Exclusive feature selection and multi-view learning for Alzheimer’s Disease

BACKGROUND: Alzheimer’s Disease (AD) is normally assessed in clinical settings using neuropsychological tests and medical procedures such as neuroimaging techniques: Magnetic Resonance Imaging (MRI) and Positron Emission Tomography (PET) among others. The latter procedures are expensive and unavailable in most nations, so early diagnosis of AD does not occur, which heavily increases the subsequent treatment costs for the patients. AIMS: This research aims to evaluate cognitive features related to dementia progression based on neuropsychological tests’ data that are related to the Alzheimer’s Disease Neuroimaging Initiative (ADNI) We utilise data related to two neuropsychological tests including the Clinical Dementia Rating Scale Sum of Boxes (CDR-SOB), and Mini-Mental State Examination (MMSE), to assess advancement of the AD. METHODS: To achieve the aim, we develop a data process called Neuropsychological Feature Assessment via Feature Selection (NFAFS) to identify impactful features using Information Gain (IG) and Pearson Correlation to assess class-feature and feature-feature correlations Later we will model a minimal subset of neuropsychological features using machine learning techniques to derive classification models. RESULTS AND IMPLICATIONS: Results obtained show key cognitive features of the MMSE are Time Orientation, Recall and Complex Attention, since they correlate with the progression class being ranked high in results of the feature selection techniques. For the CDR-SOB features, and aside from the memory feature it was difficult to identify other specific features that are signs of the dementia progression Clinicians can use specific features in a digital knowledge base to pay more attention to specific cognitive deficits related to Recall, Orientation and Complex Attention during dementia clinical evaluations in order to seek possible signs of the disease progression early.

In this work, we propose a novel Feature Selection framework, called Sparse-Modeling Based Approach for Class Specific Feature Selection (SMBA-CSFS), that simultaneously exploits the idea of Sparse Modeling and Class-Specific Feature Selection. Feature selection plays a key role in several fields (e.g., computational biology), making it possible to treat models with fewer variables which, in turn, are easier to explain, by providing valuable insights on the importance of their role, and might speed the experimental validation up. Unfortunately, also corroborated by the no free lunch theorems, none of the approaches in literature is the most apt to detect the optimal feature subset for building a final model, thus it still represents a challenge. The proposed feature selection procedure conceives a two steps approach: (a) a sparse modeling-based learning technique is first used to find the best subset of features, for each class of a training set; (b) the discovered feature subsets are then fed to a class-specific feature selection scheme, in order to assess the effectiveness of the selected features in classification tasks. To this end, an ensemble of classifiers is built, where each classifier is trained on its own feature subset discovered in the previous phase, and a proper decision rule is adopted to compute the ensemble responses. In order to evaluate the performance of the proposed method, extensive experiments have been performed on publicly available datasets, in particular belonging to the computational biology field where feature selection is indispensable: the acute lymphoblastic leukemia and acute myeloid leukemia, the human carcinomas, the human lung carcinomas, the diffuse large B-cell lymphoma, and the malignant glioma. SMBA-CSFS is able to identify/retrieve the most representative features that maximize the classification accuracy. With top 20 and 80 features, SMBA-CSFS exhibits a promising performance when compared to its competitors from literature, on all considered datasets, especially those with a higher number of features. Experiments show that the proposed approach might outperform the state-of-the-art methods when the number of features is high. For this reason, the introduced approach proposes itself for selection and classification of data with a large number of features and classes.

Identifying important biomarkers to improve disease diagnosis and treatment is a significant topic of research in bioinformatics. However, bioinformatics datasets frequently have a large number of features per sample or instance. This problem, known as “high dimensionality,” can be alleviated through the use of dimension reducing techniques such as feature (gene) selection which remove unnecessary features. There are many versions of feature selection, with varying biases and predictive abilities. However, predictive power is but one factor to consider when choosing a feature selection technique: one must also consider the technique's stability, that is, its ability to create feature subsets which remain valid in the face of changes to the data. While there has been work in determining the relative stability of different feature selection techniques, this does not always help determine whether a chosen feature selection technique will give stable feature subsets for a specific dataset. Factors such as difficulty of learning (e.g., dataset difficulty) may also influence feature selection stability, making generally-true facts about different techniques not applicable to a given dataset. In this work, we study how dataset difficulty can affect the stability of feature selection techniques, leading to good performance from bad techniques and vice versa. We use a set of twenty-six DNA microarray datasets with varying levels of difficulty of learning, along with four levels of dataset perturbation, six feature selection techniques with various levels of stability, and twelve feature subset sizes. The results show that as the dataset difficulty increases, the stability decreases. However, the relative stability between the techniques remains the same. Additionally, the more difficult the dataset, the more the stability is affected by changes to the data. We also found that unstable rankers are more affected by the transition between Easy and Moderate datasets, whereas the stable techniques are more affected by the change between Moderate and Hard datasets. Lastly, as the feature subset size increases, the stability increases and the difference between the levels of dataset difficulty decreases. Overall, we conclude that difficulty of learning must be taken into account before interpreting stability results.

