A digital assistant for shading paper sketches.

We present a mixed reality-based assistive system for shading paper sketches. Given a paper sketch made by an artist, our interface helps inexperienced users to shade it appropriately. Initially, using a simple Delaunay-triangulation based inflation algorithm, an approximate depth map is computed. The system then highlights areas (to assist shading) based on a rendering of the 2.5-dimensional inflated model of the input contour. With the help of a mixed reality system, we project the highlighted areas back to aid users. The hints given by the system are used for shading and are smudged appropriately to apply an artistic shading to the sketch. The user is given flexibility at various levels to simulate conditions such as height and light position. Experiments show that the proposed system aids novice users in creating sketches with impressive shading.

Automated Detection of Convulsive Seizures Using a Wearable Accelerometer Device.

Epileptic seizure detection requires specialized approaches such as video/electroencephalography monitoring. However, these approaches are restricted mainly to hospital setting and requires video/EEG analysis by experts, which makes these approaches resource- and labor-intensive. In contrast, we aim to develop a wireless remote monitoring system based on a single wrist-worn accelerometer device, which is sensitive to multiple types of convulsive seizures and is capable of detecting seizures with short duration. Simple time domain features including a new set of Poincar´e plot based features were extracted from the active movement events recorded using a wrist-worn accelerometer device. The best features were then selected using the area under the ROC curve analysis. Kernelized support vector data description (SVDD) was then used to classify non-seizure and seizure events. The proposed algorithm was evaluated on 5;576h of recordings from 79 patients and detected 40 (86:95%) of 46 convulsive seizures (generalized tonic-clonic (GTCS), psychogenic non-epileptic (PNES), and complex partial seizures (CPS)) from twenty patients with a total of 270 false alarms (1:16=24h). Furthermore, the algorithm showed a comparable performance (sensitivity 95:23% and false alarm rate 0:64=24h) with respect to existing unimodal and multi-modal methods for GTCS detection. The promising results shows the potential to build an ambulatory monitoring convulsive seizure detection system. A wearable accelerometer based seizure detection system would aid in continuous assessment of convulsive seizures in a timely and non-invasive manner.

Novel features for capturing temporal variations of rhythmic limb movement to distinguish convulsive epileptic and psychogenic nonepileptic seizures.

OBJECTIVE: To investigate the characteristics of motor manifestation during convulsive epileptic and psychogenic nonepileptic seizures (PNES), captured using a wrist-worn accelerometer (ACM) device. The main goal was to find quantitative ACM features that can differentiate between convulsive epileptic and convulsive PNES. METHODS: In this study, motor data were recorded using wrist-worn ACM-based devices. A total of 83 clinical events were recorded: 39 generalized tonic-clonic seizures (GTCS) from 12 patients with epilepsy, and 44 convulsive PNES from 7 patients (one patient had both GTCS and PNES). The temporal variations in the ACM traces corresponding to 39 GTCS and 44 convulsive PNES events were extracted using Poincaré maps. Two new indices-tonic index (TI) and dispersion decay index (DDI)-were used to quantify the Poincaré-derived temporal variations for every GTCS and convulsive PNES event. RESULTS: The TI and DDI of Poincaré-derived temporal variations for GTCS events were higher in comparison to convulsive PNES events (P < 0.001). The onset and the subsiding patterns captured by TI and DDI differentiated between epileptic and convulsive nonepileptic seizures. An automated classifier built using TI and DDI of Poincaré-derived temporal variations could correctly differentiate 42 (sensitivity: 95.45%) of 44 convulsive PNES events and 37 (specificity: 94.87%) of 39 GTCS events. A blinded review of the Poincaré-derived temporal variations in GTCS and convulsive PNES by epileptologists differentiated 26 (sensitivity: 70.27%) of 44 PNES events and 33 (specificity: 86.84%) of 39 GTCS events correctly. SIGNIFICANCE: In addition to quantifying the motor manifestation mechanism of GTCS and convulsive PNES, the proposed approach also has diagnostic significance. The new ACM features incorporate clinical characteristics of GTCS and PNES, thus providing an accurate, low-cost, and practical alternative to differential diagnosis of PNES.

Improved Detection and Classification of Convulsive Epileptic and Psychogenic Non-epileptic Seizures Using FLDA and Bayesian Inference.

A high number of patients with epileptic seizures (ES) are misdiagnosed due to prevalence of mimic conditions. The clinical characteristics of mimics are often similar to ES. The events mostly misdiagnosed are of psychogenic origin and are termed as psychogenic non-epileptic seizures (PNES). The gold standard for diagnosis of PNES is video-electroencephalography monitoring (VEM), which is a resource demanding process. Hence, need for a more object method of PNES diagnosis is created. Accelerometer sensors have been used previously for the diagnosis of ES. In this work, we present a new approach for detection and classification of PNES using wrist-worn accelerometer device. Various time, frequency and wavelet space features are extracted from the accelerometry signal. Feature compression is then performed using Fisher linear discriminant analysis (FLDA). A Bayesian classifier is then trained using kernel estimator method. The algorithm was trained and tested on data collected from 16 patients undergoing VEM. When tested, the algorithm detected all seizures with 20 false alarms and correctly classified 100% PNES and 75% ES, respectively of the detected seizures.

