Continual learning approaches for single cell RNA sequencing data.

Single-cell RNA sequencing data is among the most interesting and impactful data of today and the sizes of the available datasets are increasing drastically. There is a substantial need for learning from large datasets, causing nontrivial challenges, especially in hardware. Loading even a single dataset into the memory of an ordinary, off-the-shelf computer can be infeasible, and using computing servers might not always be an option. This paper presents continual learning as a solution to such hardware bottlenecks. The findings of cell-type classification demonstrate that XGBoost and Catboost algorithms, when implemented in a continual learning framework, exhibit superior performance compared to the best-performing static classifier. We achieved up to 10% higher median F1 scores than the state-of-the-art on the most challenging datasets. On the other hand, these algorithms can suffer from variations in data characteristics across diverse datasets, pointing out indications of the catastrophic forgetting problem.

Effect of distance measures on confidences of t-SNE embeddings and its implications on clustering for scRNA-seq data.

Arguably one of the most famous dimensionality reduction algorithms of today is t-distributed stochastic neighbor embedding (t-SNE). Although being widely used for the visualization of scRNA-seq data, it is prone to errors as any algorithm and may lead to inaccurate interpretations of the visualized data. A reasonable way to avoid misinterpretations is to quantify the reliability of the visualizations. The focus of this work is first to find the best possible way to predict sample-based confidence scores for t-SNE embeddings and next, to use these confidence scores to improve the clustering algorithms. We adopt an RF regression algorithm using seven distance measures as features for having the sample-based confidence scores with a variety of different distance measures. The best configuration is used to assess the clustering improvement using K-means and Density-Based Spatial Clustering of Applications with Noise (DBSCAN) based on Adjusted Rank Index (ARI), Normalized Mutual Information (NMI), and accuracy (ACC) scores. The experimental results show that distance measures have a considerable effect on the precision of confidence scores and clustering performance can be improved substantially if these confidence scores are incorporated before the clustering algorithm. Our findings reveal the usefulness of these confidence scores on downstream analyses for scRNA-seq data.

The importance of mentalization, coping mechanisms, and perceived stress in the prediction of resilience of healthcare workers.

Resilience is the process of overcoming stressors. Being able to examine the effect of the Covid epidemic on healthcare workers (HCWs) has provided us a unique opportunity to understand the impact of trauma on resilience. We aimed to investigate the relationship between stress, mentalization, and an individual's coping capacity against a real risk (Covid-19) and evaluate the predictors of resilience. 302 HCWs have enrolled in the study and completed an online questionnaire assessing demographics, perceived stress, resilience, coping, and mentalization. We utilized statistical analysis together with a Random Forest classifier to analyze the interaction between these factors extensively. We applied ten times ten-fold cross-validation and plotted Receiver Operator Characteristic (ROC) with the calculated Area Under the Curve(AUC) score and identify the most important features. Our experiments showed that the Perceived stress scale has the strongest relationship with resilience. The subject's awareness level of emotional states is an important factor that determines the level of resilience. Coping styles such as the decision of giving up is also a crucial indicator. We conclude that being aware of the risks and the mental states are the dominant factors behind the resilience levels of healthcare workers under pandemic conditions.

Exploiting lamina terminalis appearance and motion in prediction of hydrocephalus using convolutional LSTM network.

BACKGROUND: Evaluation of the lamina terminalis (LT) is crucial for non-invasive evaluation of the CSF diversion for the treatment of hydrocephalus. Together with deep learning algorithms, morphological and physiological analyses of the LT may play an important role in the management of hydrocephalus. AIM: We aim to show that exploiting the motion of LT can contribute to the evaluation of hydrocephalus using deep learning algorithms. METHODS: The dataset contains 61 True-fisp data with routine sequences 37 of which are labeled as 'hydrocephalus' and the others as 'normal condition'. A fifteen-year experienced neuroradiologist divided data into two groups. The first group, 'hydrocephalus', consists of patients with typical MRI findings (ventriculomegaly, enlargement of the third ventricular recesses and lateral ventricular horns, decreased mamillo-pontine distance, reduced frontal horn angle, thinning/elevation of the corpus callosum, and non-dilated convexity sulci), and the second group contains samples that did not show any symptoms or neurologic abnormality and labeled as 'normal condition'. The region of interest was determined by the radiologist supervisor to cover the LT. To achieve our purpose, we used both spatial and spatio-temporal analysis with two different deep learning architectures. We utilized Convolutional Neural Networks (CNN) for spatial and Convolutional Long Short-Term Memory (ConvLSTM) models for spatio-temporal analysis using an ROI around LT on sagittal True-fisp images. RESULTS: Our results show that 80.7% classification accuracy was achieved with the ConvLSTM model exploiting LT motion, whereas 76.5% and 71.6% accuracies were obtained by the 2D CNN model using all frames, and only the first frame from only spatial information, respectively. CONCLUSION: We suggest that the motion of the LT can be used as an additional attribute to the spatial information to evaluate the hydrocephalus.

Serum TNF- Related Weak Inducer of Apoptosis (TWEAK), TNF- Related Apoptosis-Inducing Ligand (TRAIL) Levels in Patients with Bipolar Depression, Major Depression and a Healthy Control Group.