The most common deadly neoplasm of the female reproductive organ is exhibited by ovarian cancer. The microarray data offers valuable diagnostic and prognostic data for ovarian cancer studies. DNA microarray technology contains high dimensions with small sample size. In high- dimensional microarray data, all genes are not related to hereditary disease. However, in diagnosing disease, these uninformative genes might lead to the misleading results. Identifying a small set of features is very essential as, an efficient and robust feature selection method can achieve significant classification accuracy. This paper proposes an Integrated Feature selection method utilizing Evolutionary Non- Dominated Radial Slots Based Algorithm (ENORA) and Consistency subset evaluation (CSE) stochastic optimization algorithms that can select the highly significant features. This proposed study used ENORA to decrease redundant features and CSE was used to discover the best smallest significant feature subset from the redundant features. The Artificial Neural Network classification model was utilized to evaluate Ovarian Cancer benchmark dataset. The outcomes of the experiment were contrasted with the conventional feature selection model of other literature. Particularly, this proposed approach attained 99.20% higher classification accuracy rate than previous study for ovarian cancer datasets with a minimal sized predictive feature subset.

How far is arterial spin labeling MRI from a clinical reality? Insights from arterial spin labeling comparative studies in Alzheimer's disease and other neurological disorders.

In many pattern recognition applications feature selection and instance selection can be used as two data preprocessing methods that aim at reducing the computational cost of the learning process. Moreover, in some cases, feature subset selection can improve the classification performance. Feature selection and instance selection can be interesting since the choice of features and instances greatly influence the performance of the learnt models as well as their training costs. In the past, unifying both problems was carried out by solving a global optimization problem using meta-heuristics. This paradigm not only does not exploit the manifold structure of data but can be computationally expensive. To the best of our knowledge, the joint use of sparse modeling representative and feature subset relevance have not been exploited by the joint feature and selection methods. In this paper, we target the joint feature and instance selection by adopting feature subset relevance and sparse modeling representative selection. More precisely, we propose three schemes for the joint feature and instance selection. The first is a wrapper technique while the two remaining ones are filter approaches. In the filter approaches, the search process adopts a genetic algorithm in which the evaluation is mainly given by a score that quantify the goodness of the features and instances. An efficient instance selection technique is used and integrated in the search process in order to adapt the instances to the candidate feature subset. We evaluate the performance of the proposed schemes using image classification where classifiers are the nearest neighbor classifier and support vector machine classifier. The study is conducted on five public image datasets. These experiments show the superiority of the proposed schemes over various baselines. The results confirm that the filter approaches leads to promising improvement on classification accuracy when both feature selection and instance selection are adopted.

High-order correlation detecting in features for diagnosis of Alzheimer’s disease and mild cognitive impairment

(Reprinted with permission from American Psychiatric Association, http://psychiatryonline.org/guidelines).

Gene selection and classification using Taguchi chaotic binary particle swarm optimization

In fuzzy C-means (FCM) clustering, each data point belongs to a cluster to a degree specified by a membership grade. FCM partitions a collection of vectors in c fuzzy groups and finds a cluster center in each group such that the dissimilarity measure is minimized. This paper presents a training algorithm for the radial basis function (RBF) network using symmetry-based Fuzzy C-means (SFCM) clustering method which is the modified version of FCM clustering method based on point symmetry distance measure. The training algorithm which uses SFCM clustering method to train the network has a number of advantages such as faster training time, more accurate predictions and reduced network architecture compared to the standard RBF networks. The proposed training algorithm has been implemented in the RBF networks created by the newrb function of MATLAB which uses gradient based iterative method as learning strategy, therefore the new network will undergo a hybrid learning process. The networks called Symmetry-based Fuzzy C-means Clustering–Radial Basis Function Network (SFCM/RBF) has been tested against the standard RBF network and the networks called standard Fuzzy C-means Clustering (FCM)-RBF network (FCM/RBF) in forecasting. The experimental models has been tested on three real world application problems, particularly in Air pollutant problem, Biochemical Oxygen Demand (BOD) problem, and Phytoplankton problem. Keywords: Fuzzy c-means clustering; SFCM; Radial basis function network; point symmetry distance; forecasting.

