Data-driven cluster analysis of insomnia disorder with physiology-based qEEG variables

Insomnia nosology: a systematic review and critical appraisal of historical diagnostic categories and current phenotypes

Insomnia nosology has significantly evolved since the Diagnostic and Statistical Manual (DSM)-III-R first distinguished between 'primary' and 'secondary' insomnia. Prior International Classification of Sleep Disorders (ICSD) nosology 'split' diagnostic phenotypes to address insomnia's heterogeneity and the DSM nosology 'lumped' them into primary insomnia, while both systems assumed causality for insomnia secondary to health conditions. In this systematic review, we discuss the historical phenotypes in prior insomnia nosology, present findings for currently proposed insomnia phenotypes based on more robust approaches, and critically appraise the most relevant ones. Electronic databases PsychINFO, PubMED, Web of Science, and references of eligible articles, were accessed to find diagnostic manuals, literature on insomnia phenotypes, including systematic reviews or meta-analysis, and assessments of the reliability or validity of insomnia diagnoses, identifying 184 articles. The data show that previous insomnia diagnoses lacked reliability and validity, leading current DSM-5-TR and ICSD-3 nosology to 'lump' phenotypes into a single diagnosis comorbid with health conditions. However, at least two new, robust insomnia phenotyping approaches were identified. One approach is multidimensional-multimethod and provides evidence for self-reported insomnia with objective short versus normal sleep duration linked to clinically relevant outcomes, while the other is multidimensional and provides evidence for two to five clusters (phenotypes) based on self-reported trait, state, and/or life-history data. Some approaches still need replication to better support whether their findings identify true phenotypes or simply different patterns of symptomatology. Regardless, these phenotyping efforts aim at improving insomnia nosology both as a classification system and as a mechanism to guide treatment.

Automatic identification of insomnia using optimal antisymmetric biorthogonal wavelet filter bank with ECG signals

Sleep is a fundamental human physiological activity required for adequate working of the human body. Sleep disorders such as sleep movement disorders, nocturnal front lobe epilepsy, insomnia, and narcolepsy are caused due to low sleep quality. Insomnia is one such sleep disorder where a person has difficulty in getting quality sleep. There is no definitive test to identify insomnia; hence it is essential to develop an automated system to identify it accurately. A few automated methods have been proposed to identify insomnia using either polysomnogram (PSG) or electroencephalogram (EEG) signals. To the best of our knowledge, we are the first to automatically detect insomnia using only electrocardiogram (ECG) signals without combining them with any other physiological signals. In the proposed study, an optimal antisymmetric biorthogonal wavelet filter bank (ABWFB) has been used, which is designed to minimize the joint duration-bandwidth localization (JDBL) of the underlying filters. The L₁-norm feature is computed from the various wavelet sub-bands coefficients of ECG signals. The L₁ norm features are fed to various supervised machine learning classifiers for the automated detection of insomnia. In this work, ECG recordings of seven insomnia patients and six normal subjects are used from the publicly available cyclic alternating pattern (CAP) sleep database. We created ten different subsets of ECG signals based on annotations of sleep-stages, namely wake (W), S1, S2, S3, S4, rapid eye moment (REM), light sleep stage (LSS), slow-wave sleep (SWS), non-rapid eye movement (NREM) and W + S1+S2+S3+S4+REM for the automated identification of insomnia. Our proposed ECG-based system obtained the highest classification accuracy of 97.87%, F1-score of 97.39%, and Cohen's kappa value of 0.9559 for K-nearest neighbour (KNN) with the ten-fold cross-validation strategy using ECG signals corresponding to the REM sleep stage. The support vector machine (SVM) yielded the highest value of 0.99 for area under the curve with the ten fold cross-validation corresponding to REM sleep stage.