are based on manual identification of immature WBCs using peripheral blood smear. These methos are time-consuming, prone to errors, and subject to inter-observer variation. Therefore, this study aims to develop a computer-aided diagnostic framework to accurately identify immature WBCs and classify them into various subtypes. The proposed framework consists of two main components: a binary classification system to classify cells into mature versus immature cells and a multiclassification system to further classify immature cells into four subtypes including myeloblast, monoblast, and promyelocytes. The proposed framework was designed to undergo four distinct phases: the first phase involves preprocessing, which includes data augmentation techniques aimed at addressing the imbalance distribution of WBCs. Augmentation encompasses geometric transformation and generation of synthetic images using convolutional Autoencoder (CAE). Feature extraction phase involves using image embedding and transfer-learning. Image embedding representation was obtained using CAE and then utilized by two pre-trained models: VGG19 and Resnet50. Classification phase was carried out using weighted ensemble of the VGG19 and Resnet50 with the optimal weights determined through Grid search. The validation phase evaluated the model performance using overall accuracy, precision, and sensitivity. The area under the ROC curve (AUC) was used to assess model discrimination ability. The proposed framework demonstrated excellent results, achieving an average accuracy of 99.5%, a sensitivity of 97.97%, and a precision of 94.12%. The model exhibited an excellent discriminatory capability, with an AUC of 99.9%. Furthermore, the proposed framework outperformed previous methods, showcasing improved performance. Key word: Acute myeloid leukemia, peripheral blood smear, immature white blood cells, deep learning, feature extraction, computer-aided diagnosis.

A Convolutional Neural Network-based gradient boosting framework for prediction of the band gap of photo-active catalysts

A recent trend in chemical synthesis is photo-catalysis, which uses photo-active catalyst materials that are semiconductor materials. A well-known electronic property of semiconducting materials is the band gap. A photo-catalyst’s desired band gap range is between 1.5 eV and 6.2 eV. A rational design and synthesis of photo-active catalysts require knowledge of the band gap as an initial screening parameter. Herein, we propose an integrated deep learning-based framework to classify the photo-active catalysts and predict their band gap using compositional features. To this extent, we have utilized the dataset extracted from the “catalyst hub” site by web scraping with the help of a Python script. Extensive data cleaning and pre-processing are done to make input data amenable for training the models. Also, more valuable features are made using two methods: (a) one hot-encoding and (b) calculating the mean of the embeddings of catalysts computed by Mat2Vec, a pre-trained transformer-based model. With the help of this generated feature set, we have proposed a two-stage deep-learning framework for classification and regression tasks. In the first stage, a 2D-Convolutional Neural Net (CNN)-based classifier is used to classify whether a catalyst belongs to the photo-active catalyst class. After the first stage screening, in the second stage, we use a 1D-VGG-based gradient boosting framework to predict the band gap of the photo-active catalyst only using compositional features as inputs. 2D-CNN for the classification task has an accuracy of 0.903 and 0.886 for the train and test datasets, respectively. Further, the proposed integrated model that uses 1D-Convolutional layers of VGG followed by the XGBoostRegressor has a test R2 of 0.750, much higher than baseline models reported in the literature.

An Integrated Deep Learning Framework for Classification of Mineral Thin Sections and Other Geo-Data, a Tutorial

