Veni, Vidi, Vici-When Is Home Video Seizure Monitoring Helpful?

Seizure detection is vital for managing epilepsy as seizures can lead to injury and even death, in addition to impacting quality of life. Prompt detection of seizures and intervention can help prevent injury and improve outcomes for individuals with epilepsy. Wearable sensors show promising results for automated detection of certain seizures, but they have limitations such as patient tolerance, impracticality for newborns, and the need for recharging. Non-contact video and audio-based technologies have become available, but a comprehensive literature review on these methods is lacking. This scoping literature review provides an overview of video and audio-based seizure detection, highlighting their potential benefits and challenges. It encompasses a thorough search and evaluation of relevant articles, summarizing methods and performances of these systems. The primary aim of this review is to examine and analyze existing research to identify patterns and gaps and establish a foundation for future advancements. We screened 7 databases using a set of standardized search criteria to minimize any potential missed articles. Four thousand four hundred eighty-seven deduplicated abstracts were screened and narrowed down to 34 studies that varied in design, algorithm methods, types of seizures detected, and performance metrics. Seizure detection sensitivity ranged from 100% to 0%, with optical flow analysis showing the highest sensitivity. The specificity of all included articles ranged from 97.7% to 60%. While limited studies reported accuracy, the highest reported was 100% using Radon Transform based technique on Dual Tree Complex Wavelet coefficients. Video and audio-based tools offer novel, noncontact approaches for detecting and monitoring seizures. Available studies are limited in sample sizes, dataset diversity, and standardized evaluation protocols, impacting the generalizability of results. Future research focusing on larger-scale investigations with diverse datasets, standardized evaluation protocols, and consistent reporting metrics is needed.

Computer vision for automated seizure detection and classification: A systematic review

Computer vision (CV) shows increasing promise as an efficient, low-cost tool for video seizure detection and classification. Here, we provide an overview of the fundamental concepts needed to understand CV and summarize the structure and performance of various model architectures used in video seizure analysis. We conduct a systematic literature review of the PubMed, Embase, and Web of Science databases from January 1, 2000 to September 15, 2023, to identify the strengths and limitations of CV seizure analysis methods and discuss the utility of these models when applied to different clinical seizure phenotypes. Reviews, nonhuman studies, and those with insufficient or poor quality data are excluded from the review. Of the 1942 records identified, 45 meet inclusion criteria and are analyzed. We conclude that the field has shown tremendous growth over the past 2 decades, leading to several model architectures with impressive accuracy and efficiency. The rapid and scalable detection offered by CV models holds the potential to reduce sudden unexpected death in epilepsy and help alleviate resource limitations in epilepsy monitoring units. However, a lack of standardized, thorough validation measures and concerns about patient privacy remain important obstacles for widespread acceptance and adoption. Investigation into the performance of models across varied datasets from clinical and nonclinical environments is an essential area for further research.

Functional Age Estimation Through Neonatal Motion Characterization Using Continuous Video Recordings

The follow-up of the development of the premature baby is a major component of its clinical care since it has been shown that it can reveal a pathology. However, no method allowing an automated and continuous monitoring of this development has been proposed. Within the framework of the Digi-NewB European project, our team wishes to offer new clinical indices qualifying the maturation of newborns. In this study, we propose a new method to characterize motor activity from video recordings. For this purpose, we have chosen to characterize the motion temporal organization by drawing inspiration from sleep organization. Thus, we propose a fully automatic process allowing to extract motion features and to combine them to estimate a functional age. By investigating two datasets, one of 28.5 hours (manually annotated) from 33 newborns and one of 4,920 hours from 46 newborns, we show that the proposed approach is relevant for monitoring in clinical routine and that the extracted features reflect the maturation of preterm newborns. Indeed, a compact and interpretable model using gestational age and three motion features (mean duration of intervals with motion, total percentage of time spent in motion and number of intervals without motion) was designed to predict post-menstrual age of newborns and showed an admissible mean absolute error of 1.3 weeks. While the temporal organization of motion was not studied clinically due to a lack of technological means, these results open the door to new developments, new investigations and new knowledge on the evolution of motion in newborns.

