A new medical decision making system: Least square support vector machine (LSSVM) with Fuzzy Weighting Pre-processing

The use of machine learning tools in medical diagnosis is increasing gradually. This is mainly because the effectiveness of classification and recognition systems has improved in a great deal to help medical experts in diagnosing diseases. This study aims at diagnosing Liver Disorder with a new hybrid machine learning method. By hybridizing LSSVM with Fuzzy Weighting Pre-processing, a method was obtained to solve this diagnosis problem via classifying Liver Disorder. Fuzzy Weighting Pre-processing stage was developed firstly in our study. This Liver Disorder dataset is a very commonly used dataset in literature relating the use of classification systems for Liver Disorder Diagnosis and it was used in this study to compare the classification performance of our proposed method with regard other studies. We obtained a classification accuracy of 94.29%, which is the highest one reached so far. This result is for Liver Disorder but it states that this method can be used confidently for other medical diseases diagnosis problems, too.     

Classification of BMD and ADHD patients using their EEG signals

Bipolar Mood Disorder (BMD) and Attention Deficit Hyperactivity Disorder (ADHD) patients mostly share clinical signs and symptoms in children; therefore, accurate distinction of these two mental disorders is a challenging issue among the psychiatric society. In this study, 43 subjects are participated including 21 patients with ADHD and 22 subjects with BMD. Their electroencephalogram (EEG) signals are recorded by 22 electrodes in two eyes-open and eyes-closed resting conditions. After a preprocessing step, several features such as band power, fractal dimension, AR model coefficients and wavelet coefficients are extracted from recorded signals. This paper is aimed to achieve a high classification rate between ADHD and BMD patients using a suitable classifier to their EEG features. In this way, we consider a piece wise linear classifier which is designed based on XCSF. Experimental results of XCSF-LDA showed a significant improvement (86.44% accuracy) compare to that of standard XCSF (78.55%). To have a fair comparison, the other state-of-art classifiers such as LDA, Direct LDA, boosted JD-LDA (BJDLDA), and XCSF are assessed with the same feature set that finally the proposed method provided a better results in comparison with the other rival classifiers. To show the robustness of our method, additive white noise with different amplitude is added to the raw signals but the results achieved by the proposed classifier empirically confirmed a higher robustness against noise compare to the other classifiers. Consequently, the proposed classifier can be considered as an effective method to classify EEG features of BMD and ADHD patients.     

Conditional Generative Adversarial Network Approach for Autism Prediction

Autism Spectrum Disorder (ASD) requires a precise diagnosis in order to be managed and rehabilitated. Non-invasive neuroimaging methods are disease markers that can be used to help diagnose ASD. The majority of available techniques in the literature use functional magnetic resonance imaging (fMRI) to detect ASD with a small dataset, resulting in high accuracy but low generality. Traditional supervised machine learning classification algorithms such as support vector machines function well with unstructured and semi structured data such as text, images, and videos, but their performance and robustness are restricted by the size of the accompanying training data. Deep learning on the other hand creates an artificial neural network that can learn and make intelligent judgments on its own by layering algorithms. It takes use of plentiful low-cost computing and many approaches are focused with very big datasets that are concerned with creating far larger and more sophisticated neural networks. Generative modelling, also known as Generative Adversarial Networks (GANs), is an unsupervised deep learning task that entails automatically discovering and learning regularities or patterns in input data in order for the model to generate or output new examples that could have been drawn from the original dataset. GANs are an exciting and rapidly changing field that delivers on the promise of generative models in terms of their ability to generate realistic examples across a range of problem domains, most notably in image-to-image translation tasks and hasn't been explored much for Autism spectrum disorder prediction in the past. In this paper, we present a novel conditional generative adversarial network, or cGAN for short, which is a form of GAN that uses a generator model to conditionally generate images. In terms of prediction and accuracy, they outperform the standard GAN. The proposed model is 74% more accurate than the traditional methods and takes only around 10 min for training even with a huge dataset.     

