Recording and Analyzing High-Density Event-Related Potentials With Infants Using the Geodesic Sensor Net

This article provides an overview of the use of the Geodesic sensor net system for high-density event-related potential (ERP) recording in infants. Some advantages and disadvantages of the system, as applied to infants, are discussed. First, we illustrate that high-density data can be recorded from infants at comparable quality to that observed with conventional (low density) ERP methods. Second, we discuss ways to utilize the greater spatial information available by applying source separation and localization procedures. In particular, we focus on the application of one recent source separation method, Independent Component Analysis (ICA). Finally, we show that source localization can be applied to infant high-density data, although this entails adopting a number of assumptions that remain to be verified. In the future, with improved source separation algorithms, we suggest that single-trial or single-subject analyses may become feasible.     

Recording and Analyzing High-Density Event-Related Potentials With Infants Using the Geodesic Sensor Net

This article provides an overview of the use of the Geodesic sensor net system for high-density event-related potential (ERP) recording in infants. Some advantages and disadvantages of the system, as applied to infants, are discussed. First, we illustrate that high-density data can be recorded from infants at comparable quality to that observed with conventional (low density) ERP methods. Second, we discuss ways to utilize the greater spatial information available by applying source separation and localization procedures. In particular, we focus on the application of one recent source separation method, Independent Component Analysis (ICA). Finally, we show that source localization can be applied to infant high-density data, although this entails adopting a number of assumptions that remain to be verified. In the future, with improved source separation algorithms, we suggest that single-trial or single-subject analyses may become feasible.     

Machine learning for survival analysis: a case study on recurrence of prostate cancer

Machine learning techniques have recently received considerable attention, especially when used for the construction of prediction models from data. Despite their potential advantages over standard statistical methods, like their ability to model non-linear relationships and construct symbolic and interpretable models, their applications to survival analysis are at best rare, primarily because of the difficulty to appropriately handle censored data. In this paper we propose a schema that enables the use of classification methods--including machine learning classifiers--for survival analysis. To appropriately consider the follow-up time and censoring, we propose a technique that, for the patients for which the event did not occur and have short follow-up times, estimates their probability of event and assigns them a distribution of outcome accordingly. Since most machine learning techniques do not deal with outcome distributions, the schema is implemented using weighted examples. To show the utility of the proposed technique, we investigate a particular problem of building prognostic models for prostate cancer recurrence, where the sole prediction of the probability of event (and not its probability dependency on time) is of interest. A case study on preoperative and postoperative prostate cancer recurrence prediction shows that by incorporating this weighting technique the machine learning tools stand beside modern statistical methods and may, by inducing symbolic recurrence models, provide further insight to relationships within the modeled data.     

Topographical Subcomponents of Electrical Brain Activity Allow to Identify Semantic Learning

We investigated the change of event-related brain activity elicited by reading meaningful or meaningless Japanese symbols in 20 healthy German adults. In a learning phase of about 20 min, subjects acquired the meaning of 20 Kanji characters. As control stimuli 20 different Kanji characters were presented. Electrical brain activity was obtained before and after learning, The mean learning performance of all subjects was 92.5% correct responses. EEG was measured simultaneously from 30 channels, artifacts were removed offline, and the data before and after learning were compared. We found five spatial principal components that accounted for 83.8% of the variance. A significant interaction between training time (before/after learning) and stimulus (learning/control) illustrates a significant relation between successful learning and topographical changes of brain activity elicited by Kanji characters. Effects that were induced by learning were seen at short latencies in the order of 100 ms. In addition, we present evidence that differences in the weighted combination of spatial components allow to identify experimental conditions successfully by linear discriminant analysis using topographical ERP data of a single time point. In conclusion, semantic meaning can be aquired rapidly and it is associated with specific changes of ERP components.     

A primer on gene expression and microarrays for machine learning researchers

Data originating from biomedical experiments has provided machine learning researchers with an important source of motivation for developing and evaluating new algorithms. A new wave of algorithmic development has been initiated with the publication of gene expression data derived from microarrays. Microarray data analysis is particularly challenging given the large number of measurements (typically in the order of thousands) that are reported for relatively few samples (typically in the order of dozens). Many data sets are now available on the web. It is important that machine learning researchers understand how data are obtained and which assumptions are necessary in the analysis. Microarray data have the potential to cause significant impact in machine learning research, not just as a rich and realistic source of cases for testing new algorithms, as has been the UCI machine learning repository in the past decades, but also as a main motivation for their development. In this article, we briefly review the biology underlying microarrays, the process of obtaining gene expression measurements, and the rationale behind the common types of analyses involved in a microarray experiment. We outline the main challenges and reiterate critical considerations regarding the construction of supervised learning models that use this type of data. The goal of this article is to familiarize machine learning researchers with data originated from gene expression microarrays.     

