No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Alzheimer’s Disease Detection using Machine Learning Techniques in 3D MR Images
... XG boosting, Gradient boosting, Ada boosting, RFC, SVM linear, Decision tree classifier, Logistic regression, and SVM radial were used to test the dataset. By using ML techniques in 3D MRI scans for AD, a method proposed by Srinivasan Aruchamy et al. [12]. To discriminate between grey and white matter using 3D pictures and identify disease at an early stage, researchers used a novel method in this study. ...
... In addition, numerous researchers have proposed novel CNN architectures for 3D MRI, addressing diverse AD stage classifications. They have also integrated a 3D convolutional network which are densely connected with 3D MRI images, resulting in a fourway categorization (3D DenseNets) [31][32][33][34][35][36]12]. Several authors have put forth a variety of deep learning approaches for Alzheimer's disease [37][38][39][40][41][42][43][44][45][46][47][48], along with alternative machine learning approaches [49][50][51][52][53][54][55][56][57][58]. ...
... The proposed DL-EDAD technique is compared with the 2D-M2IC model [34], the AD-InceptionV3 model [37], the 3D AD using ML [12], and the 3D CapsNets [36] to showcase its enhanced ability in distinguishing patients with early-stage Alzheimer's disease from those with Alzheimer's disease, as well as patients with normal control, early MCI, and late MCI. Because of the CNN-based EfficientNetB7 architecture model, the proposed AD-EDAD approach distinctly surpasses the performance of existing methods. ...
Alzheimer’s Disease (AD) is a degenerative, chronic condition of the brain for which there is now no effective treatment. However, there are medications that can slow its development. In order to stop and control the development of AD, earlier diagnosis of the disease is quintessential. Our proposed method’s primary objective is to establish a comprehensive model for the prior detection of Alzheimer’s disease and the categorization of distinct AD stages. This work employs a deep learning methodology, especially CNN. The proposed approach makes use of well-known models that have already been trained to classify medical images, like the EfficentNetB7 model, by applying the transfer learning principle. In order to achieve greater accuracy, convolutional neural networks (CNNs) are frequently scaled up as new resources become available at a fixed cost throughout the construction phase. A compound coefficient is used by the CNN architecture and scaling approach, which is the foundation of the pre-trained EfficientNetB7 model, to scale the dimensions equally. This proposed EfficientNetB7 model is quicker, easier, and more effective than other pre-trained models like VGG19 and InceptionV3. The proposed model includes simple structures that have memory requirements, provide manageable time, overfitting, and low computational complexity as well as training and inference speeds. The Alzheimer’s disease Neuroimaging Initiative (ADNI) dataset was employed for a comprehensive assessment of the method proposed, utilizing well-known performance metrics including sensitivity, specificity, and accuracy. The findings revealed that the improvised results achieved the accuracy metric when compared to existing methods. The EfficientNetB7 model has been enhanced, and this model achieves a sensitivity of 98.08%, specificity-98%, accuracy-98.2%, and F-score-98.95%, for multi-class AD stage classifications.
... However, the relationship between cognitive decline specifically in the left hemisphere and AD remains insufficiently addressed. Aruchamy et al. (2020) developed a technique to detect AD from 3D MRI scans by using first-order statistical features. It incorporates white matter and grey matter slices from three perspectives: axial, coronal, and sagittal. ...
... Linear regression Used for predicting disease progression or estimating cognitive decline based on relevant variables and time Sivakani and Ansari (2020) LASSO regression Aids in feature selection and regularization to identify important variables associated with AD Mahendran and PM (2022) RVM Delineates the relevance of AD biomarkers and their impact on classification model Pinaya et al. (2021) AdaBoost Useful for improving performance of models for AD classification by combining weak learners into a strong ensemble Aruchamy et al. (2020) and Mahendran and PM (2022) Content courtesy of Springer Nature, terms of use apply. Rights reserved. ...
... Decision stump A simple decision trees with a single split, that are used as weak learners in ensemble methods for AD classification tasks Sivakani and Ansari (2020) Gaussian process Used to model the uncertainty and make predictions in AD detection and diagnosis based on observational and measurement data (Sivakani and Ansari 2020) PCA Takes care of dimensionality reduction and feature extraction. It is useful when data from multiple modalities are fused for training Aruchamy et al. (2020), Bi et al. (2020) and Basheer et al. (2021) MLP Learns complex patterns from input biomarker features Qiu et al. (2022) EM algorithm Aids in probabilistic modeling and clustering. Handles missing data and estimates them ...
Alzheimer’s disease affects around one in every nine persons among the elderly population. Being a neurodegenerative disease, its cure has not been established till date and is managed through supportive care by the health care providers. Thus, early diagnosis of this disease is a crucial step towards its treatment plan. There exist several diagnostic procedures viz., clinical, scans, biomedical, psychological, and others for the disease’s detection. Computer-aided diagnostic techniques aid in the early detection of this disease and in the past, several such mechanisms have been proposed. These techniques utilize machine learning models to develop a disease classification system. However, the focus of these systems has now gradually shifted to the newer deep learning models. In this regards, this article aims in providing a comprehensive review of the present state-of-the-art techniques as a snapshot of the last 5 years. It also summarizes various tools and datasets available for the development of the early diagnostic systems that provide fundamentals of this field to a novice researcher. Finally, we discussed the need for exploring biomarkers, identification and extraction of relevant features, trade-off between traditional machine learning and deep learning models and the essence of multimodal datasets. This enables both medical, engineering researchers and developers to address the identified gaps and develop an effective diagnostic system for the Alzheimer’s disease.
... There was no feature selection performed. The researchers [14] conducted studies using 3D Magnetic Resonance (MR) images of the brain to detect AD. Utilizing the technique of principal component analysis (PCA), feature extraction was performed. ...