Using Multilayer Fuzzy Cognitive Maps to diagnose Autism Spectrum Disorder

Autism Spectrum Disorder (ASD) is comprised of a group of heterogeneous neurodevelopmental conditions, typically characterized by a triad of symptoms consisting of (1) impaired communication, (2) restricted interests, and (3) repetitive and stereotypical behavior pattern. An accurate and early diagnosis of autism can provide the basis for an appropriate educational and treatment program. In this work, we propose a computational model using a Multilayer Fuzzy Cognitive Map (hereafter referred to as MFCM) based on standardized behavioral assessments diagnosing the ASD (MFCM-ASD). The two standards used in the model are: the Autism Diagnostic Observation Schedule, Second Edition (ADOS2), and the Autism Diagnostic Interview Revised (ADIR). The MFCM’s are a soft computing technique characterized by robust properties that make it an effective technique for medical decision support systems. For the evaluation of the MFCM-ASD model, we have used real datasets of diagnosed cases, so as to compare against other method/approaches. Initial experiments demonstrated that the proposed model outperforms conventional Fuzzy Cognitive Maps (FCMs) for ASD diagnosis. Our MFCM-ASD model serves as a diagnostic tool required to support the medical decisions when determining the correct diagnosis of Autism in children with different cognitive characteristics.     

Action recognition using saliency learned from recorded human gaze

This paper addresses the problem of recognition and localization of actions in image sequences, by utilizing, in the training phase only, gaze tracking data of people watching videos depicting the actions in question. First, we learn discriminative action features at the areas of gaze fixation and train a Convolutional Network that predicts areas of fixation (i.e. salient regions) from raw image data. Second, we propose a Support Vector Machine-based recognition method for joint recognition and localization, in which the bounding box of the action in question is considered as a latent variable. In our formulation the optimization attempts to both minimize the classification cost and maximize the saliency within the bounding box. We show that the results obtained with the optimization where saliency within the bounding box is maximized outperform the results obtained when saliency within the bounding box is not maximized, i.e. when only classification cost is minimized. Furthermore, the results that we obtain outperform the state-of-the-art results on the UCF sports dataset.     

基于语音停顿度和平坦度的注意缺陷与多动障碍自动检测算法

针对注意缺陷与多动障碍（ADHD）临床诊断主要依靠医生主观评估,缺乏客观辅助依据的问题,提出了一种基于语音停顿度和平坦度的ADHD自动检测算法。首先,通过频带差能熵积（FDEEP）参数自动定位语音有话区间,并提取停顿度特征;然后,使用变换平均幅度平方差（TAASD）参数计算语音倍频率,并提取平坦度特征;最后,结合融合特征和支持向量机（SVM）分类器来实现ADHD的自动识别。实验共采集了17位正常对照组儿童和37位ADHD患儿的语音样本。实验结果表明,所提算法能自动检测正常儿童和ADHD患儿,识别正确率为91.38%。     

基于图神经网络的自闭症辅助诊断研究综述

自闭症谱系障碍是一种复杂的神经系统发展障碍疾病,截至目前其病因尚不明确。图神经网络作为非欧几里得空间深度学习的重要分支,在处理图结构数据的相关任务中取得优异表现,为医学领域的成像和非成像模式的集成提供了可能,因此利用图神经网络进行自闭症等脑部疾病神经成像诊断逐渐成为研究热点。阐述传统机器学习方法在自闭症疾病预测中应用,介绍图神经网络的基本分类,按照图中节点与边关系的建模方法,从基于人群图和基于个体图两个角度对图神经网络在自闭症辅助诊断中的应用进行梳理和分析,并归纳现有诊断方法的优劣势。根据目前基于图神经网络的自闭症神经成像诊断的研究现状,总结了脑神经科学领域辅助诊断技术面临的主要挑战和未来研究方向,对于自闭症等脑部疾病辅助诊断的进一步研究具有指导意义和参考价值。     

Fairy tale directed game-based training system for children with ADHD using BCI and motion sensing technologies