Recent studies have demonstrated the potential of machine learning methods for fast and accurate mineral classification based on microscope thin sections. Such methods can be extremely useful to support geoscientists during the phases of operational geology, especially when mineralogical and petrological data are fully integrated with other geological and geophysical information. In order to be effective, these methods require robust machine learning models trained on pre-labeled data. Furthermore, it is mandatory to optimize the hyper-parameters of the machine learning techniques in order to guarantee optimal classification accuracy and reliability. Nowadays, deep learning algorithms are widely applied for image analysis and automatic classification in a large range of Earth disciplines, including mineralogy, petrography, paleontology, well-log analysis, geophysical imaging, and so forth. The main reason for the recognized effectiveness of deep learning algorithms for image analysis is that they are able to quickly learn complex representations of images and patterns within them. Differently from traditional image-processing techniques based on handcrafted features, deep learning models automatically learn and extract features from the data, capturing, in almost real-time, complex relationships and patterns that are difficult to manually define. Many different types of deep learning models can be used for image analysis and classification, including fully connected deep neural networks (FCNNs), convolutional neural networks (CNNs or ConvNet), and residual networks (ResNets). In this paper, we compare some of these techniques and verify their effectiveness on the same dataset of mineralogical thin sections. We show that the different deep learning methods are all effective techniques in recognizing and classifying mineral images directly in the field, with ResNets outperforming the other techniques in terms of accuracy and precision. In addition, we compare the performance of deep learning techniques with different machine learning algorithms, including random forest, naive Bayes, adaptive boosting, support vector machine, and decision tree. Using quantitative performance indexes as well as confusion matrixes, we demonstrate that deep neural networks show generally better classification performances than the other approaches. Furthermore, we briefly discuss how to expand the same workflow to other types of images and geo-data, showing how this deep learning approach can be generalized to a multiscale/multipurpose methodology addressed to the analysis and automatic classification of multidisciplinary information. This article has tutorial purposes, too. For that reason, we will explain, with a didactical level of detail, all the key steps of the workflow.

Deep Learning Framework for Classification of Mental Stress from Multimodal Datasets

Deep Learning Framework for Classification of Mental Stress from Multimodal Datasets

A Deep-Learning Classification Framework for Reducing Communication Errors in Dynamic Hand Signaling for Crane Operation

Crane operators and signalmen play an integral role in the safe and efficient operation of cranes on a construction site. Operating a crane is a complex and demanding task that requires careful coordination between operator and signalmen in order to avoid errors that could have dire consequences, including serious injury or loss of life. Therefore, special considerations should be taken to mitigate communication errors that could occur between the two parties. Technology can play an important role in enhancing communication, and, with recent advancements in technology, human–computer interaction has emerged as an active area of research within the field of computer vision. This paper presents a framework that integrates the YOLOv4 model (for object detection) and the long short-term memory (LSTM) model (a recurrent neural network) for dynamic hand signal classification in real time. The first step is the creation of a crane signalman dynamic hand signal data set with 18 classes. The YOLOv4 model is then customized for this application by modifying the activation function. Three modified YOLOv4 models are then integrated with the LSTM model. The modified YOLOv4 integrated with LSTM is found to achieve a maximum overall accuracy of 94.8% with an inference time of 55.1 frames per second. The model is further validated with real-time dynamic hand signal classification, achieving an accuracy of 93.5% and an inference time of 44 frames per second. The proposed models show improved quality in classification accuracy as well as in processing speed in comparison to some of the most widely used models currently in use. The proposed novel framework can be used as another layer of communication to supplement current practice and reduce communication errors between crane signalmen and crane operators.

Deep Learning Multi-Class Approach for Human Fall Detection Based on Doppler Signatures.

Falling events are a global health concern with short- and long-term physical and psychological implications, especially for the elderly population. This work aims to monitor human activity in an indoor environment and recognize falling events without requiring users to carry a device or sensor on their bodies. A sensing platform based on the transmission of a continuous wave (CW) radio-frequency (RF) probe signal was developed using general-purpose equipment. The CW probe signal is similar to the pilot subcarriers transmitted by commercial off-the-shelf WiFi devices. As a result, our methodology can easily be integrated into a joint radio sensing and communication scheme. The sensing process is carried out by analyzing the changes in phase, amplitude, and frequency that the probe signal suffers when it is reflected or scattered by static and moving bodies. These features are commonly extracted from the channel state information (CSI) of WiFi signals. However, CSI relies on complex data acquisition and channel estimation processes. Doppler radars have also been used to monitor human activity. While effective, a radar-based fall detection system requires dedicated hardware. In this paper, we follow an alternative method to characterize falling events on the basis of the Doppler signatures imprinted on the CW probe signal by a falling person. A multi-class deep learning framework for classification was conceived to differentiate falling events from other activities that can be performed in indoor environments. Two neural network models were implemented. The first is based on a long-short-term memory network (LSTM) and the second on a convolutional neural network (CNN). A series of experiments comprising 11 subjects were conducted to collect empirical data and test the system's performance. Falls were detected with an accuracy of 92.1% for the LSTM case, while for the CNN, an accuracy rate of 92.1% was obtained. The results demonstrate the viability of human fall detection based on a radio sensing system such as the one described in this paper.

