Predictability analysis for an automated seizure prediction algorithm.

Epileptic seizures of mesial temporal origin are preceded by changes in signal properties detectable in the intracranial EEG. A series of computer algorithms designed to detect the changes in spatiotemporal dynamics of the EEG signals and to warn of impending seizures have been developed. In this study, we evaluated the performance of a novel adaptive threshold seizure warning algorithm (ATSWA), which detects the convergence in Short-Term Maximum Lyapunov Exponent (STLmax) values among critical intracranial EEG electrode sites, as a function of different seizure warning horizons (SWHs). The ATSWA algorithm was compared to two statistical based na?ve prediction algorithms (periodic and random) that do not employ EEG information. For comparison purposes, three performance indices "area above ROC curve" (AAC), "predictability power" (PP) and "fraction of time under false warnings" (FTF) were defined and the effect of SWHs on these indices was evaluated. The results demonstrate that this EEG based seizure warning method performed significantly better (P < 0.05) than both na?ve prediction schemes. Our results also show that the performance indexes are dependent on the length of the SWH. These results suggest that the EEG based analysis has the potential to be a useful tool for seizure warning.     

Validation of an automated computational method for skeletal muscle fibre morphometry analysis

Accurate and fast measurement of muscle fibre size and evaluation of fibre type proportions in large cross-sectional areas remains challenging as existing methods require extensive manual measurements. In this study, we assessed the fibre morphometry of ∼1000 fibres in mouse and human control and diseased muscle cross-sections. We compared fibre size, percentage fibre proportion and percentage fibre surface area results obtained by an automated method using MetaMorph® with those obtained manually using Image Pro. Data collection using MetaMorph® software was faster and produced similar results to those obtained using Image Pro. The ability to quickly and accurately measure large numbers of fibres with MetaMorph® allows the researcher to make a more precise assessment of fibre type and fibre size changes in human muscle biopsies and animal models of muscle disease.     

Characterization of the acute pulse transit time response to obstructive apneas and hypopneas in preschool children with sleep-disordered breathing

Background

Surges in heart rate (HR) and blood pressure (BP) at apnea termination contribute to the hypertension seen in obstructive sleep apnea (OSA). Because childhood OSA prevalence peaks in the preschool years, we aimed to characterize the cardiovascular response to obstructive events in preschool-aged children.

Methods

Clinically referred children aged 3–5 years were grouped by obstructive apnea–hypopnea index (OAHI) into the following: primary snoring (PS) (OAHI ? 1 event/h [n = 21]), mild OSA (OAHI > 1– ? 5 [n = 32]), and moderate to severe (MS) OSA (OAHI > 5 [n = 28]). Beat-to-beat pulse transit time (PTT), an inverse continuous indicator of BP changes, and HR were averaged during the two halves (early and late) and during the peak after (post) each obstructive event and were expressed as percentage change from late- to post-event.

Results

We analyzed 422 events consisting of 55 apneas and 367 hypopneas. A significant post-event increase in HR and fall in PTT occurred in all severity groups (P < .05 for all). A greater response was associated with OSA, nonrapid eye movement sleep (NREM), cortical arousal, hypopneas, and oxygen desaturation (P < .05 for all).

Conclusions

Obstructive events elicit acute cardiovascular changes in preschool children. Such circulatory perturbations have been implicated in the development of hypertension, and our findings complement previous studies to suggest a cumulative impact of snoring on the cardiovascular system from childhood into adulthood.     

Validation of an automated punctate mechanical stimuli delivery system designed for fMRI studies in rodents

Functional magnetic resonance imaging (fMRI) is increasingly being used for animal studies studying the transmission of nociceptive information. Application of noxious mechanical stimuli is widely used for animal and human assessment of pain processing. Any accessory hardware used in animal imaging studies must, however, be sufficiently small to fit in the magnet bore diameter and be non-magnetic. We have developed a system that can apply mechanical stimuli simultaneously with fMRI. This system consists of a standardized instrument to deliver mechanical stimuli (VonFrey monofilament) and a gas-pressured mechanical transducer. These components were integrated with a computer console that controlled the period of stimuli to match acquisition scans. Preliminary experiments demonstrated that the force-stimulus transducer did not influence MRI signal to noise ratio. Mechanical stimulation of the hindpaw significantly increased blood oxygen level dependent (BOLD) signal intensity in several midbrain regions involved in the processing of nociceptive information in the rat (p<0.001, uncorrected for multiple comparisons). This system can be applied to both animal and human imaging studies and has a wide range of applications for studies of nociceptive processing.     