We developed a game-based training system to analyse and improve the reading ability of children with Attention Deficit/Hyperactivity Disorder (ADHD). A fairy tale-based interactive narrative is used as an intervention strategy in the behaviour training system, which collects brainwaves and motion-sensing data during treatment. The system includes fairy tales as well as attention and behaviour-related tasks coupled with a brain-computer interface (BCI) and motion-sensing technology. The treatment for the children (N?=?5) diagnosed with ADHD was performed for five weeks on a weekly basis, comprised of one 20-minute long adaptation session and four 40-minute long sessions. For the quantitative analysis of the treatment, pre- and post- KNISE-BAAT and general reading questionnaires were administered. Sensing data was also recorded. In-depth post-interviews were conducted after the completion of the treatment programme for qualitative analysis. The paired-samples t-test on both reading comprehension tests indicate improvement in both reading aloud and reading comprehension. The sensing data analysis shows improvements in attention span and decreases in hyperactive behaviour over time. The analysis on the interview data supports the quantitative test results. As such, the test results indicate that this approach helps children with ADHD improve their reading ability, increase their attention span, and support behavioural inhibition.     

Assessing the importance of autistic attributes for autism screening

Autistic Spectrum Disorder (ASD) is a cognitive disease which leads to the loss of linguistic, communicative, cognitive, and social skills and abilities. Patients with ASD have diverse troubles such as sleeping problems. The role of genetic and environmental factors is of great importance in its pathophysiology. Early diagnosis provides an improved overall mental health of the patients. This study aimed to identify the important attributes for the best detection of this disorder in children, adolescents and adults. To achieve this aim, Recursive Feature Elimination and Stability Selection methods that consider important attributes for target class were proposed. The attributes collected from autism screening methods and other attributes such as age and gender were examined for the disease. The results verified the combining of feature selection method and machine learning algorithm within the frame of accuracy, sensitivity and specificity evaluation metrics.     

Physiology-based affect recognition for computer-assisted intervention of children with Autism Spectrum Disorder

Generally, an experienced therapist continuously monitors the affective cues of the children with Autism Spectrum Disorders (ASD) and adjusts the course of the intervention accordingly. In this work, we address the problem of how to make the computer-based ASD intervention tools affect-sensitive by designing therapist-like affective models of the children with ASD based on their physiological responses. Two computer-based cognitive tasks are designed to elicit the affective states of liking, anxiety, and engagement that are considered important in autism intervention. A large set of physiological indices are investigated that may correlate with the above affective states of children with ASD. In order to have reliable reference points to link the physiological data to the affective states, the subjective reports of the affective states from a therapist, a parent, and the child himself/herself were collected and analyzed. A support vector machines (SVM)-based affective model yields reliable prediction with approximately 82.9% success when using the therapist's reports. This is the first time, to our knowledge, that the affective states of children with ASD have been experimentally detected via physiology-based affect recognition technique.     

基于皮尔逊最优电极选择的ADHD患者脑电特征提取及分类研究

事件相关电位(ERP)可用于注意缺陷多动障碍儿童(ADHD)和正常儿童的脑电特征 提取与分类。首先,采用赌博任务范式,采集2 类儿童的脑电信号;其次,基于皮尔逊相关系 数算法选择最优电极,并预处理最优电极脑电信号;然后,提取预处理脑电信号的时域特征(均 值、方差、峰值)和频域特征(Theta 波段功率、Alpha 波段功率);最后,利用传统分类方法支持 向量机(SVM)、自适应增强(AdaBoost)、自举汇聚法(Bagging)、线性判别式分析(LDA)、反向传 播(BP)和组合分类器的分类方法(LDA-SVM,BP-SVM)完成对2 种脑电信号的分类。研究结果 表明,传统方法BP 分类器的分类准确率可达80.52%,组合分类器BP-SVM 的分类准确率可达 88.88%。组合分类方法能提高ADHD 儿童的分类准确率,为基于脑机接口技术的ADHD 神经 反馈康复治疗提供技术支持。     

Virtual Reality Social Cognition Training for children with high functioning autism

Virtual reality appears to be a promising and motivating platform to safely practice and rehearse social skills for children with Autism Spectrum Disorders (ASD). However, the literature to date is subject to limitations in elucidating the effectiveness of these virtual reality interventions. This study investigated the impact of a Virtual Reality Social Cognition Training to enhance social skills in children with ASD. Thirty children between the ages of 7–16 diagnosed with ASD completed 10, 1-h sessions across 5 weeks. Three primary domains were measured pre-post: emotion recognition, social attribution, attention and executive function. Results revealed improvements on measures of emotion recognition, social attribution, and executive function of analogical reasoning. These preliminary findings suggest that the use of a virtual reality platform offers an effective treatment option for improving social impairments commonly found in ASD.     