• Neural Network feature selection methodology for both numerical and categorical data. • Embedded feature selection provides model and feature subset in a single training. • Novel feature embedding for dealing with high-cardinality categorical features. • Validation of the methodology on several open and real industry dataset. • Comparison with several state-of-the-art feature selection methodologie. In an era of effortless data collection, the impact of machine learning — especially neural networks (NNs) — is undeniable. As datasets grow in size and complexity, efficiently handling mixed data types, including categorical and numerical features, becomes critical. Feature encoding and selection play a key role in improving NN performance, efficiency, interpretability, and generalisation. This paper presents GLEm-Net (Grouped Lasso with Embeddings Network), a novel NN-based approach that seamlessly integrates feature encoding and selection directly into the training process. GLEm-Net uses embedding layers to process categorical features with high cardinality, simplifying the model and improving generalisation. By extending the grouped Lasso regularisation to explicitly consider categorical features, GLEm-Net automatically identifies the most relevant features during training and returns them to the analyst. We evaluate GLEm-Net on open and proprietary industry datasets and compare it to state-of-the-art feature selection methodologies. Results show that GLEm-Net adapts to each dataset by allowing the NN to directly select subsets of most important features, offering on par performance with the best state-of-the-art feature selection methods, while eliminating the need for the external feature encoding and selection steps that are now incorporated in the NN training stage.

A new hybrid filter/wrapper algorithm for feature selection in classification

The most prevalent cause of dementia in older people is Alzheimer's disease (AD). Dementia is a neurological disease that severely limits a person's capacity to do everyday tasks. The importance of early Alzheimer's disease diagnosis can be attributed to several factors. Most of them provide for Alzheimer's disease therapies that can slow down the disease's development. The CT scan reveals a degree of generalized cortical atrophy in Alzheimer's disease patients. As a result, CT scan picture processing is critical in the early detection of Alzheimer's disease. Here, image processing is used to detect the objects in CT pictures. Edge detection is a critical first phase in image processing since it defines the discontinuities in gray-level images. The majority of them are clinic-based structural MRI images with small sample size and few scanning layers. Deep learning, on the other hand, necessitates a large amount of annotated details. This paper suggests a dataset increment approach based on a weighted mixture of positive and negative tests and a learning method with a limited number of samples to meet the realistic requirements of clinical evaluation of Alzheimer's disease. It produces a GKFCM Clustering model that can collect more image feature details and boost the model's generalization ability.

Feature selection entails identifying a subset of the most useful features that produces compatible results as the original entire set of features. A feature selection algorithm can be assessed in terms of both efficiency and effectiveness. While efficiency is concerned with the time required to find a subset of features, effectiveness is concerned with the quality of the subset of features. This paper proposes and experimentally evaluates a fast clustering-based feature selection algorithm, FAST, based on these criteria.  The FAST algorithm operates in two steps. Graph-theoretic clustering methods are used to partition characteristics into clusters in the initial stage. The most representative feature from each cluster that is strongly related to target classes is chosen in the second stage to construct a subset of features. Because the properties in various clusters are relatively independent, FAST's clustering-based technique is likely to produce a subset of valuable and independent features. We use the efficient Minimum-spanning tree clustering method to ensure FAST's efficiency. An empirical study is conducted to assess the efficiency and effectiveness of the FAST algorithm. FAST and several representative feature selection algorithms, such as FCBF, ReliefF, CFS, Consist, and FOCUS-SF, are compared to four types of well-known classifiers, including the probability-based Naive Bayes, the tree-based C4.5, the instance-based IB1, and the rule-based RIPPER, before and after feature selection. FAST not only provides smaller subsets of features but also improves the performances of the four types of classifiers, according to the findings, which were based on 35 publicly accessible real-world high-dimensional image, microarray, and text data.

Manifold regularized multi-view feature selection for social image annotation