Evaluation of machine learning approaches for cell-type identification from single-cell transcriptomics data.

Single-cell transcriptomics technologies have vast potential in advancing our understanding of cellular heterogeneity in complex tissues. While methods to interpret single-cell transcriptomics data are developing rapidly, challenges in most analysis pipeline still remain, and the major limitation is a reliance on manual annotations for cell-type identification that is time-consuming, irreproducible, and sometimes lack canonical markers for certain cell types. There is a growing realization of the potential of machine learning models as a supervised classification approach that can significantly aid decision-making processes for cell-type identification. In this work, we performed a comprehensive and impartial evaluation of 10 machine learning models that automatically assign cell phenotypes. The performance of classification methods is estimated by using 20 publicly accessible single-cell RNA sequencing datasets with different sizes, technologies, species and levels of complexity. The performance of each model for within dataset (intra-dataset) and across datasets (inter-dataset) experiments based on the classification accuracy and computation time are both evaluated. Besides, the sensitivity to the number of input features, different annotation levels and dataset complexity was also been estimated. Results showed that most classifiers perform well on a variety of datasets with decreased accuracy for complex datasets, while the Linear Support Vector Machine (linear-SVM) and Logistic Regression classifier models have the best overall performance with remarkably fast computation time. Our work provides a guideline for researchers to select and apply suitable machine learning-based classification models in their analysis workflows and sheds some light on the potential direction of future improvement on automated cell phenotype classification tools based on the single-cell sequencing data.

EEG-based trial-by-trial texture classification during active touch.

Trial-by-trial texture classification analysis and identifying salient texture related EEG features during active touch that are minimally influenced by movement type and frequency conditions are the main contributions of this work. A total of twelve healthy subjects were recruited. Each subject was instructed to use the fingertip of their dominant hand's index finger to rub or tap three textured surfaces (smooth flat, medium rough, and rough) with three levels of movement frequency (approximately 2, 1 and 0.5 Hz). EEG and force data were collected synchronously during each touch condition. A systematic feature selection process was performed to select temporal and spectral EEG features that contribute to texture classification but have low contribution towards movement type and frequency classification. A tenfold cross validation was used to train two 3-class (each for texture and movement frequency classification) and a 2-class (movement type) Support Vector Machine classifiers. Our results showed that the total power in the mu (8-15 Hz) and beta (16-30 Hz) frequency bands showed high accuracy in discriminating among textures with different levels of roughness (average accuracy > 84%) but lower contribution towards movement type (average accuracy < 65%) and frequency (average accuracy < 58%) classification.

Sample Entropy on Multidistance Signal Level Difference for Epileptic EEG Classification.

Epilepsy is a disorder of the brain's nerves as a result of excessive brain cell activity. It is generally characterized by the recurrent unprovoked seizures. This neurological abnormality can be detected and evaluated using Electroencephalogram (EEG) signal. Many algorithms have been applied to achieve high performance for the EEG classification of epileptic. However, the complexity and randomness of EEG signals become a challenge to researchers in applying the appropriate algorithms. In this research, sample entropy on Multidistance Signal Level Difference (MSLD) was applied to obtain the characteristic of EEG signals, especially towards the epilepsy patients. The test was performed on three classes of EEG data: EEG signals of epilepsy patient in ictal (seizure), interictal conditions (occurring between seizures) and normal EEG signals from healthy subjects with a closed eye condition. In this study, classification and verification were done using the Support Vector Machine (SVM) method. Through the 5-fold cross-validation, experimental results showed the highest accuracy of 97.7%.

Predicting Autism Spectrum Disorder Using Domain-Adaptive Cross-Site Evaluation.

The advances in neuroimaging methods reveal that resting-state functional fMRI (rs-fMRI) connectivity measures can be potential diagnostic biomarkers for autism spectrum disorder (ASD). Recent data sharing projects help us replicating the robustness of these biomarkers in different acquisition conditions or preprocessing steps across larger numbers of individuals or sites. It is necessary to validate the previous results by using data from multiple sites by diminishing the site variations. We investigated partial least square regression (PLS), a domain adaptive method to adjust the effects of multicenter acquisition. A sparse Multivariate Pattern Analysis (MVVPA) framework in a leave one site out cross validation (LOSOCV) setting has been proposed to discriminate ASD from healthy controls using data from six sites in the Autism Brain Imaging Data Exchange (ABIDE). Classification features were obtained using 42 bilateral Brodmann areas without presupposing any prior hypothesis. Our results showed that using PLS, SVM showed poorer accuracies with highest accuracy achieved (62%) than without PLS but not significantly. The regions occurred in two or more informative connections are Dorsolateral Prefrontal Cortex, Somatosensory Association Cortex, Primary Auditory Cortex, Inferior Temporal Gyrus and Temporopolar area. These interrupted regions are involved in executive function, speech, visual perception, sense and language which are associated with ASD. Our findings may support early clinical diagnosis or risk determination by identifying neurobiological markers to distinguish between ASD and healthy controls.

Inferring imagined speech using EEG signals: a new approach using Riemannian manifold features.