Alzheimer's disease (AD) is a common and well-known neurodegenerative condition that causes cognitive impairment. In the field of medicine, it is the "nervous system" disorder that has received the most attention. Despite this extensive research, there is no treatment or strategy to slow or stop its spread. Nevertheless, there are a variety of options (medication and non-medication alternatives) that may aid in the treatment of AD symptoms at their various phases, thereby enhancing the patient's quality of life. As AD advances over time, it is necessary to treat patients at their various stages appropriately. As a result, detecting and classifying AD phases prior to symptom treatment can be beneficial. Approximately twenty years ago, the rate of progress in the field of machine learning (ML) accelerated dramatically. Using ML methods, this study focuses on early AD identification. The "Alzheimer's Disease Neuroimaging Initiative" (ADNI) dataset was subjected to exhaustive testing for AD identification. The purpose was to classify the dataset into three groups: AD, "Cognitive Normal" (CN), and "Late Mild Cognitive Impairment" (LMCI). In this paper, we present the ensemble model Logistic Random Forest Boosting (LRFB), representing the ensemble of “Logistic Regression” (LR), “Random Forest” (RF), and “Gradient Boost” (GB). The proposed LRFB outperformed LR, RF, GB, “k-Nearest Neighbour” (k-NN), “Multi-Layer Perceptron” (MLP), “Support Vector Machine” (SVM), “AdaBoost” (AB), “Naïve Bayes” (NB), “XGBoost” (XGB), “Decision Tree” (DT), and other ensemble ML models with respect to the performance metrics “Accuracy” (Acc), “Recall” (Rec), “Precision” (Prec), and “F1-Score” (FS).
... In the confusion matrix, four different terminologies are used. False positives and False negatives are present, as well as True positives and True negatives [11]. Further accuracy and parameters were calculated for the classification model. ...
Alzheimer’s disease is the most common form of dementia affecting the brain’s parts. A broad term used to describe illnesses and conditions that causes a deterioration in memory, language, and other cognitive abilities severe enough to interface with daily life is “dementia”. According to estimates, this disease affects 6.2 million Americans and 5 million people in India aged 65and older. In 2019, the most recent year for which data are available, official death certificates reported 121,499 deaths from AD, making Alzheimer’s the “sixth leading cause of death in the country and the fifth leading cause of death for people 65 and older”. In this paper, we suggest several machine Learning algorithms like Decision trees, SVM, Logistic regression, and Naive Bayes identify AD at an early stage. The Alzheimer's Disease Neuroimaging Initiative (ADNI) and the Open Access Series of Imaging Investigations (OASIS) provide data sets white used to detect the disease in its early stage. The datasets consist of longitudinal MRI data (age, gender, mini mental status, CDR) By taking into account many factors in each method, such as precision, F1 Score, Recall, and specificity are calculated. The results obtained 93.7% of maximum accuracy for the Decision Tree Algorithm.
... Then, the CNN was introduced in order to extract the characteristics of the segmented image for the classification. The method in [56] used a 3D structural brain MRI images to isolate the MRI images of white and gray matter, extract 2D slices and pick main slices from them for extraction of features. In order to calculate the first-order statistical features, feature extraction is applied on top of these slices and the prominent feature vectors produced by PCA were selected. ...
Detection of Alzheimer's disease (AD) is one of the most potent and daunting activities in the processing of medical imagery. The survey of recent AD detection techniques in the last 10 years is described in this paper. The AD detection process involves various steps, namely preprocessing, feature extraction, feature selection, dimensionality reduction, segmentation and classification. In this study, we reviewed the latest findings and possible patterns as well as their main contributions. Different types of AD detection techniques are also discussed. Based on the applied algorithms and methods , and the evaluated databases (e.g., ADNI and OASIS), the performances of the most relevant AD detection techniques are compared and discussed.
... Other studies have used voxel-based MRI image feature extraction with machine learning algorithms for Alzheimer detection [18]. The proposed method consists of 3-D structural brain MR images separating the white and grey matter MR images, extracting 2D slices in the coronal, sagittal and axial directions and selecting the key slices from them to do feature extraction. ...
div>
This paper applies Ensemble Learning models for the early detection of Alzheimer’s disease in elderly adults. The publicly available dataset from the Open Access Series of Imaging Studies (OASIS) Database is used. A novel longitudinal MRI data-based machine learning model is proposed in the paper, which takes account of features like- Mini-Mental State Examination (MMSE) score and years of education to make a generalized classifier. Our proposed model achieved a 5-fold cross-validation area under the curve (AUC) score of 89.93% and accuracy of 94.64%. We show that our results quantitatively outperform the state-of-the-art in Alzheimer’s disease detection. We then compared our results to other previous state-of-the-art research and our model’s metrics surpasses them.
</div
... Aruchamy et al. [11] present novel techniques for detection of AD utilizing first order statistic features from 3D-brain Magnetic Resonance Imaging (MRI). The presented work utilizes 3D structural brain MRI for separating the white as well as grey matter MRI, remove 2D slices under the axial directions, coronal, and sagittal and elect the key slice for carrying out feature extraction. ...
Presently, cognitive Internet of Things (CIoT) with cloud computing (CC) enabled intelligent healthcare models are developed, which enables communication with intelligent devices, sensor modules, and other stakeholders in the healthcare sector to avail effective decision making. On the other hand, Alzheimer disease (AD) is an advanced and degenerative illness which injures the brain cells, and its earlier detection is necessary for suitable interference by healthcare professional. In this aspect, this paper presents a new Oriented Features from Accelerated Segment Test (FAST) with Rotated Binary Robust Independent Elementary Features (BRIEF) Detector (ORB) with optimal artificial neural network (ORB-OANN) model for AD diagnosis and classification on the CIoT based smart healthcare system. For initial pre-processing, bilateral filtering (BLF) based noise removal and region of interest (RoI) detection processes are carried out. In addition, the ORB-OANN model includes ORB based feature extractor and principal component analysis (PCA) based feature selector. Moreover, artificial neural network (ANN) model is utilized as a classifier and the parameters of the ANN are optimally chosen by the use of salp swarm algorithm (SSA). A comprehensive experimental analysis of the ORB-OANN model is carried out on the benchmark database and the obtained results pointed out the promising outcome of the ORB-OANN technique in terms of different measures.
