Automatic Early Detection Research Articles

Early automatic detection of Parkinson's disease based on sleep recordings.

Idiopathic rapid-eye-movement (REM) sleep behavior disorder (iRBD) is most likely the earliest sign of Parkinson's Disease (PD) and is characterized by REM sleep without atonia (RSWA) and consequently increased muscle activity. However, some muscle twitching in normal subjects occurs during REM sleep. There are no generally accepted methods for evaluation of this activity and a normal range has not been established. Consequently, there is a need for objective criteria. In this study we propose a full-automatic method for detection of RSWA. REM sleep identification was based on the electroencephalography and electrooculography channels, while the abnormal high muscle activity was detected from the electromyography channels, in this case the submentalis combined with left and right anterior tibialis. RSWA was identified by considering it an outlier problem, in which the number of outliers during REM sleep was used as a quantitative measure of muscle activity. The proposed method was able to automatically separate all iRBD test subjects from healthy elderly controls and subjects with periodic limb movement disorder. The proposed work is considered a potential automatic method for early detection of PD.

Multisensor Network System for Wildfire Detection Using Infrared Image Processing

This paper presents the next step in the evolution of multi-sensor wireless network systems in the early automatic detection of forest fires. This network allows remote monitoring of each of the locations as well as communication between each of the sensors and with the control stations. The result is an increased coverage area, with quicker and safer responses. To determine the presence of a forest wildfire, the system employs decision fusion in thermal imaging, which can exploit various expected characteristics of a real fire, including short-term persistence and long-term increases over time. Results from testing in the laboratory and in a real environment are presented to authenticate and verify the accuracy of the operation of the proposed system. The system performance is gauged by the number of alarms and the time to the first alarm (corresponding to a real fire), for different probability of false alarm (PFA). The necessity of including decision fusion is thereby demonstrated.

Monitoring the Health of Large Interconnected Power Systems: A Near Real-Time Perspective

Large interconnected power systems are exposed to many system disturbances that may impair system security and result in wide-spread damage. This paper discusses the development of advanced health monitoring systems for large interconnected power systems. The use of adaptive, data processing techniques to provide automatic early detection of system deterioration in large power systems is investigated. Areas of improvement in real-time system monitoring are identified and algorithms to detect both, oscillatory and abrupt changes in system behavior are proposed and tested. The applicability of the proposed monitoring system is evaluated using actual measurements from significant events.

Automatic detection of AutoPEEP during controlled mechanical ventilation

BackgroundDynamic hyperinflation, hereafter called AutoPEEP (auto-positive end expiratory pressure) with some slight language abuse, is a frequent deleterious phenomenon in patients undergoing mechanical ventilation. Although not readily quantifiable, AutoPEEP can be recognized on the expiratory portion of the flow waveform. If expiratory flow does not return to zero before the next inspiration, AutoPEEP is present. This simple detection however requires the eye of an expert clinician at the patient’s bedside. An automatic detection of AutoPEEP should be helpful to optimize care.MethodsIn this paper, a platform for automatic detection of AutoPEEP based on the flow signal available on most of recent mechanical ventilators is introduced. The detection algorithms are developed on the basis of robust non-parametric hypothesis testings that require no prior information on the signal distribution. In particular, two detectors are proposed: one is based on SNT (Signal Norm Testing) and the other is an extension of SNT in the sequential framework. The performance assessment was carried out on a respiratory system analog and ex-vivo on various retrospectively acquired patient curves.ResultsThe experiment results have shown that the proposed algorithm provides relevant AutoPEEP detection on both simulated and real data. The analysis of clinical data has shown that the proposed detectors can be used to automatically detect AutoPEEP with an accuracy of 93% and a recall (sensitivity) of 90%.ConclusionsThe proposed platform provides an automatic early detection of AutoPEEP. Such functionality can be integrated in the currently used mechanical ventilator for continuous monitoring of the patient-ventilator interface and, therefore, alleviate the clinician task.