Visual event-related brain potentials in 4-month-old infants at risk for neurodevelopmental impairments

The recording of event-related potentials (ERPs) is an electrophysiologic technique that has been used to evaluate the functional maturation of neural pathways responsible for recognition memory systems in infants and children. The purpose of this study was to evaluate ERP correlates of visual recognition memory in 4-month-old infants at risk for later cognitive impairments. We compared ERPs using a test of shape recognition at 4 months of age (adjusted for prematurity) in 16 high-risk, neonatal intensive care unit (NICU) survivors and 16 healthy full-term infants. ERPs were recorded while infants were familiarized with one stimulus (a red cross, 15 trials), then tested with 60 trials of this familiar stimulus and a novel stimulus (a red corkscrew). Both the NICU and control groups' ERPs demonstrated evidence of differential processing of the two stimuli, but the NICU groups' ERP patterns were distinctly different from those of the control group. In the NICU group, the novel stimulus elicited parietal positivity at 1000–1700 ms poststimulus, whereas in the control group the novel stimulus elicited occipital and frontal negativity at 500–1700 ms poststimulus. The ERP pattern demonstrated by the NICU group was atypical as it has not been previously described in healthy infants. The results of the study indicate that the ERP technique can be used to demonstrate altered patterns of neural activity during tasks of visual recognition memory in high-risk infants. We speculate that the atypical ERP patterns described in this study may indicate that patterns of synaptic organization were altered by neonatal events. © 1997 John Wiley & Sons, Inc. Dev Psychobiol 30 : 11–28, 1997     

Principal component analyses and scalp distribution of the auditory P150-250 and N250-550 to speech contrasts in Mexican and American infants

We report a Principal Component Analysis (PCA) and the scalp distribution of the normalized peak amplitude values for speech-related auditory Event-related Potentials (ERP) P150-250 and N250-550 in 7-, 11-, and 20-month-old American infants learning English and in 10-13-month-old Mexican infants learning Spanish. After assessing the infant auditory ERP P-N complex using PCA, we evaluated the topographic distribution of each of the discriminatory phases to native and non-native CV-syllabic contrasts used in Spanish and English. We found that the first two Principal Components for each contrast type across ages showing a maximization of differences between the P150-250 and the N250-550 waves, explain more than 70% of the variance. The scalp distributions of the P150-250 and N250-550 components also differed, the P150-250 showing a frontal and anterior temporal distribution, and the N250-550 a more posterior distribution. The older infants showed a broader distribution of responses, particularly for the N250-550. There were no differences in the topographies of the components between same-aged Mexican and American infants. We discuss the perceptual/linguistic functions that each component may reflect during development and across the two cultures.     

Machine Learning EEG to Predict Cognitive Functioning and Processing Speed Over a 2-Year Period in Multiple Sclerosis Patients and Controls

Event-related potentials (ERPs) show promise to be objective indicators of cognitive functioning. The aim of the study was to examine if ERPs recorded during an oddball task would predict cognitive functioning and information processing speed in Multiple Sclerosis (MS) patients and controls at the individual level. Seventy-eight participants (35 MS patients, 43 healthy age-matched controls) completed visual and auditory 2- and 3-stimulus oddball tasks with 128-channel EEG, and a neuropsychological battery, at baseline (month 0) and at Months 13 and 26. ERPs from 0 to 700 ms and across the whole scalp were transformed into 1728 individual spatio-temporal datapoints per participant. A machine learning method that included penalized linear regression used the entire spatio-temporal ERP to predict composite scores of both cognitive functioning and processing speed at baseline (month 0), and months 13 and 26. The results showed ERPs during the visual oddball tasks could predict cognitive functioning and information processing speed at baseline and a year later in a sample of MS patients and healthy controls. In contrast, ERPs during auditory tasks were not predictive of cognitive performance. These objective neurophysiological indicators of cognitive functioning and processing speed, and machine learning methods that can interrogate high-dimensional data, show promise in outcome prediction.     