Development and validation of Auto-Neo-electroencephalography (EEG) to estimate brain age and predict report conclusion for electroencephalography monitoring data in neonatal intensive care units

Background

Electroencephalography (EEG) monitoring is widely used in neonatal intensive care units (NICUs). However, conventional EEG report generation processes are time-consuming and labor-intensive. Therefore, an automatic, objective, and comprehensive pipeline for brain age estimation and EEG report conclusion prediction is urgently needed to assist clinician’s decision-making.

Methods

We recruited patients who underwent EEG monitoring from the NICU at Children’s Hospital of Fudan University from Jan. 2016 to Mar. 2018. A total of 1,851 subjects were enrolled, including the patient’s conceptional age (CA) and the clinical EEG report conclusion (normal, slightly abnormal, moderately abnormal, or severely abnormal). A total of 1,591 subjects were used to generate predictive models and 260 were used as the validation dataset. We developed Auto-Neo-EEG (an automatic prediction system to assist clinical neonatal EEG report generation), including signal feature extraction, supervised machine learning realized by gradient boosted models, to estimate brain age and predict EEG report conclusion.

Results

The predicted results from the validation dataset were compared with the clinical observations to assess the performance. In the independent validation dataset, the model could achieve accordance 0.904 on estimating brain age for neonates with normal clinical EEG report conclusion, and differences between the predicted and observed brain age were strongly related with EEG report conclusion abnormality. Further, as for the EEG report conclusion prediction, the model could achieve area under the curve (AUC) of 0.984 for severely abnormal situations, and 0.857 for moderately abnormal ones.

Conclusions

The Auto-Neo-EEG has the high accuracy of estimating brain age and EEG report conclusion, which can potentially greatly accelerate the EEG report generation processes assist in clinical decision making.

Video-Based Detection of Generalized Tonic-Clonic Seizures Using Deep Learning

Timely detection of seizures is crucial to implement optimal interventions, and may help reduce the risk of sudden unexpected death in epilepsy (SUDEP) in patients with generalized tonic-clonic seizures (GTCSs). While video-based automated seizure detection systems may be able to provide seizure alarms in both in-hospital and at-home settings, earlier studies have primarily employed hand-designed features for such a task. In contrast, deep learning-based approaches do not rely on prior feature selection and have demonstrated outstanding performance in many data classification tasks. Despite these advantages, neural network-based video classification has rarely been attempted for seizure detection. We here assessed the feasibility and efficacy of automated GTCSs detection from videos using deep learning. We retrospectively identified 76 GTCS videos from 37 participants who underwent long-term video-EEG monitoring (LTM) along with interictal video data from the same patients, and 10 full-night seizure-free recordings from additional patients. Using a leave-one-subject-out cross-validation approach (LOSO-CV), we evaluated the performance to detect seizures based on individual video frames (convolutional neural networks, CNNs) or video sequences [CNN+long short-term memory (LSTM) networks]. CNN+LSTM networks based on video sequences outperformed GTCS detection based on individual frames yielding a mean sensitivity of 88% and mean specificity of 92% across patients. The average detection latency after presumed clinical seizure onset was 22 seconds. Detection performance increased as a function of training dataset size. Collectively, we demonstrated that automated video-based GTCS detection with deep learning is feasible and efficacious. Deep learning-based methods may be able to overcome some limitations associated with traditional approaches using hand-crafted features, serve as a benchmark for future methods and analyses, and improve further with larger datasets.