OBJECTIVE: In this paper, we investigate the suitability of imagined speech for brain-computer interface (BCI) applications. APPROACH: A novel method based on covariance matrix descriptors, which lie in Riemannian manifold, and the relevance vector machines classifier is proposed. The method is applied on electroencephalographic (EEG) signals and tested in multiple subjects. MAIN RESULTS: The method is shown to outperform other approaches in the field with respect to accuracy and robustness. The algorithm is validated on various categories of speech, such as imagined pronunciation of vowels, short words and long words. The classification accuracy of our methodology is in all cases significantly above chance level, reaching a maximum of 70% for cases where we classify three words and 95% for cases of two words. SIGNIFICANCE: The results reveal certain aspects that may affect the success of speech imagery classification from EEG signals, such as sound, meaning and word complexity. This can potentially extend the capability of utilizing speech imagery in future BCI applications. The dataset of speech imagery collected from total 15 subjects is also published.

A review and experimental study on the application of classifiers and evolutionary algorithms in EEG-based brain-machine interface systems.

OBJECTIVE: Considering the importance and the near-future development of noninvasive brain-machine interface (BMI) systems, this paper presents a comprehensive theoretical-experimental survey on the classification and evolutionary methods for BMI-based systems in which EEG signals are used. APPROACH: The paper is divided into two main parts. In the first part, a wide range of different types of the base and combinatorial classifiers including boosting and bagging classifiers and evolutionary algorithms are reviewed and investigated. In the second part, these classifiers and evolutionary algorithms are assessed and compared based on two types of relatively widely used BMI systems, sensory motor rhythm-BMI and event-related potentials-BMI. Moreover, in the second part, some of the improved evolutionary algorithms as well as bi-objective algorithms are experimentally assessed and compared. MAIN RESULTS: In this study two databases are used, and cross-validation accuracy (CVA) and stability to data volume (SDV) are considered as the evaluation criteria for the classifiers. According to the experimental results on both databases, regarding the base classifiers, linear discriminant analysis and support vector machines with respect to CVA evaluation metric, and naive Bayes with respect to SDV demonstrated the best performances. Among the combinatorial classifiers, four classifiers, Bagg-DT (bagging decision tree), LogitBoost, and GentleBoost with respect to CVA, and Bagging-LR (bagging logistic regression) and AdaBoost (adaptive boosting) with respect to SDV had the best performances. Finally, regarding the evolutionary algorithms, single-objective invasive weed optimization (IWO) and bi-objective nondominated sorting IWO algorithms demonstrated the best performances. SIGNIFICANCE: We present a general survey on the base and the combinatorial classification methods for EEG signals (sensory motor rhythm and event-related potentials) as well as their optimization methods through the evolutionary algorithms. In addition, experimental and statistical significance tests are carried out to study the applicability and effectiveness of the reviewed methods.

Multi-feature classifiers for burst detection in single EEG channels from preterm infants.

OBJECTIVE: The study of electroencephalographic (EEG) bursts in preterm infants provides valuable information about maturation or prognostication after perinatal asphyxia. Over the last two decades, a number of works proposed algorithms to automatically detect EEG bursts in preterm infants, but they were designed for populations under 35 weeks of post menstrual age (PMA). However, as the brain activity evolves rapidly during postnatal life, these solutions might be under-performing with increasing PMA. In this work we focused on preterm infants reaching term ages (PMA ⩾36 weeks) using multi-feature classification on a single EEG channel. APPROACH: Five EEG burst detectors relying on different machine learning approaches were compared: logistic regression (LR), linear discriminant analysis (LDA), k-nearest neighbors (kNN), support vector machines (SVM) and thresholding (Th). Classifiers were trained by visually labeled EEG recordings from 14 very preterm infants (born after 28 weeks of gestation) with 36-41 weeks PMA. MAIN RESULTS: The most performing classifiers reached about 95% accuracy (kNN, SVM and LR) whereas Th obtained 84%. Compared to human-automatic agreements, LR provided the highest scores (Cohen's kappa = 0.71) using only three EEG features. Applying this classifier in an unlabeled database of 21 infants ⩾36 weeks PMA, we found that long EEG bursts and short inter-burst periods are characteristic of infants with the highest PMA and weights. SIGNIFICANCE: In view of these results, LR-based burst detection could be a suitable tool to study maturation in monitoring or portable devices using a single EEG channel.

Grading hypoxic-ischemic encephalopathy severity in neonatal EEG using GMM supervectors and the support vector machine.

OBJECTIVE: This work presents a novel automated system to classify the severity of hypoxic-ischemic encephalopathy (HIE) in neonates using EEG. METHODS: A cross disciplinary method is applied that uses the sequences of short-term features of EEG to grade an hour long recording. Novel post-processing techniques are proposed based on majority voting and probabilistic methods. The proposed system is validated with one-hour-long EEG recordings from 54 full term neonates. RESULTS: An overall accuracy of 87% is achieved. The developed grading system has improved both the accuracy and the confidence/quality of the produced decision. With a new label 'unknown' assigned to the recordings with lower confidence levels an accuracy of 96% is attained. CONCLUSION: The statistical long-term model based features extracted from the sequences of short-term features has improved the overall accuracy of grading the HIE injury in neonatal EEG. SIGNIFICANCE: The proposed automated HIE grading system can provide significant assistance to healthcare professionals in assessing the severity of HIE. This represents a practical and user friendly implementation which acts as a decision support system in the clinical environment. Its integration with other EEG analysis algorithms may improve neonatal neurocritical care.