Principal Component Analyses and Scalp Distribution of the Auditory P150–250 and N250–550 to Speech Contrasts in Mexican and American Infants

We report a Principal Component Analysis (PCA) and the scalp distribution of the normalized peak amplitude values for speech-related auditory Event-related Potentials (ERP) P150–250 and N250–550 in 7-, 11-, and 20-month-old American infants learning English and in 10–13-month-old Mexican infants learning Spanish. After assessing the infant auditory ERP P-N complex using PCA, we evaluated the topographic distribution of each of the discriminatory phases to native and non-native CV-syllabic contrasts used in Spanish and English. We found that the first two Principal Components for each contrast type across ages showing a maximization of differences between the P150–250 and the N250–550 waves, explain more than 70% of the variance. The scalp distributions of the P150–250 and N250–550 components also differed, the P150–250 showing a frontal and anterior temporal distribution, and the N250–550 a more posterior distribution. The older infants showed a broader distribution of responses, particularly for the N250–550. There were no differences in the topographies of the components between same-aged Mexican and American infants. We discuss the perceptual/linguistic functions that each component may reflect during development and across the two cultures.     

A novel artificial neural network method for biomedical prediction based on matrix pseudo-inversion

Biomedical prediction based on clinical and genome-wide data has become increasingly important in disease diagnosis and classification. To solve the prediction problem in an effective manner for the improvement of clinical care, we develop a novel Artificial Neural Network (ANN) method based on Matrix Pseudo-Inversion (MPI) for use in biomedical applications. The MPI-ANN is constructed as a three-layer (i.e., input, hidden, and output layers) feed-forward neural network, and the weights connecting the hidden and output layers are directly determined based on MPI without a lengthy learning iteration. The LASSO (Least Absolute Shrinkage and Selection Operator) method is also presented for comparative purposes. Single Nucleotide Polymorphism (SNP) simulated data and real breast cancer data are employed to validate the performance of the MPI-ANN method via 5-fold cross validation. Experimental results demonstrate the efficacy of the developed MPI-ANN for disease classification and prediction, in view of the significantly superior accuracy (i.e., the rate of correct predictions), as compared with LASSO. The results based on the real breast cancer data also show that the MPI-ANN has better performance than other machine learning methods (including support vector machine (SVM), logistic regression (LR), and an iterative ANN). In addition, experiments demonstrate that our MPI-ANN could be used for bio-marker selection as well.     

A support vector machine using the lazy learning approach for multi-class classification

Support vector machines can be used in a new machine learning technique based on statistical learning. In this paper, we develop least squares support vector machines (LS-SVMs) using the lazy learning approach to classify data in unclassifiable regions in the case of multi-class classification. LS-SVMs use a set of linear equations while SVMs use a quadratic programming problem. The lazy learning approach is a local and memory-based technique. Therefore, it is an alternative technique to fuzzy inference systems. Our studies show that LS-SVMs with the lazy learning approach can give comparable results to fuzzy LS-SVMs for multi-class classification.     

Machine learning in bioinformatics: a brief survey and recommendations for practitioners

Machine learning is used in a large number of bioinformatics applications and studies. The application of machine learning techniques in other areas such as pattern recognition has resulted in accumulated experience as to correct and principled approaches for their use. The aim of this paper is to give an account of issues affecting the application of machine learning tools, focusing primarily on general aspects of feature and model parameter selection, rather than any single specific algorithm. These aspects are discussed in the context of published bioinformatics studies in leading journals over the last 5 years. We assess to what degree the experience gained by the pattern recognition research community pervades these bioinformatics studies. We finally discuss various critical issues relating to bioinformatic data sets and make a number of recommendations on the proper use of machine learning techniques for bioinformatics research based upon previously published research on machine learning.     

A novel ensemble machine learning for robust microarray data classification

Microarray data analysis and classification has demonstrated convincingly that it provides an effective methodology for the effective diagnosis of diseases and cancers. Although much research has been performed on applying machine learning techniques for microarray data classification during the past years, it has been shown that conventional machine learning techniques have intrinsic drawbacks in achieving accurate and robust classifications. This paper presents a novel ensemble machine learning approach for the development of robust microarray data classification. Different from the conventional ensemble learning techniques, the approach presented begins with generating a pool of candidate base classifiers based on the gene sub-sampling and then the selection of a sub-set of appropriate base classifiers to construct the classification committee based on classifier clustering. Experimental results have demonstrated that the classifiers constructed by the proposed method outperforms not only the classifiers generated by the conventional machine learning but also the classifiers generated by two widely used conventional ensemble learning methods (bagging and boosting).     