Fast Detection of Unexpected Sound Intensity Decrements as Revealed by Human Evoked Potentials

The detection of deviant sounds is a crucial function of the auditory system and is reflected by the automatically elicited mismatch negativity (MMN), an auditory evoked potential at 100 to 250 ms from stimulus onset. It has recently been shown that rarely occurring frequency and location deviants in an oddball paradigm trigger a more negative response than standard sounds at very early latencies in the middle latency response of the human auditory evoked potential. This fast and early ability of the auditory system is corroborated by the finding of neurons in the animal auditory cortex and subcortical structures, which restore their adapted responsiveness to standard sounds, when a rare change in a sound feature occurs. In this study, we investigated whether the detection of intensity deviants is also reflected at shorter latencies than those of the MMN. Auditory evoked potentials in response to click sounds were analyzed regarding the auditory brain stem response, the middle latency response (MLR) and the MMN. Rare stimuli with a lower intensity level than standard stimuli elicited (in addition to an MMN) a more negative potential in the MLR at the transition from the Na to the Pa component at circa 24 ms from stimulus onset. This finding, together with the studies about frequency and location changes, suggests that the early automatic detection of deviant sounds in an oddball paradigm is a general property of the auditory system.

Selection of Optimal Wavelength Features for Decay Detection in Citrus Fruit Using the ROC Curve and Neural Networks

Early automatic detection of fungal infections in post-harvest citrus fruits is especially important for the citrus industry because only a few infected fruits can spread the infection to a whole batch during operations such as storage or exportation, thus causing great economic losses. Nowadays, this detection is carried out manually by trained workers illuminating the fruit with dangerous ultraviolet lighting. The use of hyperspectral imaging systems makes it possible to advance in the development of systems capable of carrying out this detection process automatically. However, these systems present the disadvantage of generating a huge amount of data, which must be selected in order to achieve a result that is useful to the sector. This work proposes a methodology to select features in multi-class classification problems using the receiver operating characteristic curve, in order to detect rottenness in citrus fruits by means of hyperspectral images. The classifier used is a multilayer perceptron, trained with a new learning algorithm called extreme learning machine. The results are obtained using images of mandarins with the pixels labelled in five different classes: two kinds of sound skin, two kinds of decay and scars. This method yields a reduced set of features and a classification success rate of around 89%.

Structural health monitoring of fixed offshore structures using the bicoherence function of ambient vibration measurements

This paper investigates the use of the bicoherence function of measured structural acceleration to provide automatic early detection of damage in an offshore structure. A detailed simulation model is developed, including realistic wave loading, and is used to assess the performance of the method. The method is shown through analytical and numerical analysis to be insensitive to typical operating parameter variations and to variations in wave excitation force. It is demonstrated that very small changes in stiffness of individual structural members are detectable from measurements of global structural motion.

INTELLIGENT DIAGNOSIS OF CARDIOVASCULAR DISEASES UTILIZING ECG SIGNALS

Early automatic detection of cardiovascular diseases is of great importance to provide timely treatment and reduce fatality rate. Although many efforts have been devoted to detecting various arrhythmias, classification of other common cardiovascular diseases still lacks comprehensive and intensive studies. This work aims at developing an automatic diagnosis system for myocardial infarction, valvular heart disease, cardiomyopathy, hypertrophy, and bundle branch block, based on the clinic recordings provided by PTB Database. The proposed diagnosis system consists of the components as baseline wander reduction, beat segmentation, feature extraction, feature reduction and classification. The selected features are the location, amplitude and width of each wave, exactly the parameters of ECG dynamical model. We also propose a mean shift algorithm based method to extract these features. To demonstrate the availability and efficacy of the proposed system, we use a total of 13,564 beats to conduct a large scale experiment, where only 25% beats are utilized to train the eigenvectors of generalized discriminant analysis in the feature reduction phase and 25% beats are applied to train the support vector machine in the classification phase. The average sensitivity, specificity and positive predicitivity for the test set, containing 75% beats, are respectively 96.06%, 99.32% and 97.29%.

Automatic identification of epileptic electroencephalography signals using higher-order spectra

Epilepsy is a pathological condition characterized by the spontaneous and unforeseeable occurrence of seizures, during which the perception or behaviour of patients is disturbed. An automatic early detection of the seizure onsets would help the patients and observers to take appropriate precautions. Various methods have been proposed to predict the onset of seizures based on electroencephalography (EEG) recordings. The use of non-linear features motivated by the higher-order spectra (HOS) has been reported to be a promising approach to differentiate between normal, background (pre-ictal), and epileptic EEG signals. In this work, these features are used to train both a Gaussian mixture model classifier and a support vector machine classifier. Results show that the classifiers were able to achieve 93.11 per cent and 92.67 per cent classification accuracy respectively, with selected HOS-based features. About 2 h of EEG recordings from ten patients were used in this study.