... It was observed that the efficiency of detection of the AD could b increased by focussing on the regions affected by the brain during the AD. Srinivasan et.al [10] proposed a new method for the detection of AD in three dimensional MR images. Initially, the Grey matter and white matter of the brain volume is extracted for analysis of AD. ...
... In future studies, they suggested to use it on X-ray images for detecting the breast cancer. In 2020, Aruchamy et al. [22], suggested a method to detect the Alzheimer's disease using MRI. They applied different classifier on Grey and White matter because they get affected by Alzheimer's disease. ...
In this paper machine learning algorithms have been discussed and analyzed. It has been discussed considering computational aspects in different domains. These algorithms have the capability of building mathematical and analytical model. These models may be helpful in the decision-making process. This paper elaborates the computational analysis in three different ways. The background and analytical aspect have been presented with the learning application in the first phase. In the second phase detail literature has been explored along with the pros and cons of the applied techniques in different domains. Based on the literatures, gap identification and the limitations have been discussed and highlighted in the third phase. Finally, computational analysis has been presented along with the machine learning results in terms of accuracy. The results mainly focus on the exploratory data analysis, domain applicability and the predictive problems. Our systematic analysis shows that the applicability of machine learning is wide and the results may be improved based on these algorithms. It is also inferred from the literature analysis that at the applicability of machine learning algorithm has the capability in the performance improvement. The main methods discussed here are classification and regression trees (CART), logistic regression, naïve Bayes (NB), k-nearest neighbors (KNN), support vector machine (SVM) and decision tree (DT). The domain covered mainly are disease detection, business intelligence, industry automation and sentiment analysis.
Alzheimer's disease (AD) is a degenerative condition that can cause anything from a slight loss of memory to total loss of consciousness and speech. Early detection has a critical role in maintaining the patient's quality of life. Despite a wealth of studies on AD diagnosis, early and correct diagnosis is most beneficial to patients. Because machine learning (ML) models may identify abnormalities early on, they have become indispensable in the diagnosis of diseases such as AD. ML and computer-aided diagnostics (CAD) have been combined, and this has enhanced AD detection—especially when integrating with MRI data. ML methods are preferred because they produce results quickly and accurately. The goal of this research is to create an automated AD detection system that is more sophisticated and accurate by integrating data from many modalities. The goal of this strategy is to lower the rate of incorrect diagnoses while offering a more thorough diagnostic, emphasizing accuracy, sensitivity, and specificity.
Recent years have seen an uptick in the use of computed tomography (CT) and magnetic resonance imaging (MRI) scans to create three-dimensional images of the human body for use in medical image processing studies. Immunisations and medical treatment do not work to prevent or treat chronic diseases. Some examples of chronic ailments are asthma, cancer, heart disease, diabetes, and Alzheimer's disease. Alzheimer's disease is a progressive neurodegenerative illness that destroys both memory and personality over time. Without regular checks, diseases like Alzheimer's could not be seen until they've progressed to a fatal level. Millions of individuals throughout the globe are living with Alzheimer's disease, which is a leading cause of death. The first indicator of Alzheimer's disease may be moderate cognitive and/or behavioural impairment, followed by preclinical illness and, ultimately, full-blown Alzheimer's disease. This machine learning model outperforms state-of-the-art medical ailment prediction methods. Most machine learning algorithms for Alzheimer's disease identification are limited to low-dimensional feature spaces because of the sparsity problem. Research in this article examines the feasibility of using several methods such as deep learning, machine learning, and transfer learning approaches to create an early Alzheimer's disease diagnosis.
Alzheimer's disease is a neurodegenerative condition that has a aging population. Early diagnosis and intervention are critical. A valuable tool for the detection .This research aims to develop a deep learning-based approach for the automatic recognition in MRI images. In this study, architecture is employed to extract meaningful features from MRI images. Alzheimer's and non-Alzheimer's MRI images is utilized. The proposed deep learning approach in Alzheimer's disease recognition. The model achieves a high level of accuracy, sensitivity, and specificity in classifying MRI images. Potential to assist medical professionals in the early diagnosis and monitoring of Alzheimer's disease, ultimately contributing to improved patient care and outcomes
As the global population ages, the prevalence of dementia caused by Alzheimer's disease is expected to increase, posing a significant healthcare challenge. Dementia is a leading cause of disability and dependency among older people, and it also has broad physical, psychological, social, and economic impacts on individuals, their caregivers, families, and communities. Despite the existence of treatments to manage dementia symptoms, a lack of awareness and understanding of the condition often leads to stigmatization and difficulties in accessing diagnosis and care. Neuroimaging is a viable method for reaching this objective because early Alzheimer's disease detection is essential for successful intervention. Although dementia cannot be cured, controlling symptoms can enhance the quality of life for those who have the condition. The objective is to employ structural magnetic resonance imaging (sMRI) information to examine deep learning algorithms for the identification of significant biomarkers related to Alzheimer's disease. The suggested method employs Convolutional Neural Networks (CNNs) to classify brain images into one of categories of Alzheimer disease as Non Demented, Very Mild Demented, Mild Demented and Moderate Demented.