Deep learning framework for classification of COVID-19 from Chest X-rays

The Coronavirus Disease 2019 (COVID-19) has emerged as the most pressing health concern in recent years. In addition, a wide variety of severe acute respiratory syndrome-related COVID-19 strains are now in the process of developing (SARS). Therefore, the identification of all of these variations utilising a Real-time polymerase chain reaction (RT-PCR) test is a challenging endeavour that takes a significant amount of time. Due of the intricate structure of the chest x-ray (CXR) picture, traditional approaches were unable to correctly categorise the COVID-19 at an early stage. As a result, the primary emphasis of this paper is placed on the use of an artificial intelligence strategy that is based on deep learning convolutional neural networks (DLCNN) for the classification of COVID-19 illness. In the beginning, the hybrid features are extracted from the CXR dataset by using stacked auto ensemble (SAE), a technique that retrieved features by partitioning data into various subsets. This approach was used. After that, a DLCNN model is trained with these characteristics in order to classify COVID-19 for each test CXR picture. According to the findings of the simulations, the suggested classification led to superior results in terms of both subjectivity and objectivity in comparison to both traditional machine learning and deep learning approaches.

Towards Classification of Architectural Styles of Chinese Traditional Settlements Using Deep Learning: A Dataset, a New Framework, and Its Interpretability

The classification of architectural style for Chinese traditional settlements (CTSs) has become a crucial task for developing and preserving settlements. Traditionally, the classification of CTSs primarily relies on manual work, which is inefficient and time consuming. Inspired by the tremendous success of deep learning (DL), some recent studies attempted to apply DL networks such as convolution neural networks (CNNs) to achieve automated classification of the architecture styles. However, these studies suffer overfitting problems of the CNNs, leading to inferior classification performance. Moreover, most of the studies apply the CNNs as a black box providing limited interpretability. To address these limitations, a new DL classification framework is proposed in this study to overcome the overfitting problem by transfer learning and learning-based data augmentation technique (i.e., AutoAugment). Furthermore, we also employ class activation map (CAM) visualization technique to help understand how the CNN classifiers work to abstract patterns from the input. Specifically, due to a lack of architectural style datasets for the CTSs, a new annotated dataset is first established with six representative classes. Second, several representative CNNs are leveraged to benchmark the new dataset. Third, to address the overfitting problem of the CNNs, a new DL framework is proposed which combines transfer learning and AutoAugment to improve the classification performance. Extensive experiments are conducted on the new dataset to demonstrate the effectiveness of our framework. The proposed framework achieves much better performance than baselines, greatly mitigating the overfitting problem. Additionally, the CAM visualization technique is harnessed to explain what and how the CNN classifiers implicitly learn for recognizing a specified architectural style.

Deep Learning Framework for Classification of Emoji Based Sentiments

Recent patterns of human sentiments are highly influenced by emoji based sentiments (EBS). Social media users are widely using emoji based sentiments (EBS) in between text messages, tweets and posts. Although tiny pictures of emoji contains sufficient information to be considered for construction of classification model; but due to the wide range of dissimilar, heterogynous and complex patterns of emoji with similar meanings (SM) have become one of the significant research areas of machine vision. This paper proposes an approach to provide meticulous assistance to social media application (SMA) users to classify the EBS sentiments. Proposed methodology consists upon three layers where first layer deals with data cleaning and feature selection techniques to detect dissimilar emoji patterns (DEP) with similar meanings (SM). In first sub step we input set of emoji, in second sub step every emoji has to qualify user defined threshold, in third sub step algorithm detects every emoji by considering as objects and in fourth step emoji images are cropped, after data cleaning these tiny images are saved as emoji images. In second step we build classification model by using convolutional neural networks (CNN) to explore hidden knowledge of emoji datasets. In third step we present results visualization by using confusion matrix and other estimations. This paper contributes (1) data cleaning method to detect EBS; (2) highest classification accuracy for emoji classification measured as 97.63%.

Boosting automated sleep staging performance in big datasets using population subgrouping.