Onset Detection of Epileptic Seizures From Accelerometry Signal.

Epileptic seizures are the result of any abnormal asynchronous firing of cortical neurons. Seizures are abrupt and pose a risk of injury and fatal harm to the patient. Epilepsy affects patients quality of life (QOL) and imposes financial, social, and physical burden on the patient. The unpredictability associated with seizures further adds to the reduced QOL and increases dependence on caregivers and family members. A seizure triggered alarm system can reduce the risk of seizure-related injuries and aid in improving patient's QOL. This study presents real-time onset detection of seizures from accelerometry signal. An automated approach based on statistical machine learning is employed to learn the onset of seizures. To search for the optimal parameter that simultaneously maximizes detection sensitivity (sens) while minimizing false alarm rate (FAR) and latency, the epoch length is varied from $t=\{1,~10s\}$. Linear and non-linear time-varying dynamical patterns were extracted from every epoch using Poincaré plot analysis. The correlation patterns were learned using a kernalized support vector data descriptor. The preliminary analysis on accelerometry data collected from 8 epileptic patients with 9 generalized tonicclonic seizures (GTCS) shows promising results. The proposed algorithm detected all GTCS events (sens: 100%, FAR: 1. 09/24h) at 8s from onset. The proposed algorithm can lead to a sensitive, specific, and a relatively short-latency detection system for real-time remote monitoring of epileptic patients.

Detection of generalized tonic-clonic seizures using short length accelerometry signal.

Epileptic seizures are characterized by the excessive and abrupt electrical discharge in the brain. This asynchronous firing of neurons causes unprovoked convulsions which can be a cause of sudden unexpected death in epilepsy (SUDEP). Remote monitoring of epileptic patients can help prevent SUDEP. Systems based on wearable accelerometer sensors have shown to be effective in ambulatory monitoring of epileptic patients. However, these systems have a trade-off between seizure duration and the false alarm rate (FAR). The FAR of the system decreases as we increase the seizure duration. Further, multiple sensors are used in conjugation to improve the overall performance of the detection system. In this study, we propose a system based on single wrist-worn accelerometer sensor capable of detecting seizures with short duration (≥ 10s). Seizure detection was performed by employing machine learning approach such as kernelized support vector data description (SVDD). The proposed approach is validated on data collected from 12 patients, corresponding to approximately 966h of recording under video-telemetry unit. The algorithm resulted in a seizure detection sensitivity of 95.23% with a mean FAR of 0.72=24h.

Longitudinal multiple sclerosis lesion segmentation: Resource and challenge.

In conjunction with the ISBI 2015 conference, we organized a longitudinal lesion segmentation challenge providing training and test data to registered participants. The training data consisted of five subjects with a mean of 4.4 time-points, and test data of fourteen subjects with a mean of 4.4 time-points. All 82 data sets had the white matter lesions associated with multiple sclerosis delineated by two human expert raters. Eleven teams submitted results using state-of-the-art lesion segmentation algorithms to the challenge, with ten teams presenting their results at the conference. We present a quantitative evaluation comparing the consistency of the two raters as well as exploring the performance of the eleven submitted results in addition to three other lesion segmentation algorithms. The challenge presented three unique opportunities: (1) the sharing of a rich data set; (2) collaboration and comparison of the various avenues of research being pursued in the community; and (3) a review and refinement of the evaluation metrics currently in use. We report on the performance of the challenge participants, as well as the construction and evaluation of a consensus delineation. The image data and manual delineations will continue to be available for download, through an evaluation website2 as a resource for future researchers in the area. This data resource provides a platform to compare existing methods in a fair and consistent manner to each other and multiple manual raters.

Gaussian mixture model for the identification of psychogenic non-epileptic seizures using a wearable accelerometer sensor.

Any abnormal hypersynchronus activity of neurons can be characterized as an epileptic seizure (ES). A broad class of non-epileptic seizures is comprised of Psychogenic non-epileptic seizures (PNES). PNES are paroxysmal events, which mimics epileptic seizures and pose a diagnostic challenge with epileptic seizures due to their clinical similarities. The diagnosis of PNES is done using video-electroencephalography (VEM) monitoring. VEM being a resource intensive process calls for alternative methods for detection of PNES. There is now an emerging interest in the use of accelerometer based devices for the detection of seizures. In this work, we present an algorithm based on Gaussian mixture model (GMM's) for the identification of PNES, ES and normal movements using a wrist-worn accelerometer device. Features in time, frequency and wavelet domain are extracted from the norm of accelerometry signal. All events are then classified into three classes i.e normal, PNES and ES using a parametric estimate of the multivariate normal probability density function. An algorithm based on GMM's allows us to accurately model the non-epileptic and epileptic movements, thus enhancing the overall predictive accuracy of the system. The new algorithm was tested on data collected from 16 patients and showed an overall detection accuracy of 91% with 25 false alarms.