A broad term used to describe diseases and conditions that cause deterioration in memory, language, and other mental capacities sufficiently extreme to communicate with day-to-day existence is "dementia". Alzheimer's disease is the most well-known type of Dementia influencing the mind's parts. As per range, this disorder influences 6.2 million Americans and 5 million individuals in India matured 65 and more seasoned. In 2019, the latest year for which information is accessible, official passing declarations revealed 121,499 deaths from Promotion, Alzheimer's, the "6th driving reason for death in the nation". In this paper, we propose AI calculations like Decision trees (DT), SVM, Linear regression, and Naive Bayes determines Promotion at the beginning phase. The Alzheimer's Sickness Neuroimaging Drive (ADNI) and the Open Access Series of Imaging Examinations give informational collections used to identify the disease in its beginning phase. The datasets comprise longitudinal X-ray information (age, orientation, small-scale mental status, CDR). By taking into; account many variables in every strategy, for example, accuracy, F1 Score, Review, and explicitness are determined. The outcomes acquired 93.7% of the greatest precision for the DT Calculation.
This model of recovering useful data and models from the information is called KDD (Knowledge Discovery of Database), which includes specific steps like information finding, grouping and change review. AI analyses are called managed and independently. A supervised learning analysis uses insight to predict new or unseeable information, though unaided measures can draw impedances from informative clusters. Supervised learning is additionally described as arrangement. This review uses clustering methods to deliver a more precise area with the class. The clustering analyses have been applied to the Indian Diabetes Data-set of the PIMA of the National Institute of Diabetes, Stomach related and kidney diseases which contains information on diabetic women
Dementia is a neurological disease that that encompasses a wide range of conditions like verbal communication, problem-solving, and other judgment abilities that are severely sufficient to interfere with daily life. It is among the leading causes of vulnerability among the elderly all over the world. A considerable amount of research has been conducted in this area so that we can perform early detection of the disease, yet further research into its betterment is still an emerging trend. This article compares the performance of multiple machine learning models for dementia detection and classification using brain MRI data, including support vector machine, random forest, AdaBoost, and XGBoost. Meanwhile, the research conducts a systematic assessment of papers for the clinical categorization of dementia using ML algorithms and neuroimaging data. The authors used 373 participants from the OASIS database. Among the tested models, RF model exhibited the best performance with 83.92% accuracy, 87.5% precision, 81.67% recall, 84.48% F1-score, 81.67% sensitivity, and 88.46% specificity.
Alzheimer is a kind of dementia that affects reasoning and social aptitudes which results in constant decline of individual’s capacity to work. The development of amyloid protein and tau protein in cerebrum stimulates cell passing. The prevention and early prediction is a big challenge. The proposed model addresses this need with the analysis of parameters such as Mini-Mental State Examination score and so on. In addition to these parameters, one more parameter dice coefficient value that shows difference between normal and affected scan images. The SVM and CNN are applied on various neuro parameters and evaluated for prediction accuracy.KeywordsNeuro declineDementiaAlzheimer’s disease and machine learning algorithms
Monitoring healthcare is a major issue, that requires attention. In underdeveloped countries, the number of nurses for patients is relatively low, and the accessibility of 24-h medical supervision is also ambiguous, resulting in the incidence of easily avoidable deaths as well as urgent situations, causing a disturbance in the health sector. The medical dispensers that are currently available are expensive, and devices that combine a reminder and a dispenser are hard to come by. The major goal of the medicine reminder is to automatically transmit an alarm and dispense medicine to the correct individual at the stated time from a single machine. An automatic medicine distributor is developed for persons who take medicine without expert direction. It is used by a single patient or by a group of patients. It discharges the individual of the error-prone job of injecting the incorrect medicine at the incorrect time. The main goal line is to retain the device simple usage and affordable. Working software is trustworthy and steady. The older age population will benefit greatly from the device because it can substitute expensive medicinal treatment and the money spent on a personal nurse.KeywordsInternet of ThingsMedical update frameworksGSM moduleArduino
Among the population of India, nearly 1.6 million children are blind or visually impaired by birth. The challenges faced by the visually impaired persons are high in day-to-day life. There are number of challenges faced by them like requiring guidance for the usage of public transport, to walk independently. One of the most common problems is the dependency on others for the purchase of household items. Hence, a unique solution is required which makes the people to identify and purchase the products in the supermarket without dependency. This research focuses on a portable camera-based assistance that helps them to identify the grocery items along with the framework to read the text on the labels. This experimental system consists of three modules: object detection and classification; reading the text from the label; text to audio converter. The input to the camera is fed by the Raspberry Pi kit. The captured video is segregated as pictures of the grocery items. The proposed deep learning algorithm for the detection of object is You only look once (YOLO). The accuracy of the detection is pre-dominantly high compared to the existing algorithms. The captured image of the grocery item label is preprocessed to blur the unwanted text. Optical character recognition (OCR) is applied over the label to convert as the text where machine can read. At last, the processed text is converted as audio for the guidance of the people. Thus, a flexible supervisory mechanism with user-friendliness is developed for helping the visually impaired people.KeywordsDeep learningVisually impairedSmart super market assistanceOptical character recognition
A prevailing disease which strikes a multitude of people in the present era is Alzheimer’s Disease. Alzheimer’s Disease is a neurodegenerative disorder that causes the death of brain cells destroying memory and thinking abilities gradually. To limit the number of Alzheimer’s illness, a few frameworks have been recommended to some extent in recent times. One such computer aided framework is the use of machine learning for the digital diagnosis of Alzheimer’s. Machine Learning recognizes the extent of certain diseases and informs the experts about the irregularities. This paper investigates the application of Machine Learning models in Alzheimer’s Disease detection.