Transductive reliability estimation for medical diagnosis

In the past decades, machine learning (ML) tools have been successfully used in several medical diagnostic problems. While they often significantly outperform expert physicians (in terms of diagnostic accuracy, sensitivity, and specificity), they are mostly not being used in practice. One reason for this is that it is difficult to obtain an unbiased estimation of diagnose's reliability. We discuss how reliability of diagnoses is assessed in medical decision-making and propose a general framework for reliability estimation in machine learning, based on transductive inference. We compare our approach with a usual (machine learning) probabilistic approach as well as with classical stepwise diagnostic process where reliability of diagnose is presented as its post-test probability. The proposed transductive approach is evaluated on several medical datasets from the University of California (UCI) repository as well as on a practical problem of clinical diagnosis of the coronary artery disease (CAD). In all cases, significant improvements over existing techniques are achieved.     

Comparative experiments on learning information extractors for proteins and their interactions

OBJECTIVE: Automatically extracting information from biomedical text holds the promise of easily consolidating large amounts of biological knowledge in computer-accessible form. This strategy is particularly attractive for extracting data relevant to genes of the human genome from the 11 million abstracts in Medline. However, extraction efforts have been frustrated by the lack of conventions for describing human genes and proteins. We have developed and evaluated a variety of learned information extraction systems for identifying human protein names in Medline abstracts and subsequently extracting information on interactions between the proteins. METHODS AND MATERIAL: We used a variety of machine learning methods to automatically develop information extraction systems for extracting information on gene/protein name, function and interactions from Medline abstracts. We present cross-validated results on identifying human proteins and their interactions by training and testing on a set of approximately 1000 manually-annotated Medline abstracts that discuss human genes/proteins. RESULTS: We demonstrate that machine learning approaches using support vector machines and maximum entropy are able to identify human proteins with higher accuracy than several previous approaches. We also demonstrate that various rule induction methods are able to identify protein interactions with higher precision than manually-developed rules. CONCLUSION: Our results show that it is promising to use machine learning to automatically build systems for extracting information from biomedical text. The results also give a broad picture of the relative strengths of a wide variety of methods when tested on a reasonably large human-annotated corpus.     

基于心电图的心肌梗死识别分类研究

心肌梗死（MI）是一种严重的心脏病,症状前的健康检查可以发现早期的MI。心电图（ECG）是一种常用的无创健康检查诊断工具。一些使用ECG预测MI的研究存在基于私人数据集、样本量小、分析方法简单等不足。为了解决这些问题,本研究提出在英国最大的开放采集生物信息资源平台UK Biobank上进行MI的首次基准预测实验,涵盖基于临床特征的机器学习方法和基于ECG信号的深度学习方法。结果显示,基于临床特征的AUC为0.690,深度学习使用原始ECG信号的AUC为0.728,提升近4%。证明深度学习基于原始ECG信号能学习到比临床特征更多的信息。另外,对XGBoost和ResNet方法的结果进行了初步的可解释性分析,发现ST波与MI的关联更密切。     

基于机器学习的神经精神疾病辅助诊断研究进展

【摘要】神经影像技术被广泛应用于研究大脑结构和功能异常与神经精神疾病之间的相关性。与传统的统计学分析方法不同,机器学习模型能对神经影像学数据进行个体化预测,发掘潜在的生物学标记物。神经精神疾病辅助诊断包含数据预处理和机器学习算法。数据预处理是一种人为的特征工程,为机器学习算法提供量化特征;机器学习算法包含特征降维、模型训练和模型评估。鲁棒的机器学习算法可以实现对不同数据集的准确预测,并提供对预测结果贡献大的特征,作为潜在的生物学标记物。本文综述了近年来基于机器学习的神经精神疾病辅助诊断研究进展,从数据预处理、机器学习算法和生物学标记物3个角度进行介绍,并展望未来的研究方向。 【关键词】神经精神疾病;神经影像;机器学习;辅助诊断     

Integrating Natural Language Processing and Machine Learning Algorithms to Categorize Oncologic Response in Radiology Reports

Journal of Digital Imaging - A significant volume of medical data remains unstructured. Natural language processing (NLP) and machine learning (ML) techniques have shown to successfully extract...