BACKGROUND: A low-grade inflammation is presumed to be related to the etiopathogenesis of major depressive disorder (MDD) and bipolar disorder. Tumor necrosis factor (TNF) superfamily members have roles in the pathogenesis of neuropsychiatric disorders because of the relationship with inflammation and neurogenesis. The aim of this study was to investigate the serum TNF-related weak inducer of apoptosis (TWEAK) and TNF-related apoptosis-inducing ligand (TRAIL) levels in patients with bipolar depression (BD), MDD and a healthy control (HC) group to determine any differences between MDD and BD in terms of inflammation biomarkers. SUBJECTS AND METHODS: After a 12-hour overnight fast, 5 milliliter (mL) samples of fasting blood were obtained from the participants. The TWEAK and TRAIL plasma levels were calculated using ELISA kits. RESULTS: The TWEAK levels were found to be higher in the BD group than in the HC group (p=0.03). No statistically significant differences were determined between the BD vs MDD and MDD vs HC groups (p=0.17, p=0.37, respectively). There were no statistically significant differences between the three groups (BD vs HC; BD vs MDD; MDD vs HC) in terms of TRAIL levels (p=0.21). CONCLUSION: To the best of our knowledge, this study is the first to have explored TWEAK levels in patients with BD. The higher TWEAK levels in BD than in the control group is compatible with the inflammation hypothesis of BD. Limitations of the study were the differences in medications of the patient groups and that it was a cross-sectional study. There is a need for further longitudinal studies with larger sample size and medication-free patients.

An Overview of Deep Learning Algorithms and Their Applications in Neuropsychiatry.

Deep learning (DL) algorithms have achieved important successes in data analysis tasks, thanks to their capability of revealing complex patterns in data. With the advance of new sensors, data storage, and processing hardware, DL algorithms start dominating various fields including neuropsychiatry. There are many types of DL algorithms for different data types from survey data to functional magnetic resonance imaging scans. Because of limitations in diagnosing, estimating prognosis and treatment response of neuropsychiatric disorders; DL algorithms are becoming promising approaches. In this review, we aim to summarize the most common DL algorithms and their applications in neuropsychiatry and also provide an overview to guide the researchers in choosing the proper DL architecture for their research.

Comparison of morphometric parameters in prediction of hydrocephalus using random forests.

Ventricles of the human brain enlarge with aging, neurodegenerative diseases, intrinsic, and extrinsic pathologies. The morphometric examination of neuroimages is an effective approach to assess structural changes occurring due to diseases such as hydrocephalus. In this study, we explored the effectiveness of commonly used morphological parameters in hydrocephalus diagnosis. For this purpose, the effect of six common morphometric parameters; Frontal Horns' Length (FHL), Maximum Lateral Length (MLL), Biparietal Diameter (BPD), Evans' Ratio (ER), Cella Media Ratio (CMR), and Frontal Horns' Ratio (FHR) were compared in terms of their importance in predicting hydrocephalus using a Random Forest classifier. The experimental results demonstrated that hydrocephalus can be detected with 91.46 % accuracy using all of these measurements. The accuracy of classification using only CMR and FHL reached up to 93.33 %. In terms of individual performances, CMR and FHL were the top performers whereas BPD and FHR did not contribute as much to the overall accuracy.

Quantitative error prediction of medical image registration using regression forests.

Predicting registration error can be useful for evaluation of registration procedures, which is important for the adoption of registration techniques in the clinic. In addition, quantitative error prediction can be helpful in improving the registration quality. The task of predicting registration error is demanding due to the lack of a ground truth in medical images. This paper proposes a new automatic method to predict the registration error in a quantitative manner, and is applied to chest CT scans. A random regression forest is utilized to predict the registration error locally. The forest is built with features related to the transformation model and features related to the dissimilarity after registration. The forest is trained and tested using manually annotated corresponding points between pairs of chest CT scans in two experiments: SPREAD (trained and tested on SPREAD) and inter-database (including three databases SPREAD, DIR-Lab-4DCT and DIR-Lab-COPDgene). The results show that the mean absolute errors of regression are 1.07  ±â€¯ 1.86 and 1.76  ±â€¯ 2.59 mm for the SPREAD and inter-database experiment, respectively. The overall accuracy of classification in three classes (correct, poor and wrong registration) is 90.7% and 75.4%, for SPREAD and inter-database respectively. The good performance of the proposed method enables important applications such as automatic quality control in large-scale image analysis.

Confidence Estimation for Medical Image Registration Based On Stereo Confidences.

In this paper, we propose a novel method to estimate the confidence of a registration that does not require any ground truth, is independent from the registration algorithm and the resulting confidence is correlated with the amount of registration error. We first apply a local search to match patterns between the registered image pairs. Local search induces a cost space per voxel which we explore further to estimate the confidence of the registration similar to confidence estimation algorithms for stereo matching. We test our method on both synthetically generated registration errors and on real registrations with ground truth. The experimental results show that our confidence measure can estimate registration errors and it is correlated with local errors.