Advancement in medical field increases population of ageing society and monitor their health using sensors without disturbing their routine works. At the same time, elderly persons suffer from various age-related cognitive impairments. This work focuses on Alzheimer Disease (AD) which is one of the important causes of dementia. Though there are many existing machine learning models and deep learning approaches are developed for AD detection, handling class imbalance and inconsistencies are the primary issues faced by these standard classification models. The main objective of this paper is to construct an enhanced model which takes both class imbalance and inconsistencies in Alzheimer disease detection at its earlier stages using deep learning classification model. The proposed Enhanced Deep Hierarchical Classification Model (EDHCM) comprised of three different layers functions in a hierarchal manner of parent–child relationship using CASAS dataset which is a smart home resident’s daily activity collected using sensor devices. The first layer is the flat neural network whose functionality is enhanced using analytical hierarchal processing to fine-tune the parameters. The next layer is the hierarchal embedding layer which uses softmax to classify the severity of Alzheimer dataset. Final layer is hierarchical loss network which improves the learning model by punishing it based on layer loss and dependency loss. The simulation results explored the efficiency of the EDHCM by producing highest rate of accurate detection rate with less false alarms in AD detection compared to Support Vector Machine, Artificial Neural Network, Multilayer perceptron and Deep Neural Network.
Despite decades of research, at present there is no curative therapy for Alzheimer's disease. Changes in the way new drugs are tested appear to be necessary. Three changes are presented here and will be discussed. The first change is that Alzheimer's disease must be considered a disease of four major pathological processes, not one. The four processes are: 1) vascular hypoperfusion of the brain with associated mitochondrial dysfunction, 2) destructive protein inclusions, 3) uncontrolled oxidative stress, and 4) proinflammatory immune processes secondary to microglial and astrocytic dysfunction in the brain. The second change recommended is to alter the standard cognitive measurement tools used to quantify mental decline in test patients. Specifically the Dementia Severity Rating Scale (DSRS) should supersede Mini-Mental State Examination (MMSE) and other popular tests, and a measurement scale developed in research should be used to produce a linear and non-irregular baseline. Finally, accepting the concept that four etiologies cause Alzheimer's disease leads to the last necessary change, that new therapies must be employed directed against all four causes, likely as a combination. There are drugs ready to be employed in such a combinations which are available and used clinically for other purposes so can be used "off label" and one such combination is suggested.
In this study, a probabilistic model, named as BayGmmKda, is proposed for flood susceptibility assessment in a study area in central Vietnam. The new model is a Bayesian framework constructed by a combination of a Gaussian mixture model (GMM), radial-basis-function Fisher discriminant analysis (RBFDA), and a geographic information system (GIS) database. In the Bayesian framework, GMM is used for modeling the data distribution of flood-influencing factors in the GIS database, whereas RBFDA is utilized to construct a latent variable that aims at enhancing the model performance. As a result, the posterior probabilistic output of the BayGmmKda model is used as flood susceptibility index. Experiment results showed that the proposed hybrid framework is superior to other benchmark models, including the adaptive neuro-fuzzy inference system and the support vector machine. To facilitate the model implementation, a software program of BayGmmKda has been developed in MATLAB. The BayGmmKda program can accurately establish a flood susceptibility map for the study region. Accordingly, local authorities can overlay this susceptibility map onto various land-use maps for the purpose of land-use planning or management.
Pathological brain detection by computer vision is now attracting intense attentions from academic fields. Nevertheless, most of recent methods suffer from low-accuracy. This study combined two successful techniques: pseudo Zernike moment and kernel support vector machine. Three open datasets were downloaded and used. The 10 times of K-fold stratified cross validation showed our method using 19-order pseudo Zernike moments achieved perfect classification on the first dataset. It achieved a sensitivity of 99.93 ± 0.23%, a specificity of 98.50 ± 2.42%, and an accuracy of 99.75 ± 0.32% on the second dataset. It achieved a sensitivity of 99.64 ± 0.42%, a specificity of 98.29 ± 2.76%, and an accuracy of 99.45 ± 0.38% on the third dataset. This approach performs better than eleven state-of-the-art smart pathological brain detection methods.
Background: Missing values are commonly encountered on the Mini Mental State Examination (MMSE), particularly when administered to frail older people. This presents challenges for MMSE scoring in research settings. We sought to describe missingness in MMSEs administered in Long Term Care Facilities (LTCF), and to compare and contrast approaches to dealing with missing items.
Methods: As part of the Care and Construction project in Nova Scotia, Canada, LTCF residents completed an MMSE. Different methods of dealing with missing values (e.g., use of raw scores, raw scores/number of items attempted, scale-level multiple imputation (MI), and blended approaches) are compared to item-level MI.
Results: 320 residents living in 23 LTCF were administered the MMSE. 38% of participants were aged >85 years; 73% were women. 122 (38.2%) had at least one missing item. Data were not Missing Completely at Random (MCAR), 2 (1110) =1351, p <. 001. Using Raw scores for those missing <6 items in combination with scale-level MI resulted in the regression coefficients and standard errors closest to item-level MI.
Conclusions: Patterns of missing items often suggest systematic problems such as trouble with manual dexterity, literacy or visual impairment. While these observations may be relatively easy to take into account in clinical settings, non-random missingness presents challenges for research, and must be considered in statistical analyses. We present suggestions for dealing with missing MMSE data based on the extent of missingness and the goal of analyses.
Article available at http://www.sciencedirect.com/science/article/pii/S0917504016301538
© 2016 The Authors. Publishing services by Elsevier B.V. on behalf of The Japan Epidemiological
Association. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/ licenses/by-nc-nd/4.0/).
Purpose:
To evaluate the value of quantitative volumetric functional MR imaging in early assessment of response to yttrium-90-labeled ((90)Y) transarterial radioembolization (TARE) in patients with hypo-vascular liver metastases.
Materials and methods:
Seventy four metastatic lesions in 14 patients with hypo-vascular liver metastases after TARE were included in this retrospective study. Diffusion and contrast-enhanced MR imaging was performed before and early after treatment. All MR images were analyzed by two experienced radiologists. Response by anatomic metrics (RECIST, mRECIST, EASL) and functional metrics (ADC and arterial and venous enhancement) were reported in targeted and non-targeted lesions. A two-sample paired t test was used to compare the changes after TARE. A p value of <0.05 was considered statistically significant.