Validation of the NetSCID: an automated web-based adaptive version of the SCID

The present study developed and validated a configurable, adaptive, web-based version of the Structured Clinical Interview for DSM, the NetSCID. The validation included 24 clinicians who administered the SCID and 230 participants who completed the paper SCID and/or the NetSCID. Data-entry errors, branching errors, and clinician satisfaction were quantified. Relative to the paper SCID, the NetSCID resulted in far fewer data-entry and branching errors. Clinicians ‘preferred’ using the NetSCID and found that the NetSCID was easier to administer.     

High inter-reviewer variability of spike detection on intracranial EEG addressed by an automated multi-channel algorithm

ObjectiveThe goal of this study was to determine the consistency of human reviewer spike detection and then develop a computer algorithm to make the intracranial spike detection process more objective and reliable.MethodsThree human reviewers marked interictal spikes on samples of intracranial EEGs from 10 patients. The sensitivity, precision and agreement in channel ranking by activity were calculated between reviewers. A computer algorithm was developed to parallel the way human reviewers detect spikes by first identifying all potential spikes on each channel using frequency filtering and then block scaling all channels at the same time in order to exclude potential spikes that fall below an amplitude and slope threshold. Its performance was compared to the human reviewers on the same set of patients.ResultsHuman reviewers showed surprisingly poor inter-reviewer agreement, but did broadly agree on the ranking of channels for spike activity. The computer algorithm performed as well as the human reviewers and did especially well at ranking channels from highest to lowest spike frequency.ConclusionsOur algorithm showed good agreement with the different human reviewers, even though they demonstrated different criteria for what constitutes a ‘spike’ and performed especially well at the clinically important task of ranking channels by spike activity.SignificanceAn automated, objective method to detect interictal spikes on intracranial recordings will improve both research and the surgical management of epilepsy patients.     

Validation of the Spanish version of the EURO-D scale: an instrument for detecting depression in older people

OBJECTIVE: To standardize the EURO-D scale as a method for detecting depression in Spanish older people. METHODS: In a first phase, a sample of 1080 community older people was assessed with the GMS-AGECAT by lay interviewers. In a second phase, all the probable cases and a similar number of randomly selected probable non-cases were assessed using DSM-III-R diagnosis by psychiatrists. To test reliability, internal consistency with the Cronbach alpha coefficient and test-retest reliability (Kappa value) were obtained. Validity, sensitivity and specificity, predictive values, Receiver Operating Characteristic (ROC) curve, and Specific Likelihood Ratios (SSLR) were calculated. RESULTS: A high internal consistency was obtained in both phases of the study (alpha = 0.75; alpha = 0.79, respectively). Test-retest reliability was acceptable (weighted kappa = 0.60). For the prediction of DSM-III-R diagnosis validity coefficients (cut-off point 3/4) were: sensitivity 91.8%, specificity 76.6% and area under the ROC curve 0.92 [95% Confidence Intervals (0.89-0.95)]. Illiteracy and cognitive difficulties had a small negative effect on the performance of the scale. CONCLUSIONS: The EURO-D scale is a reliable and valid instrument for detecting probable cases of depression in older people in Spain. Illiteracy and cognitive difficulties should be considered when interpreting the results.     

Validation of the Retardation Rating Scale for detecting depression in geriatric inpatients