Remote Monitoring for Advisories: Automatic Early Detection of Silent Lead Failure

Pacing and Clinical ElectrophysiologyVolume 32, Issue 4 p. 525-527 Remote Monitoring for Advisories: Automatic Early Detection of Silent Lead Failure NIRAJ VARMA M.D., NIRAJ VARMA M.D. From Cardiac Electrophysiology and Pacing, Cleveland Clinic, Cleveland, OhioSearch for more papers by this author NIRAJ VARMA M.D., NIRAJ VARMA M.D. From Cardiac Electrophysiology and Pacing, Cleveland Clinic, Cleveland, OhioSearch for more papers by this author First published: 23 March 2009 https://doi.org/10.1111/j.1540-8159.2009.02314.xCitations: 37 Address for reprints: Niraj Varma, M.D., F.R.C.P., Cardiac Electrophysiology, Cardiovascular Medicine, Desk J2-2, Cleveland Clinic, 9500 Euclid Avenue, Cleveland, OH 44195. Fax: 216-445-6161; e-mail: varman@ccf.org Read the full textAboutPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Share a linkShare onFacebookTwitterLinked InRedditWechat Citing Literature Volume32, Issue4April 2009Pages 525-527 RelatedInformation

Fault diagnosis of machines via parameter estimation and knowledge processing—Tutorial paper

For further improvement of the reliability and safety of machines the automatic early detection and localization of faults is of high interest. The conventional approach is to monitor some important variables like temperature, pressure, vibration and to generate alarms if certain limits are exceeded. However, developing internal faults are in this way only detected at a rather late stage. By applying static and dynamic process models and common process input and output measurements the inherent relationships and redundancies can be used to detect faults earlier and to localize them better. Changes in process and signal parameters are very well suited for fault detection. The paper describes a general methodology for machines and other processes by using few measurements, dynamic process and signal models and parameter estimation to generate analytical symptoms. Heuristic symptoms observed by the operator or based on, e.g. the process history are another source for fault diagnosis. Both kinds of symptoms are weighted by confidence measures and are fed into a knowledge based fault diagnosis procedure. The diagnosis is performed by forward and backward chaining according to causal fault-symptom trees. The considered machines consist of a motor, a drive chain and a working process or load. They may be electrical motor or combustion engine driven pumps, fans or machine tools with gear or belt drive chains. As one example, experimental results are shown for a DC motor powered feed drive of a machine tool. A summary of practical results with other machines is also given.

Fault Diagnosis of Machines via Parameter Estimation and Knowledge Processing

Abstract For further improvement of the reliability and safety of machines the automatic early detection and localization of faults is of high interest. The conventional approach is to monitor some important variables like temperatures, pressures, vibrations and to generate alarms if certain limits are exeeded. However, developing internal faults are in this way only detected at a rather late stage. By applying static and dynamic process models and common process input and output measurements the inherent relationships and redundancies can be used to detect faults earlier and to localize them better. Changes in process and signal parameters are very well suited for fault detection. The paper describes a general methodology for machines and other processes by using few measurements, dynamic process and signal models and parameter estimation to generate symptoms. The detected symptoms are fed into a knowledge based fault diagnosis procedure. Analytical and heuristic knowledge is treated via fault-symptom trees, process history and plausibility measures. The considered machines consist of a motor, a drive chain and a working process or load. They may be electrical motor or combustion engine driven pumps, fans or machine tools with gear or belt drive chains. The described methodology was developed and tested experimentally for several machines. As one example, experimental results are shown for a d.c. motor powered feed drive of a machine tool.

The orienting response and stimulus significance: Further comments

Siddle's (1979) assertion that the significance hypothesis is based on tautological argument regarding the effects of intensity decrease refer only to two papers which first suggested the need for research into determinants other than mismatch. Studies testing specific predictions followed in various laboratories, and it is their results that support the hypothesis. The hypothesis never depended on intensity decrease, and the charge of nonreplicability is incorrect. Siddle's argument concerning response stereotypy cannot account for the data showing response to significant ‘novelty’ coupled with nonresponse to nonsignificant ‘novelty’ in the same OR component, or for that showing consistent differences in various components between randomly selected groups. Arguments based on trait-lability are inadequate since tests of the lability hypothesis have confounded trait factors with the differences in stimulus assessment they hope to test. The significance hypothesis emphasizes stimulus information in the OR, significance serving to limit access to central processing channels to those signals whose information is relevant for the organism. OR components are seen as long-latency responses elicited after automatic early detection of uncertainty and significance, and reflect higher processing of significant information. A distinction between involuntary and voluntary phases of attention is acknowledged, but it is questioned whether OR components can be assigned to the former because of the frequent failure of ‘automatic’ ORs to innocuous stimuli, the long latency of most components, and failure to demonstrate structural differences between the classes of response.