Test feature classifiers: performance and applications

In this paper, we present a class of test feature classifiers (TFCs). We discuss the properties and performance of the proposed classifiers and describe cases when a 100% recognition rate on test data can be achieved. When the number of features increases, classes having no more than a polynomial number of instances (in the number of features) are the only cases possible to process. We prove that for almost all pairs of classes with a polynomial number of instances, a 100% recognition rate on any test data can be achieved. To test the performance of the classifiers, we apply them to both artificial and real data. For the real data, we use the well known breast cancer, phoneme, and satimage databases, which are recognized to be difficult classification problems. Our experimental results show that the proposed classifiers not only have a high recognition ability but, also, the ability to achieve a 100% recognition rate in difficult classification problems.     

Stockwell transform of time-series of fMRI data for diagnoses of attention deficit hyperactive disorder

Attention deficit hyperactivity disorder (ADHD) is a common brain disorder among children. It presents various symptoms, hence, utilizing the information obtained from functional magnetic resonance imaging (fMRI) time-series data can be useful. Finding functional connections in typically developed control (TDC) and ADHD patients can be helpful in classification. The aim of this paper is to present a multifold method for the study of fMRI data to diagnose ADHD patients. In the proposed method, first, by applying the Stockwell transform (ST), we obtain detailed information about the time-series of the region of interests (ROIs) in the time and frequency domains. ST provides information about the variations of each ROI during the time. Thereafter, time-frequency domains are partitioned into sub-matrices and then, their fuzzy entropies are calculated as features. Next, discriminative features are chosen by using the two-sample Kolmogorov–Smirnov (K–S) test. Finally, the data are classified by the leave-one-out cross-validation (LOOCV) method using the support vector machine (SVM) classifier. To see the effectiveness of the proposed method, the experiments are performed on the ADHD-200 database. We consider different scenarios including classification of TDCs and ADHDs as well as classification of ADHD subtypes. We also assess the performance by considering the age and sex as phenotypic information. The proposed method gives good results in the classification procedure and identifying the connection paths between ROIs. The results indicate that the proposed method can distinguish ADHD disorder in a more accurate manner in comparison with other methods. The connectivity paths show that there is a reduction in the input of cerebellar regions and the left mid orbitofrontal cortex in ADHDs compared to TDCs.     

Impact of mainstream classroom setting on attention of children with autism spectrum disorder: an eye-tracking study

It has long been reported that children with autism spectrum disorder (ASD) exhibit attention difficulties while learning. They tend to focus on irrelevant information and can easily be distracted. As a result, they are often confined to a one-to-one teaching environment, with fewer distractions and social interactions than would be present in a mainstream educational setting. In recent years, inclusive mainstream schools have been growing in popularity due to government policies on equality rights. Therefore, it is crucial to investigate attentional patterns of children with ASD in mainstream schools. This study aims to explore the attentional behaviors of children with ASD in a virtual reality simulated classroom. We analyzed four eye-gaze behaviors and performance scores of 45 children: children with ASD (ASD n = 20) and typically developing children (TD n = 25) when performing attention tasks. The gaze behaviors included time to first fixate (TTFF), first fixation duration (FFD), average fixation duration (AFD) and the sum of fixation count (SFC) on fourteen areas of interest (AOIs) in the classroom. Our results showed that children with ASD exhibit similar gaze behaviors to TD children, but with significantly lower performance scores and SFC on the target AOI. These findings showed that classroom settings can influence attentional patterns and the academic performance of children with ASD. Further studies are needed on different modalities for supporting the attention of children with ASD in a mainstream setting.