Evolution of Entropy in Art Painting Based on the Wavelet Transform

Quantitative studies of art and aesthetics are representative of interdisciplinary research. In this work, we conducted a large-scale quantitative study of 36,000 paintings covering both Eastern and Western paintings. The information entropy and wavelet entropy of the images were calculated based on their complexity and energy. Wavelet energy entropy is a feature that can characterize rich information in images, and this is the first study to introduce this feature into aesthetic analysis of art paintings. This study shows that the process of entropy change coincides with the development process of art painting. Further, the experimental results demonstrate an important change in the evolution of art painting, and since the rise of modern art in the twentieth century, the entropy values in painting have started to become diverse. In comparison with Western paintings, Eastern paintings have distinct low entropy characteristics in which the wavelet entropy feature of the images has better results in the machine learning classification task of Eastern and Western paintings (i.e., the F1 score can reach 97%). Our study can be the basis for future quantitative analysis and comparative research in the context of Western and Eastern art aesthetics.

THE ESTIMATION OF RESPIRATION RATE BASED ON THE AMPLIFICATION OF RESPIRATION MOTION IN VIDEO

Standard instrument for the clinical diagnosis of sleep apnea is large and based on invasive method, which is not comfortable and not suitable for daily inspection. A video-based measurement method for the respiration rate (RR) is therefore proposed, which is meaningful to the home diagnosis of sleep apnea. We proposed a novel method for the visualization and calculation of RR from a video containing a sleeping person. The video was decomposed by spatio-temporal Laplacian pyramid method into multiresolution image sequences, which were filtered by an infinite-impulse-response bandpass filter to extract the respiration movement in the video. The respiration movement was amplified, and fused into the original video. On the other hand, the signal intensity of the filtering results was compared between layers of Laplacian pyramid to identify the layer with the strongest movement caused by respiration. A morphological calculation was conducted on the image reshaped from the filtered results in this layer, to find the region of interest (ROI) with most significant movement of respiration. The image intensity in the ROI was spatially averaged into a one-dimensional signal, of which the frequency domain was analyzed to obtain RR. The ROI and the calculation results for RR were visualized on the video with enhanced respiration movement. Ten videos lasting 30–60[Formula: see text]s were recorded by a general webcam. The respiration movement of the subject was successfully extracted and amplified, no matter the posture was supine or side lying. The thoracic and abdominal parts were generally identified as ROI in all postures. RR was calculated by the frequency domain analysis for the averaged image intensity in ROI with the error no more than 1 time per minute, and further, as well as ROI, was fused into the amplified video. The region of respiration movement and RR is calculated by the noncontact method, and well visualized in a video. The method provides a novel screening tool for the population suspected of sleep apnea, and is meaningful to the home diagnosis of sleep illness.

Voxyvi: A system for long-term audio and video acquisitions in neonatal intensive care units

BACKGROUND

In the European Union, 300,000 newborn babies are born prematurely every year. Their care is ensured in Neonatal Intensive Care Units (NICU) where vital signs are constantly monitored. In addition, other descriptors such as motion, facial and vocal activities have been shown to be essential to assess neurobehavioral development.

AIM

In the scope of the European project Digi-NewB, we aimed to develop and evaluate a new audio-video device designed to non-invasively acquire multi-modal data (audio, video and thermal images), while fitting the wide variety of bedding environment in NICU.

METHODS

Firstly, a multimodal system and associated software and guidelines to collect data in neonatal intensive care unit were proposed. Secondly, methods for post-evaluation of the acquisition phase were developed, including the study of clinician feedback and a qualitative analysis of the data.

RESULTS

The deployment of 19 acquisition devices in six French hospitals allowed to record more than 500 newborns of different gestational and postmenstrual ages. After the acquisition phase, clinical feedback was mostly positive. In addition, quality of more than 300 recordings was inspected and showed that 77% of the data is exploitable. In depth, the percentage of sole presence of the newborn was estimated at 62% within recordings.