OBJECTIVES: Validation in the elderly of the Retardation Rating Scale (RRS), which includes items related to motor and mental retardation but not vegetative items, and may be particularly well-suited for the diagnosis of depression in the elderly. METHODS: One hundred and sixty-five geriatric inpatients (105 depressed), aged 65 and over, without dementia, neuroleptic medication and increased risk of slowed mobility, were assessed with the RRS and three validated 'gold-standard' scales for geriatric depression (Hamilton Depression Rating Scale, Montgomery and Asberg Depression Rating Scale, Geriatric Depression Scale). Factor analysis used varimax rotation, Cronbach's, Spearman's and Ferguson's coefficients and the Mann-Whitney U-test to evaluate construct and internal consistency. Convergent validity and Receiver Operating Characteristics curves were also analyzed. RESULTS: Factor analysis retained three interpretable domains: (1) motor items (45% of the variance); (2) mental items and (3) the cognitive items. Internal consistency was high (alpha = 0.91). Each item was strongly correlated with the total RRS score and associated with depression. The RRS showed good convergent validity and its total score increased with depression severity. A cut-off score of 10 yielded 79% sensitivity and 80% specificity, with 80% of the patients properly classified, that is 15% more than standard observer scales. CONCLUSION: RRS is a valid screening tool for depression and improves recognition of depression in geriatric inpatients.     

Validation of a semi-automated scintigraphic technique for detecting episodic, real-time colonic flow

The relationships between the movement of colonic content and regional pressures have only been partially defined. During the analysis of a combined colonic scintigraphic and manometric study, a quantitative technique for determining discrete, episodic, real-time colonic flow was developed. Our aim was to validate this technique through the construction of a computer-generated phantom model of known antegrade and retrograde motility. The anthropoid phantom was rasterized into a 6-mm voxel model to create a 3D voxel phantom of the colon with four distinct colonic segments. Associating a time/activity curve with each segment simulated dynamic behaviour. Activity in the model was based on data obtained from human colonic scintigraphic recordings using 30 MBq of (99m)Tc sulphur colloid. The flow was simulated by modifying the input time/activity functions to represent episodes of net flow of 2%, 5% or 10% of segmental content. Our quantitative technique was applied to the phantom model to measure the accuracy with which simulated flows were detected. Our quantitative technique proved to be a sensitive and specific means of detecting the presence and the magnitude of discrete episodes of colonic flow and therefore, should improve our ability to correlate colonic flow and motor patterns.     

Validation of Addenbrooke's cognitive examination for detecting early dementia in a Japanese population

There is a clear need for brief, but sensitive and specific, cognitive screening instruments for dementia. We assessed the diagnostic accuracy of the Japanese version of Addenbrooke's Cognitive Examination (ACE) in identifying early dementia in comparison with the conventional Mini-Mental State Examination (MMSE). Standard tests for evaluating dementia screening tests were applied. A total of 201 subjects (Alzheimer's disease (AD) = 65, frontotemporal dementia (FTD) = 24, vascular dementia = 26, dementia with Lewy bodies = 11, mild cognitive impairment (MCI) = 13, and controls = 62) participated in this study. The reliability of the ACE was very good (alpha coefficient = 0.82). In our patient series, the sensitivity for diagnosing dementia with an ACE score of ≤ 74 was 0.889 with a specificity of 0.987, and the sensitivity of an ACE score of ≤ 80 was 0.984 with a specificity of 0.867. The Japanese version of the ACE is a very accurate instrument for the detection of early dementia, and should be widely used in clinical practice.     

An automated system for the mapping and quantitative analysis of immunocytochemistry of an inducible nuclear protein

We describe here an automated system that accurately maps tissue sections stained by immunocytochemistry for an inducible nuclear protein. The sections are scanned with a computer-controlled microscope setup hooked to a CCD camera. Raw images captured at high resolution are filtered using highly selective criteria for the recognition of labeled cell nuclei. The total population of recognized labeled nuclei is then divided into separate bins, according to their labeling intensities. Finally, information about both the position and labeling intensity of labeled nuclei is represented in average density maps. The system was optimized for the quantitative mapping of neuronal cells expressing the inducible gene ZENK in the brain of songbirds, in response to stimulation with song, but should be of general applicability for the mapping of inducible nuclear proteins.     