Results:
The anatomic metrics did not show any significant changes in both targeted and non-targeted groups. Targeted lesions demonstrated an increase in mean volumetric ADC (23.4%; p = 0.01), a decrease in arterial and venous enhancement (-22.9% and -6.7%, respectively; p < 0.001 and p = 0.002, respectively) 1 month after treatment. Twenty one responding lesions (42%) by RECIST at 6 months demonstrated a significant increase in volumetric ADC (37.2%; p = 0.01), decrease in arterial and venous enhancement (-58.5% and -23.9%, respectively; p < 0.001) at 1 month post-treatment. Responding lesions did not change significantly by anatomic metrics.
Conclusions:
RECIST, mRECIST, and EASL criteria failed to stratify lesions into responders and non-responders early after TARE in hypo-vascular liver metastasis. Quantitative volumetric functional MR imaging could be a promising tool as a biomarker for predicting early response and can potentially be utilized in clinical trials.
Functional magnetic resonance imaging (fMRI) is one of the most promising non-invasive techniques for early Alzheimer's disease (AD) diagnosis. In this paper, we explore the application of different machine learning techniques to the classification of fMRI data for this purpose. The functional images were firstly preprocessed using the statistical parametric mapping toolbox to output individual maps of statistically activated voxels. A fast filter was applied afterwards to select voxels commonly activated across demented and non-demented groups. Four feature ranking selection techniques were embedded into a wrapper scheme using an inner-outer loop for the selection of relevant voxels. The wrapper approach was guided by the performance of six pattern recognition models, three of which were ensemble classifiers based on stochastic searches. Final classification performance was assessed from the nested internal and external cross-validation loops taking several voxel sets ordered by importance. Numerical performance was evaluated using statistical tests, and the best combination of voxel selection and classification reached a 97.14% average accuracy. Results repeatedly pointed out Brodmann regions with distinct activation patterns between demented and non demented profiles, indicating that the machine learning analysis described is a powerful method to detect differences in several brain regions between both groups.
Background
Timely diagnosis of Alzheimer’s disease (AD) refers to a diagnosis at the stage when patients come to the attention of clinicians because of concerns about changes in cognition, behavior, or functioning and can be still free of dementia and functionally independent.
Objectives
To comprehensively review existing scientific evidence on the benefits and potential challenges of making a timely diagnosis of AD.
Methods
Relevant studies were identified by searching electronic databases (Medline, Embase) and bibliographies for studies published in English between 1 January 2000 and 2 June 2014 on the consequences of a timely diagnosis of AD.
Results
Nine studies were identified that investigated the consequences of diagnosing AD at the initial stages; none were specifically focused on prodromal AD. A timely diagnosis potentially offers the opportunities of early intervention, implementation of coordinated care plans, better management of symptoms, patient safety, cost savings, and postponement of institutionalization. Barriers to making a timely diagnosis include stigma, suicide risk, lack of training, diagnostic uncertainty, shortage of specialized diagnostic services, and the reluctance of healthcare providers to make a diagnosis when no effective disease-modifying options are available.
Conclusions
Despite its potential benefits, few published studies have explored the advantages or risks of a timely diagnosis of AD. In light of the cultural shift toward diagnosis at the initial stage of the disease continuum, when the patient does not yet have dementia, more investigations are needed to evaluate the benefits and address the barriers that may impede making a timely AD diagnosis.
Due to the aging of the global population and the derivative increase in aging-related non-communicable diseases and their economic burden, there is an urgent need to promote research on aging and aging-related diseases as a way to improve healthy and productive longevity for the elderly population. To accomplish this goal, we advocate the following policies: 1) Increasing funding for research and development specifically directed to ameliorate degenerative aging processes and to extend healthy and productive lifespan for the population; 2) Providing a set of incentives for commercial, academic, public and governmental organizations to foster engagement in such research and development; and 3) Establishing and expanding coordination and consultation structures, programs and institutions involved in aging-related research, development and education in academia, industry, public policy agencies and at governmental and supra-governmental levels.
Voxel-based morphometry (VBM) has been proven capable of capturing cerebral gray matter asymmetries with a high (voxel-wise) regional specificity. However, a standardized reference on how to conduct voxel-wise asymmetry analyses is missing. This protocol provides the scientific community with a carefully developed guide describing, in 12 distinct steps, how to take structural images from data pre-processing, via statistical analysis, to the final interpretation of the significance maps. Key adaptations compared with the standard VBM workflow involve establishing a voxel-wise hemispheric correspondence, capturing the direction and degree of asymmetry and preventing a blurring of information across hemispheres. The workflow incorporates the most recent methodological developments, including high-dimensional spatial normalization and partial volume estimations. Although the protocol is primarily designed to enable relatively inexperienced users to conduct a voxel-based asymmetry analysis on their own, it may also be useful to experienced users who wish to efficiently adapt their existing scripts or pipelines.
In this letter we discuss a least squares version for support vector machine (SVM) classifiers. Due to equality type constraints in the formulation, the solution follows from solving a set of linear equations, instead of quadratic programming for classical SVM''s. The approach is illustrated on a two-spiral benchmark classification problem.