     

Online Affect Detection and Robot Behavior Adaptation for Intervention of Children With Autism

Investigation into robot-assisted intervention for children with autism spectrum disorder (ASD) has gained momentum in recent years. Therapists involved in interventions must overcome the communication impairments generally exhibited by children with ASD by adeptly inferring the affective cues of the children to adjust the intervention accordingly. Similarly, a robot must also be able to understand the affective needs of these children—an ability that the current robot-assisted ASD intervention systems lack—to achieve effective interaction that addresses the role of affective states in human–robot interaction and intervention practice. In this paper, we present a physiology-based affect-inference mechanism for robot-assisted intervention where the robot can detect the affective states of a child with ASD as discerned by a therapist and adapt its behaviors accordingly. This paper is the first step toward developing “understanding” robots for use in future ASD intervention. Experimental results with six children with ASD from a proof-of-concept experiment (i.e., a robot-based basketball game) are presented. The robot learned the individual liking level of each child with regard to the game configuration and selected appropriate behaviors to present the task at his/her preferred liking level. Results show that the robot automatically predicted individual liking level in real time with 81.1% accuracy. This is the first time, to our knowledge, that the affective states of children with ASD have been detected via a physiology-based affect recognition technique in real time. This is also the first time that the impact of affect-sensitive closed-loop interaction between a robot and a child with ASD has been demonstrated experimentally.      

基于深度学习的卫星图像识别分类方法

卫星遥感技术是一种非常重要的地球空间监测技术.卫星遥感图像经过处理后具有数据量大和数据类型复杂多样的特点,传统方法进行识别分类耗费大量人力物力.为了降低工作量,并为后续处理提供便利,本文将深度学习算法应用于卫星图像的识别分类中,设计了一种基于VGGNet的识别分类方法,利用除雾算法对训练数据进行数据增强处理,并添加岭回归正则化层,利用标签之间的相关性进行预测,使得方法达到90%以上的F2 score,并在实验部分进行了对比验证.最后利用此方法搭建了一个基于Django的在线识别分类展示系统.     

基于可视化异类特征优选融合的滚动轴承故障诊断

针对故障诊断过程中基于简单的多类故障特征联合决策存在特征集维数多、数据冗余、故障识别率不高的缺点,提出了一种基于异类特征优选融合的故障诊断方法。该方法根据多类特征数据的轮廓图,分析各维特征数据的聚类特性,去除聚类性弱、对故障区分无益的冗余特征维度,仅保留聚类性强的特征维度用于故障识别。在轴承故障诊断实验中,选用故障信号时域统计量和小波包能量两类多维特征进行优选融合,并采用反向传播（BP）神经网络进行故障模式识别。故障识别率达到100%,显著高于无特征优选的故障诊断方法。实验结果表明所提出的方法简便易行,可以显著提高故障识别率。     

基于最优传输的多中心自闭症谱系障碍诊断

融合来自多个中心的医学数据能够增加样本数量，有助于研究自闭症谱系障碍(Autism spectrum disorder, ASD)的病理变化。因此，如何有效地利用多中心数据，提高对ASD诊断的准确性受到了越来越多的关注。然而，以往的大部分研究忽略了多中心数据的异质性(如受试者群体和扫描参数的不同)，这可能会降低模型在多中心数据上对疾病诊断的性能。为了解决这一问题，提出一种基于联合分布最优传输(Joint distribution optimal transport, JDOT)的领域自适应模型鉴别ASD。选择一个中心作为目标域，其余的中心作为源域，假设两个域的联合特征、标签空间分布之间存在非线性映射，利用最优传输方法交替优化传输矩阵和分类器。结果表明，在多中心静息态功能磁共振成像(resting state functional magnetic resonance imaging, rs-fMRI)数据上，该模型能够有效提高对ASD鉴别的准确性。     

Multi-class pattern classification using neural networks

Multi-class pattern classification has many applications including text document classification, speech recognition, object recognition, etc. Multi-class pattern classification using neural networks is not a trivial extension from two-class neural networks. This paper presents a comprehensive and competitive study in multi-class neural learning with focuses on issues including neural network architecture, encoding schemes, training methodology and training time complexity. Our study includes multi-class pattern classification using either a system of multiple neural networks or a single neural network, and modeling pattern classes using one-against-all, one-against-one, one-against-higher-order, and P-against-Q. We also discuss implementations of these approaches and analyze training time complexity associated with each approach. We evaluate six different neural network system architectures for multi-class pattern classification along the dimensions of imbalanced data, large number of pattern classes, large vs. small training data through experiments conducted on well-known benchmark data.