Current approaches to automated sleep staging from the electroencephalogram (EEG) rely on constructing a large labeled training and test corpora by aggregating data from different individuals. However, many of the subjects in the training set may exhibit changes in the EEG that are very different from the subjects in the test set. Training an algorithm on such data without accounting for this diversity can cause underperformance. Moreover, test data may have unexpected sensor misplacement or different instrument noise and spectral responses. This work proposes a novel method to learn relevant individuals based on their similarities effectively. The proposed method embeds all training patients into a shared and robust feature space. Individuals who share strong statistical relationships and are similar based on their EEG signals are clustered in this feature space before being passed to a deep learning framework for classification. Using 994 patient EEGs from the 2018 Physionet Challenge (≈6,561 h of recording), we demonstrate that the clustering approach significantly boosts performance compared to state-of-the-art deep learning approaches. The proposed method improves, on average, a precision score from 0.72 to 0.81, a sensitivity score from 0.74 to 0.82, and a Cohen's Kappa coefficient from 0.64 to 0.75 under 10-fold cross-validation.

A deep learning classification framework for early prediction of team-based academic performance

The prediction of the performance of teams during the collaboration of their members is considered a difficult task for various reasons. However, an accurate prediction would assist educators and learning experts in the extraction of useful knowledge for designing learning interventions for improving teams’ outcomes. The main objective of the current paper is to explore and propose a Deep Neural Network (DNN) framework for binary classification with two hidden layers, for the early prediction of teams’ performance in the domain of software engineering. The framework was evaluated by using different activation functions (Sigmoid, ReLu, and Tanh) and optimizers (Adagrad and Adadelta). A dataset created from over 30000 entries grouped in 74 teams was used for training and evaluating the framework. Additionally, the SHapley Additive exPlanations (SHAP) approach was used to interpret the framework and extract the most important features having a positive or negative impact on its final prediction. Among other conclusions, it was shown that the prediction accuracy of the framework, when the Adadelta and Adagrad optimizers were used, was found to be 76.73% and 82.39%, respectively, while its overall learning performance was 80.76% and 86.57%, meaning that it is capable of predicting teams’ performance adequately and accurately.

CNN-based temporal detection of motion saliency in videos

The problem addressed in this paper appertains to the domain of motion saliency in videos. However this is a new problem since we aim to extract the temporal segments of the video where motion saliency is present. It turns out to be a frame-based classification problem. A frame will be classified as dynamically salient if it contains local motion departing from its context. Temporal motion saliency detection is relevant for applications where one needs to trigger alerts or to monitor dynamic behaviours from videos. It can also be viewed as a prerequisite before computing motion saliency maps. The proposed approach handles situations with a mobile camera. It involves two main stages consisting first in cancelling the global motion due to the camera movement, then in applying a deep learning classification framework. We have investigated two ways of implementing the first stage, based on image warping, and on residual flow respectively. Experiments on real videos demonstrate that we can obtain accurate classification in highly challenging situations.

Deep Boltzmann machine algorithm for accurate medical image analysis for classification of cancerous region

In this research work, a deep learning algorithm is applied to the medical domain to deliver a better healthcare system. For this, a deep learning framework for classification the region of interest pattern of complex hyperspectral medical images is proposed. The performance of computer-aided diagnosis by verifying the region in hyperspectral image by pre and post-cancerous region classification is enhanced. For this a deep Boltzmann machine (DBM) architecture of the bipartite structure as an unsupervised generative model was developed. The performance of DBM is compared with deep convolutional neural network architecture. For implementation, a three-layer unsupervised network with a backpropagation structure is used. From the presented dataset, image patches are collected and classified into two classes, namely non-informative and discriminative classes as labelled classes. The spatial information is used for classification and spectral-spatial representation of class labels is formed. In the labelled classes, the accuracy, false-positive predictions, sensitivity are obtained for the proposed fully-connected network. By the proposed cognitive computation technique an accuracy of 95.5% with 93.5% sensitivity was obtained. From the obtained classification, accuracy and success rate DBM provide a better classification of complex images compared to traditional convolution network.