The Open Access Series of Imaging Studies is a series of neuroimaging data sets that are publicly available for study and analysis. The present MRI data set consists of a longitudinal collection of 150 subjects aged 60 to 96 years all acquired on the same scanner using identical sequences. Each subject was scanned on two or more visits, separated by at least 1 year for a total of 373 imaging sessions. Subjects were characterized using the Clinical Dementia Rating (CDR) as either nondemented or with very mild to mild Alzheimer's disease. Seventy-two of the subjects were characterized as nondemented throughout the study. Sixty-four of the included subjects were characterized as demented at the time of their initial visits and remained so for subsequent scans, including 51 individuals with CDR 0.5 similar level of impairment to individuals elsewhere considered to have “mild cognitive impairment.” Another 14 subjects were characterized as nondemented at the time of their initial visit (CDR 0) and were subsequently characterized as demented at a later visit (CDR > 0). The subjects were all right-handed and include both men (n = 62) and women (n = 88). For each scanning session, three or four individual T1-weighted MRI scans were obtained. Multiple within-session acquisitions provide extremely high contrast to noise, making the data amenable to a wide range of analytic approaches including automated computational analysis. Automated calculation of whole-brain volume is presented to demonstrate use of the data for measuring differences associated with normal aging and Alzheimer's disease.
This paper develops a new neural network architecture for modeling spatial distributions (i.e. distributions on ) which is more computationally efficient than a traditional fully-connected feedforward architecture. The design of the architecture takes advantage of the specific structure of limit order books. The new architecture, which we refer to as a ‘spatial neural network’, yields a low-dimensional model of price movements deep into the limit order book, allowing more effective use of information from deep in the limit order book (i.e. many levels beyond the best bid and best ask). The spatial neural network models the joint distribution of the state of the limit order book at a future time conditional on the current state of the limit order book. The spatial neural network outperforms status quo models such as the naive empirical model, logistic regression (with nonlinear features), and a standard neural network architecture. Both neural networks strongly outperform the logistic regression model. Due to its more effective use of information deep in the limit order book, the spatial neural network especially outperforms the standard neural network in the tail of the distribution, which is important for risk management applications. The models are trained and tested on nearly 500 U.S. stocks. Techniques from deep learning such as dropout are employed to improve performance. Due to the significant computational challenges associated with the large amount of data, models are trained with a cluster of 50 GPUs.
Magnetic resonance image analysis for detection of some neurodegenerative diseases are investigated and reported in this work. It is based on discrete wavelet transform in combination with different fractal analysis. The proposed action consists of two stages. In the first stage consists of image preprocessing which includes image enhancement, the region of extraction and skull stripping. In the second stage, the preprocessed image is converted to wavelet domain, and the following study is performed for feature extraction using fractal analysis. Finally proposed work has experimented with a publically available dataset of images having the mild cognitive impairment, Alzheimer's and healthy patient. The support vector machine based classifier has achieved the classification accuracy of 89.7 +- 0.6. This work will pave to develop a system for computer-aided diagnosis of some neurological diseases.
This book provides an introduction to the analysis of multivariate data. It describes multivariate probability distributions, the preliminary analysisof a large -scale set of data, princ iple component and factor analysis, traditional normal theory material, as well as multidimensional scaling andcluster analysis. Introduction to Multivariate Analysis provides a reasonable blend oftheory and practice. Enough theory is given to introduce the concepts andto make the topics mathematically interesting. In addition the authors discussthe use (and misuse) of the techniques in pra ctice and present appropriatereal-life examples from a variety of areas includ ing agricultural research, sociology and crim inology. The book should be suitable both for researchworkers and as a text for students taking a course on multivariate analysis.
Motivation:
Heatmap is a popular visualization technique in biology and related fields. In this study, we extend heatmaps within the framework of matrix visualization (MV) by incorporating a covariate adjustment process through the estimation of conditional correlations. MV can explore the embedded information structure of high-dimensional large-scale data sets effectively without dimension reduction. The benefit of the proposed covariate-adjusted heatmap is in the exploration of conditional association structures among the subjects or variables that cannot be done with conventional MV.
Results:
For adjustment of a discrete covariate, the conditional correlation is estimated by the within and between analysis. This procedure decomposes a correlation matrix into the within- and betweencomponent matrices. The contribution of the covariate effects can then be assessed through the relative structure of the between-component to the original correlation matrix while the within-component acts as a residual. When a covariate is of continuous nature, the conditional correlation is equivalent to the partial correlation under the assumption of a joint normal distribution. A test is then employed to identify the variable pairs which possess the most significant differences at varying levels of correlation before and after a covariate adjustment. In addition, a z-score significance map is constructed to visualize these results. A simulation and three biological datasets are employed to illustrate the power and versatility of our proposed method.
Availability:
GAP is available to readers and is free to non-commercial applications. The installation instructions, the user's manual, and the detailed tutorials can be found at http://gap.stat.sinica.edu.tw/Software/GAP.
Contact:
cchen@stat.sinica.edu.tw.
Supplementary information:
Supplementary data are available at Bioinformatics online.
This paper describes a new paradigm of machine learning, in which Intelligent Teacher is involved. During training stage, Intelligent Teacher provides Student with information that contains, along with classification of each example, additional privileged information (for example, explanation) of this example. The paper describes two mechanisms that can be used for significantly accelerating the speed of Student's learning using privileged information: (1) correction of Student's concepts of similarity between examples, and (2) direct Teacher-Student knowledge transfer.
Explored is the utility of modelling brain magnetic resonance images as a fractal object for the classification of healthy brain images against those with Alzheimer's disease (AD) or mild cognitive impairment (MCI). More precisely, fractal multi-scale analysis is used to build feature vectors from the derived Hurst's exponents. These are then classified by support vector machines (SVMs). Three experiments were conducted: in the first the SVM was trained to classify AD against healthy images. In the second experiment, the SVM was trained to classify AD against MCI and, in the third experiment, a multiclass SVM was trained to classify all three types of images. The experimental results, using the 10-fold cross-validation technique, indicate that the SVM achieved 97.08% ± 0.05 correct classification rate, 98.09% ± 0.04 sensitivity and 96.07% ± 0.07 specificity for the classification of healthy against MCI images, thus outperforming recent works found in the literature. For the classification of MCI against AD, the SVM achieved 97.5% ± 0.04 correct classification rate, 100% sensitivity and 94.93% ± 0.08 specificity. The third experiment also showed that the multiclass SVM provided highly accurate classification results. The processing time for a given image was 25 s. These findings suggest that this approach is efficient and may be promising for clinical applications.
