A Developed LSTM-Ladder-Network-Based Model for Sleep Stage Classification.

Sleep staging is crucial for diagnosing sleep-related disorders. The heavy and time-consuming task of manual staging can be released by automatic techniques. However, the automatic staging model would have a relatively poor performance when working on unseen new data due to individual differences. In this research, a developed LSTM-Ladder-Network (LLN) model is proposed for automatic sleep stage classification. Several features are extracted for each epoch and combined with the following epochs to form a cross-epoch vector. The long short-term memory (LSTM) network is added into the basic ladder network (LN) to learn the sequential information of adjacent epochs. The developed model is implemented based on a transductive learning scheme to avoid the issue of accuracy loss caused by individual differences. In this process, the labeled data pre-trains the encoder, and the unlabeled data re- fine the model parameters by minimizing the reconstruction loss. The proposed model is evaluated on the data from public database and hospital. Comparison experiments were conducted where the developed LLN model achieved rather satisfied performance while dealing with the unseen new data. The obtained results demonstrate the effectiveness of the proposed approach in addressing individual differences. This can improve the quality of automatic sleep staging when assessed on different individuals and has strong application potential as a computer aided approach for sleep staging.

Automatic interpretation and writing report of the adult waking electroencephalogram.

Automatic interpretation of the EEG has so far been faced with significant difficulties because of a large amount of spatial as well as temporal information contained in the EEG, continuous fluctuation of the background activity depending on changes in the subject's vigilance and attention level, the occurrence of paroxysmal activities such as spikes and spike-and-slow-waves, contamination of the EEG with a variety of artefacts and the use of different recording electrodes and montages. Therefore, previous attempts of automatic EEG interpretation have been focussed only on a specific EEG feature such as paroxysmal abnormalities, delta waves, sleep stages and artefact detection. As a result of a long-standing cooperation between clinical neurophysiologists and system engineers, we report for the first time on a comprehensive, computer-assisted, automatic interpretation of the adult waking EEG. This system analyses the background activity, intermittent abnormalities, artefacts and the level of vigilance and attention of the subject, and automatically presents its report in written form. Besides, it also detects paroxysmal abnormalities and evaluates the effects of intermittent photic stimulation and hyperventilation on the EEG. This system of automatic EEG interpretation was formed by adopting the strategy that the qualified EEGers employ for the systematic visual inspection. This system can be used as a supplementary tool for the EEGer's visual inspection, and for educating EEG trainees and EEG technicians.

An automatic spike detection system based on elimination of false positives using the large-area context in the scalp EEG.

Most automatic spike detection systems in the scalp electroencephalogram (EEG) focused on the characteristics of "spike." However, the characteristics of "false positives" (FPs) have not been fully studied. In this paper, we proposed a system that contains a series of algorithms to eliminate FPs and a template method to confirm spikes. The system used large area context available on 49 channels from two common montages. The impact of slow-waves after spikes was taken into consideration, as well as the information from single channel, multichannel, and whole recording. Two types of FPs were identified in this paper. The ones from typical artifacts were identified by analysis of background EEG activities, and the ones from other EEG activities were declared by spatial and temporal characteristics of spike activities. Finally, a multichannel template method was used to assess the performance of the proposed system. The system was evaluated using 17 routine EEG recordings. Spike activities were observed in six of them. Effective multichannel templates were extracted from four recordings containing frequent spikes. The least selectivity was 92.6% and the most false positive rate was 0.26 min(-1). Proposed algorithms for elimination of FPs are also suitable for other algorithms to enhance performance since most FPs can be identified while few true spikes are eliminated.

Automatic interpretation of hyperventilation-induced electroencephalogram constructed in the way of qualified electroencephalographer's visual inspection.

Quantitative analysis and detection of electroencephalogram (EEG) recordings during evoked activities is essential for clinical diagnosis on neurological disorders. However, the process of interpreting EEG is time consuming for electroencephalographers (EEGers). In this study, an automatic EEG interpretation system constructed in the way of qualified EEGer's visual inspection was proposed. The system was applied to interpret hyperventilation-induced EEG automatically. The final results of automatic interpretation were compared with EEGer's visual inspection, and showed high consistence.

Technical quality evaluation of EEG recording based on electroencephalographers' knowledge.

The aim of this study is to develop a technical quality evaluation system of electroencephalogram (EEG) recording in order to acquire technically satisfactory EEG records, which may contribute to the accuracy improvement of EEG interpretation. In our developed system, the evaluation of EEG recording comprises the detection of technical artifacts and physiological status, which indicates the recording status objectively. In addition, the caution signals to users are generated in the system according to the undesired status detected. The information displayed to users includes the updated EEG records and instant evaluation results. Two examples of evaluation results are introduced in this paper, illustrating unsatisfactory records and artifact free records, respectively. The experimental results are proposed to verify the effectiveness of the technical quality evaluation of EEG recording. The implementation of the technical quality evaluation of EEG recording is helpful to acquire technically satisfactory EEG records, which may improve the accuracy of results in both the visual and the automatic EEG interpretation, and ease the laborious work of EEG technicians in the recording progress.

Adaptive EEG spike detection: determination of threshold values based on conditional probability.

Determination of the threshold value for automatic EEG spike detection was investigated adopting conditional probability. An adaptive spike detection method was constructed and evaluated. A discriminant function for detecting spikes was obtained by conditional probability calculated from the EEG spike data. The relationship among false-negatives, false-positives and threshold values for the discriminant function was investigated. An adaptive detection algorithm was developed by combining different threshold values. False-negative and false-positive rates for spike detection depended on the threshold values. The adaptive spike detection algorithm achieved a high detection rate and accuracy. The advantage of the proposed method is to construct an adaptive detection algorithm by combining the threshold values according to the purpose of spike detection. Since the threshold can be easily changed in the proposed method, it is practically effective for clinical use.