CONCLUSIONS

This study demonstrates that audio-video acquisitions are feasible on a large scale in real life in NICU. The experience also allowed us to make a clear observation of the requirements and challenges that will have to be overcome in order to set up audio-video monitoring methods.

Seizures in the neonate: A review of etiologies and outcomes

Neonatal seizures occur in their majority in close temporal relation to an acute brain injury or systemic insult, and are accordingly defined as acute symptomatic or provoked seizures. However less frequently, unprovoked seizures may also present in the neonatal period as secondary to structural brain abnormalities, thus corresponding to structural epilepsies, or to genetic conditions, thus corresponding to genetic epilepsies. Unprovoked neonatal seizures should be thus considered as the clinical manifestation of early onset structural or genetic epilepsies that often have the characteristics of early onset epileptic encephalopathies. In this review, we address the conundrum of neonatal seizures including acute symptomatic, remote symptomatic, provoked, and unprovoked seizures, evolving to post-neonatal epilepsies, and neonatal onset epilepsies. The different clinical scenarios involving neonatal seizures, each with their distinct post-neonatal evolution are presented. The structural and functional impact of neonatal seizures on brain development and the concept of secondary epileptogenesis, with or without a following latent period after the acute seizures, are addressed. Finally, we underline the need for an early differential diagnosis between an acute symptomatic seizure and an unprovoked seizure, since it is associated with fundamental differences in clinical evolution. These are crucial aspects for neonatal management, counselling and prognostication. In view of the above aspects, we provide an outlook on future strategies and potential lines of research in this field.

Epileptic Seizure Detection and Experimental Treatment: A Review

One-fourths of the patients have medication-resistant seizures and require seizure detection and treatment continuously to cope with sudden seizures. Seizures can be detected by monitoring the brain and muscle activities, heart rate, oxygen level, artificial sounds, or visual signatures through EEG, EMG, ECG, motion, or audio/video recording on the human head and body. In this article, we first discuss recent advances in seizure sensing, signal processing, time- or frequency-domain analysis, and classification algorithms to detect and classify seizure stages. Then, we show a strong potential of applying recent advancements in non-invasive brain stimulation technology to treat seizures. In particular, we explain the fundamentals of brain stimulation approaches, including (1) transcranial magnetic stimulation (TMS), (2) transcranial direct current stimulation (tDCS), (3) transcranial focused ultrasound stimulation (tFUS), and how to use them to treat seizures. Through this review, we intend to provide a broad view of both recent seizure diagnoses and treatments. Such knowledge would help fresh and experienced researchers to capture the advancements in sensing, detection, classification, and treatment seizures. Last but not least, we provide potential research directions that would attract seizure researchers/engineers in the field.

Video and audio processing in paediatrics: a review

OBJECTIVE

Video and sound acquisition and processing technologies have seen great improvements in recent decades, with many applications in the biomedical area. The aim of this paper is to review the overall state of the art of advances within these topics in paediatrics and to evaluate their potential application for monitoring in the neonatal intensive care unit (NICU).

APPROACH

For this purpose, more than 150 papers dealing with video and audio processing were reviewed. For both topics, clinical applications are described according to the considered cohorts-full-term newborns, infants and toddlers or preterm newborns. Then, processing methods are presented, in terms of data acquisition, feature extraction and characterization.

MAIN RESULTS

The paper first focuses on the exploitation of video recordings; these began to be automatically processed in the 2000s and we show that they have mainly been used to characterize infant motion. Other applications, including respiration and heart rate estimation and facial analysis, are also presented. Audio processing is then reviewed, with a focus on the analysis of crying. The first studies in this field focused on induced-pain cries and the newest ones deal with spontaneous cries; the analyses are mainly based on frequency features. Then, some papers dealing with non-cry signals are also discussed.

SIGNIFICANCE

Finally, we show that even if recent improvements in digital video and signal processing allow for increased automation of processing, the context of the NICU makes a fully automated analysis of long recordings problematic. A few proposals for overcoming some of the limitations are given.