Nepali Text Document Classification Using Deep Neural Network

An automated text classification is a well-studied problem in text mining which generally demands the automatic assignment of a label or class to a particular text documents on the basis of its content. To design a computer program that learns the model form training data to assign the specific label to unseen text document, many researchers has applied deep learning technologies. For Nepali language, this is first attempt to use deep learning especially Recurrent Neural Network (RNN) and compare its performance to traditional Multilayer Neural Network (MNN). In this study, the Nepali texts were collected from online News portals and their pre-processing and vectorization was done. Finally deep learning classification framework was designed and experimented for ten experiments: five for Recurrent Neural Network and five for Multilayer Neural Network. On comparing the result of the MNN and RNN, it can be concluded that RNN outperformed the MNN as the highest accuracy achieved by MNN is 48 % and highest accuracy achieved by RNN is 63%.

3D-CNN based discrimination of schizophrenia using resting-state fMRI.

This study reports a framework to discriminate patients with schizophrenia and normal healthy control subjects, based on magnetic resonance imaging (MRI) of the brain. Resting-state functional MRI data from a total of 144 subjects (72 patients with schizophrenia and 72 healthy controls) was obtained from a publicly available dataset using a three-dimensional convolution neural network 3D-CNN based deep learning classification framework and ICA based features. We achieved 98.09 ± 1.01% ten-fold cross-validated classification accuracy with a p-value < 0.001 and an area under the curve (AUC) of 0.9982 ± 0.015. In addition, differences in functional connectivity between the two groups were statistically analyzed across multiple resting-state networks. The disconnection between the visual and frontal network was prominent in patients, while they showed higher connectivity between the default mode network and other task-positive/ cerebellar networks. These ICA functional network maps served as highly discriminative three-dimensional imaging features for the discrimination of schizophrenia in this study. Due to the very high AUC, this research with more validation on the cross diagnosis and publicly available dataset, may be translated in future as an adjunct tool to assist clinicians in the initial screening of schizophrenia.

An integration of deep learning with feature embedding for protein-protein interaction prediction.

Protein–protein interactions are closely relevant to protein function and drug discovery. Hence, accurately identifying protein–protein interactions will help us to understand the underlying molecular mechanisms and significantly facilitate the drug discovery. However, the majority of existing computational methods for protein–protein interactions prediction are focused on the feature extraction and combination of features and there have been limited gains from the state-of-the-art models. In this work, a new residue representation method named Res2vec is designed for protein sequence representation. Residue representations obtained by Res2vec describe more precisely residue-residue interactions from raw sequence and supply more effective inputs for the downstream deep learning model. Combining effective feature embedding with powerful deep learning techniques, our method provides a general computational pipeline to infer protein–protein interactions, even when protein structure knowledge is entirely unknown. The proposed method DeepFE-PPI is evaluated on the S. Cerevisiae and human datasets. The experimental results show that DeepFE-PPI achieves 94.78% (accuracy), 92.99% (recall), 96.45% (precision), 89.62% (Matthew’s correlation coefficient, MCC) and 98.71% (accuracy), 98.54% (recall), 98.77% (precision), 97.43% (MCC), respectively. In addition, we also evaluate the performance of DeepFE-PPI on five independent species datasets and all the results are superior to the existing methods. The comparisons show that DeepFE-PPI is capable of predicting protein–protein interactions by a novel residue representation method and a deep learning classification framework in an acceptable level of accuracy. The codes along with instructions to reproduce this work are available from https://github.com/xal2019/DeepFE-PPI.

MalDeep: A Deep Learning Classification Framework against Malware Variants Based on Texture Visualization

The increasing sophistication of malware variants such as encryption, polymorphism, and obfuscation calls for the new detection and classification technology. In this paper, MalDeep, a novel malware classification framework of deep learning based on texture visualization, is proposed against malicious variants. Through code mapping, texture partitioning, and texture extracting, we can study malware classification in a new feature space of image texture representation without decryption and disassembly. Furthermore, we built a malware classifier on convolutional neural network with two convolutional layers, two downsampling layers, and many full connection layers. We adopt the dataset, from Microsoft Malware Classification Challenge including 9 categories of malware families and 10868 variant samples, to train the model. The experiment results show that the established MalDeep has a higher accuracy rate for malware classification. In particular, for some backdoor families, the classification accuracy of the model reaches over 99%. Moreover, compared with other main antivirus software, MalDeep also outperforms others in the average accuracy for the variants from different families.

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.