Noise sensitivity is known as a key related issue of AdaBoost algorithm. Previous works exhibit that AdaBoost is prone to be overfitting in dealing with the noisy data sets due to its consistent high weights assignment on hard-to-learn instances (mislabeled instances or outliers). In this paper, a new boosting approach, named noise-detection based AdaBoost (ND-AdaBoost), is exploited to combine classifiers by emphasizing on training misclassified noisy instances and correctly classified non-noisy instances. Specifically, the algorithm is designed by integrating a noise-detection based loss function into AdaBoost to adjust the weight distribution at each iteration. A k-nearest-neighbor (k-NN) and an expectation maximization (EM) based evaluation criteria are both constructed to detect noisy instances. Further, a regeneration condition is presented and analyzed to control the ensemble training error bound of the proposed algorithm which provides theoretical support. Finally, we conduct some experiments on selected binary UCI benchmark data sets and demonstrate that the proposed algorithm is more robust than standard and other types of AdaBoost for noisy data sets.
Research on neuropathology indicates that Alzheimer's disease is characterized by loss of neurons and synapses in the cerebral cortex and other subregions, which can be measured by the thickness of cortex from the magnetic resonance imaging (MRI). A classification method based on Gaussian mixture model (GMM) under Bayesian framework is proposed to facilitate the automated diagnosis of Alzheimer's disease based on the cortical thickness, and EM algorithm is employed to solve the parameters of Gaussian mixture model. The experiment shows that our method is outstanding over the common supervised learning methods.
This paper presents a novel computer-aided diagnosis (CAD) technique for the early diagnosis of the Alzheimer's disease (AD) based on nonnegative matrix factorization (NMF) and support vector machines (SVM) with bounds of confidence. The CAD tool is designed for the study and classification of functional brain images. For this purpose, two different brain image databases are selected: a single photon emission computed tomography (SPECT) database and positron emission tomography (PET) images, both of them containing data for both Alzheimer's disease (AD) patients and healthy controls as a reference. These databases are analyzed by applying the Fisher discriminant ratio (FDR) and nonnegative matrix factorization (NMF) for feature selection and extraction of the most relevant features. The resulting NMF-transformed sets of data, which contain a reduced number of features, are classified by means of a SVM-based classifier with bounds of confidence for decision. The proposed NMF-SVM method yields up to 91% classification accuracy with high sensitivity and specificity rates (upper than 90%). This NMF-SVM CAD tool becomes an accurate method for SPECT and PET AD image classification.
Brain MR Image Analysis using Discrete wavelet Transform with Fractal Feature Analysis
- A Srinivasan
- Battacharjee
- G Prasad
- Sanyal
Srinivasan A, Battacharjee P, Prasad A, Sanyal G. Brain MR Image
Analysis using Discrete wavelet Transform with Fractal Feature Analysis. In2018 Second International Conference on Electronics, Communication and Aerospace Technology (ICECA) 2018 Mar 29 (pp. 1660-1664). IEEE.
Automated skull stripping in brain MR images
- S Aruchamy
- Kumar
- G Bhattacharjee
- Sanyal
Aruchamy S, Kumar RK, Bhattacharjee P, Sanyal G. Automated skull
stripping in brain MR images. In2016 3rd International Conference on
Computing for Sustainable Global Development (INDIACom) 2016 Mar
16 (pp. 2043-2047). IEEE.
Voxel-based diagnosis of Alzheimer’s disease using classifier ensembles
- R Armaanzas
- Iglesias
- L Da Morales
- Alonso-Nanclares
Armaanzas R, Iglesias M, Morales DA, Alonso-Nanclares L. Voxelbased diagnosis of Alzheimer's disease using classifier ensembles. IEEE
journal of biomedical and health informatics. 2016 Mar 4;21(3):778-84.
Brain tumor segmentation using K-means clustering and fuzzy C-means algorithms and its area calculation and disease prediction using naive-bayes algorithm
- R Mandhare
- P Nawale
- Kulkarni
- S Ansari
- Kakade
Mandhare R, Nawale J, Kulkarni P, Ansari A, Kakade S. Brain tumor
segmentation using K-means clustering and fuzzy C-means algorithms
and its area calculation and disease prediction using naive-bayes algorithm. Int J Innov Res Sci Eng Technol. 2018;7(6).
NMF-SVM based CAD tool applied to functional brain images for the diagnosis of Alzheimer’s disease
- P Padilla
- Lpez
- Grriz
- D Ramirez
- I Salas-Gonzalez
- Alvarez
Padilla P, Lpez M, Grriz JM, Ramirez J, Salas-Gonzalez D, Alvarez
I. NMF-SVM based CAD tool applied to functional brain images for
the diagnosis of Alzheimer's disease. IEEE Transactions on medical
imaging. 2011 Sep 12;31(2):207-16.
Gaussian mixture models
- D Reynolds
Reynolds D. Gaussian mixture models. Encyclopedia of biometrics.
2015:827-32.
Automated diagnosis of Alzheimer’s disease using Gaussian mixture model based on cortical thickness
- S Song
- Z Lu
- Pan
A new direction for Alzheimer's research. Neural regeneration research
- J D Weinstein
Weinstein JD. A new direction for Alzheimer's research. Neural regeneration research. 2018 Feb;13(2):190.
Automated skull stripping in brain MR images
- aruchamy
Brain tumor segmentation using K-means clustering and fuzzy C-means algorithms and its area calculation and disease prediction using naive-bayes algorithm
- mandhare