Toward a Personalized Real-Time Diagnosis in Neonatal Seizure Detection

The problem of creating a personalized seizure detection algorithm for newborns is tackled in this paper. A probabilistic framework for semi-supervised adaptation of a generic patient-independent neonatal seizure detector is proposed. A system that is based on a combination of patient-adaptive (generative) and patient-independent (discriminative) classifiers is designed and evaluated on a large database of unedited continuous multichannel neonatal EEG recordings of over 800 h in duration. It is shown that an improvement in the detection of neonatal seizures over the course of long EEG recordings is achievable with on-the-fly incorporation of patient-specific EEG characteristics. In the clinical setting, the employment of the developed system will maintain a seizure detection rate at 70% while halving the number of false detections per hour, from 0.4 to 0.2 FD/h. This is the first study to propose the use of online adaptation without clinical labels, to build a personalized diagnostic system for the detection of neonatal seizures.

Non-EEG seizure detection systems and potential SUDEP prevention: State of the art: Review and update

PURPOSE

Detection of, and alarming for epileptic seizures is increasingly demanded and researched. Our previous review article provided an overview of non-invasive, non-EEG (electro-encephalography) body signals that can be measured, along with corresponding methods, state of the art research, and commercially available systems. Three years later, many more studies and devices have emerged. Moreover, the boom of smart phones and tablets created a new market for seizure detection applications.

METHOD

We performed a thorough literature review and had contact with manufacturers of commercially available devices.

RESULTS

This review article gives an updated overview of body signals and methods for seizure detection, international research and (commercially) available systems and applications. Reported results of non-EEG based detection devices vary between 2.2% and 100% sensitivity and between 0 and 3.23 false detections per hour compared to the gold standard video-EEG, for seizures ranging from generalized to convulsive or non-convulsive focal seizures with or without loss of consciousness. It is particularly interesting to include monitoring of autonomic dysfunction, as this may be an important pathophysiological mechanism of SUDEP (sudden unexpected death in epilepsy), and of movement, as many seizures have a motor component.

CONCLUSION

Comparison of research results is difficult as studies focus on different seizure types, timing (night versus day) and patients (adult versus pediatric patients). Nevertheless, we are convinced that the most effective seizure detection systems are multimodal, combining for example detection methods for movement and heart rate, and that devices should especially take into account the user's seizure types and personal preferences.

Monitoring of newborns at high risk for brain injury

Due to the increasing number of surviving preterm newborns and to the recognition of therapeutic hypothermia as the current gold standard in newborns with hypoxic-ischaemic encephalopathy, there has been a growing interest in the implementation of brain monitoring tools in newborns at high risk for neurological disorders.Among the most frequent neurological conditions and presentations in the neonatal period, neonatal seizures and neonatal status epilepticus, paroxysmal non-epileptic motor phenomena, hypoxic-ischaemic encephalopathy, white matter injury of prematurity and stroke require specific approaches to diagnosis. In this review we will describe the characteristics, aims, indications and limitations of routinely available diagnostic techniques such as conventional and amplitude-integrated EEG, evoked potentials, cranial ultrasound and brain MRI. We will conclude by briefly outlining potential future perspectives from research studies.

A neural network based infant monitoring system to facilitate diagnosis of epileptic seizures

In this paper, we propose an infant monitoring system that automatically detects epileptic seizures in domestic and hospital environments. The proposed system measures the movements and electroencephalogram (EEG) signals of an infant using a video camera and an electroencephalograph. Seizure features are then extracted from the video images and EEG signals, and the evaluation indices based on medical knowledge are calculated from the features. The system employs a probabilistic neural network for the automatic detection of seizures, thereby allowing the choice/combination of evaluation indices appropriate for the environment via network training. We tested the system in simulated domestic and hospital environments. The validity of the proposed system was reinforced by the results of comparisons with clinical diagnoses.