SPiQE: An automated analytical tool for detecting and characterising fasciculations in amyotrophic lateral sclerosis

ObjectivesFasciculations are a clinical hallmark of amyotrophic lateral sclerosis (ALS). Compared to concentric needle EMG, high-density surface EMG (HDSEMG) is non-invasive and records fasciculation potentials (FPs) from greater muscle volumes over longer durations. To detect and characterise FPs from vast data sets generated by serial HDSEMG, we developed an automated analytical tool.MethodsSix ALS patients and two control patients (one with benign fasciculation syndrome and one with multifocal motor neuropathy) underwent 30-minute HDSEMG from biceps and gastrocnemius monthly. In MATLAB we developed a novel, innovative method to identify FPs amidst fluctuating noise levels. One hundred repeats of 5-fold cross validation estimated the model’s predictive ability.ResultsBy applying this method, we identified 5,318 FPs from 80 minutes of recordings with a sensitivity of 83.6% (+/? 0.2 SEM), specificity of 91.6% (+/? 0.1 SEM) and classification accuracy of 87.9% (+/? 0.1 SEM). An amplitude exclusion threshold (100 μV) removed excessively noisy data without compromising sensitivity. The resulting automated FP counts were not significantly different to the manual counts (p = 0.394).ConclusionWe have devised and internally validated an automated method to accurately identify FPs from HDSEMG, a technique we have named Surface Potential Quantification Engine (SPiQE).SignificanceLongitudinal quantification of fasciculations in ALS could provide unique insight into motor neuron health.     

Convergence analysis of an augmented algorithm for fully complex-valued neural networks

This paper presents an augmented algorithm for fully complex-valued neural network based on Wirtinger calculus, which simplifies the derivation of the algorithm and eliminates the Schwarz symmetry restriction on the activation functions. A unified mean value theorem is first established for general functions of complex variables, covering the analytic functions, non-analytic functions and real-valued functions. Based on so introduced theorem, convergence results of the augmented algorithm are obtained under mild conditions. Simulations are provided to support the analysis.     

Validation of a swallowing disturbance questionnaire for detecting dysphagia in patients with Parkinson's disease.

Underreporting of swallowing disturbances by Parkinson's disease (PD) patients may lead to delay in diagnosis and treatment, alerting the physician to an existing dysphagia only after the first episode of aspiration pneumonia. We developed and validated a swallowing disturbance questionnaire (SDQ) for PD patients and compared its findings to an objective assessment. Fifty-seven PD patients (mean age 69 +/- 10 years) participated in this study. Each patient was queried about experiencing swallowing disturbances and asked to complete a self-reported 15-item "yes/no" questionnaire on swallowing disturbances (24 replied "no"). All study patients underwent a physical/clinical swallowing evaluation by a speech pathologist and an otolaryngologist. The 33 patients who complained of swallowing disturbances also underwent fiberoptic endoscopyic evaluation of swallowing (FEES). According to the ROC test, the "optimal" score (where the sensitivity and specificity curves cross) is 11 (sensitivity 80.5%, specificity 81.3%). Using the SDQ questionnaire substantially reduced Type I errors (specifically, an existing swallowing problem missed by the selected cutoff point). On the basis of the SDQ assessment alone, 12 of the 24 (50%) noncomplaining patients would have been referred to further evaluation that they otherwise would not have undergone. The SDQ emerged as a validated tool to detect early dysphagia in PD patients.     

Detection of REM sleep behaviour disorder by automated polysomnography analysis

ObjectiveEvidence suggests Rapid-Eye-Movement (REM) Sleep Behaviour Disorder (RBD) is an early predictor of Parkinson’s disease. This study proposes a fully-automated framework for RBD detection consisting of automated sleep staging followed by RBD identification.MethodsAnalysis was assessed using a limited polysomnography montage from 53 participants with RBD and 53 age-matched healthy controls. Sleep stage classification was achieved using a Random Forest (RF) classifier and 156 features extracted from electroencephalogram (EEG), electrooculogram (EOG) and electromyogram (EMG) channels. For RBD detection, a RF classifier was trained combining established techniques to quantify muscle atonia with additional features that incorporate sleep architecture and the EMG fractal exponent.ResultsAutomated multi-state sleep staging achieved a 0.62 Cohen’s Kappa score. RBD detection accuracy improved from 86% to 96% (compared to individual established metrics) when using manually annotated sleep staging. Accuracy remained high (92%) when using automated sleep staging.ConclusionsThis study outperforms established metrics and demonstrates that incorporating sleep architecture and sleep stage transitions can benefit RBD detection. This study also achieved automated sleep staging with a level of accuracy comparable to manual annotation.SignificanceThis study validates a tractable, fully-automated, and sensitive pipeline for RBD identification that could be translated to wearable take-home technology.