Challenges in Building of Deep Learning Models for Glioblastoma Segmentation: Evidence from Clinical Data.

In this article, we compare the performance of a state-of-the-art segmentation network (UNet) on two different glioblastoma (GB) segmentation datasets. Our experiments show that the same training procedure yields almost twice as bad results on the retrospective clinical data compared to the BraTS challenge data (in terms of Dice score). We discuss possible reasons for such an outcome, including inter-rater variability and high variability in magnetic resonance imaging (MRI) scanners and scanner settings. The high performance of segmentation models, demonstrated on preselected imaging data, does not bring the community closer to using these algorithms in clinical settings. We believe that a clinically applicable deep learning architecture requires a shift from unified datasets to heterogeneous data.

CT-Based COVID-19 triage: Deep multitask learning improves joint identification and severity quantification.

The current COVID-19 pandemic overloads healthcare systems, including radiology departments. Though several deep learning approaches were developed to assist in CT analysis, nobody considered study triage directly as a computer science problem. We describe two basic setups: Identification of COVID-19 to prioritize studies of potentially infected patients to isolate them as early as possible; Severity quantification to highlight patients with severe COVID-19, thus direct them to a hospital or provide emergency medical care. We formalize these tasks as binary classification and estimation of affected lung percentage. Though similar problems were well-studied separately, we show that existing methods could provide reasonable quality only for one of these setups. We employ a multitask approach to consolidate both triage approaches and propose a convolutional neural network to leverage all available labels within a single model. In contrast with the related multitask approaches, we show the benefit from applying the classification layers to the most spatially detailed feature map at the upper part of U-Net instead of the less detailed latent representation at the bottom. We train our model on approximately 1500 publicly available CT studies and test it on the holdout dataset that consists of 123 chest CT studies of patients drawn from the same healthcare system, specifically 32 COVID-19 and 30 bacterial pneumonia cases, 30 cases with cancerous nodules, and 31 healthy controls. The proposed multitask model outperforms the other approaches and achieves ROC AUC scores of 0.87±0.01 vs. bacterial pneumonia, 0.93±0.01 vs. cancerous nodules, and 0.97±0.01 vs. healthy controls in Identification of COVID-19, and achieves 0.97±0.01 Spearman Correlation in Severity quantification. We have released our code and shared the annotated lesions masks for 32 CT images of patients with COVID-19 from the test dataset.

Optimizing Connectivity-Driven Brain Parcellation Using Ensemble Clustering.

This work addresses the problem of constructing a unified, topologically optimal connectivity-based brain atlas. The proposed approach aggregates an ensemble partition from individual parcellations without label agreement, providing a balance between sufficiently flexible individual parcellations and intuitive representation of the average topological structure of the connectome. The methods exploit a previously proposed dense connectivity representation, first performing graph-based hierarchical parcellation of individual brains, and subsequently aggregating the individual parcellations into a consensus parcellation. The search for consensus-based on the hard ensemble (HE) algorithm-approximately minimizes the sum of cluster membership distances, effectively estimating a pseudo-Karcher mean of individual parcellations. Computational stability, graph structure preservation, and biological relevance of the simplified representation resulting from the proposed parcellation are assessed on the Human Connectome Project data set. These aspects are assessed using (1) edge weight distribution divergence with respect to the dense connectome representation, (2) interhemispheric symmetry, (3) network characteristics' stability and agreement with respect to individually and anatomically parcellated networks, and (4) performance of the simplified connectome in a biological sex classification task. Ensemble parcellation was found to be highly stable with respect to subject sampling, outperforming anatomical atlases and other connectome-based parcellations in classification as well as preserving global connectome properties. The HE-based parcellation also showed a degree of symmetry comparable with anatomical atlases and a high degree of spatial contiguity without using explicit priors.

Phytochemical characterization of an adaptogenic preparation from Rhodiola heterodonta.

The phytochemical constituents of a biologically active, standardized, 80% ethanol extract of Rhodiola heterodonta were characterized. The extract was fractionated over a Sephadex LH-20 column to afford two main fractions representing two classes of secondary metabolites: phenylethanoids and proanthocyanidins. This fractionation facilitated the identification and quantification of individual compounds in the fractions and sub-fractions using HPLC, and LC-MS. The major compounds in the phenylethanoid fraction were heterodontoside, tyrosol methyl ether, salidroside, viridoside, mongrhoside, tyrosol, and the cyanogenic glucoside rhodiocyanoside A. These seven compounds comprised 17.4% of the EtOH extract. Proanthocyanidins ranged from oligomers to polymers based on epigallocatechin and gallate units. The main identified oligomeric compounds in the proanthocyanidin fraction were epigallocatechin gallate, epigallocatechin-epigallocatechin-3-O-gallate and 3-O-galloyl-epigallocatechin-epigallocatechin-3-O-gallate, which constituted 1.75% of the ethanol extract. Tyrosol methyl ether, mongrhoside, and the two proanthocyanidin dimers were reported for the first time from this species in this study. Intraperitoneal injection of the 80% ethanol extract increased survival time of mice under hypoxia by 192%, as an indication of adaptogenic activity.

Plant extracts from central Asia showing antiinflammatory activities in gene expression assays.

Plant natural products remain a good resource for the discovery of novel pharmaceuticals. A mouse macrophage-based quantitative, reverse transcription polymerase chain reaction (qRT-PCR) system was optimized to screen plant extracts for antiinflammatory activities using three well known genetic markers of inflammation. Plants used for extraction were taxonomically identified and vouchered species from two Central Asian countries, Uzbekistan and Kyrgyzstan, collected through the International Cooperative Biodiversity Groups (ICBG) program. The mRNA expression of cyclooxygenase-2, interleukin 1beta and inducible nitric oxide synthase genes in RAW macrophages was determined quantitatively in response to treatment with plant extracts applied at 100 microg/mL. The screening of 1000 extracts from 449 plant species belonging to 68 plant families resulted in 75 extracts (7.5%) showing strong (75% or higher inhibition) activity against at least one target gene. Many extracts showed qualitative and quantitative differences in the levels of activities against each target gene. Extracts identified from this screen were able to reduce inflammatory symptoms in vivo, thereby validating the screening approach.

In vitro and in vivo anti-inflammatory activity of a seed preparation containing phenethylisothiocyanate.

Winter cress (Barbarea verna) seed preparations rich in phenethylisothiocyanate (PEITC) had strong in vivo and in vitro anti-inflammatory activity, significantly reducing the size of carrageenan-induced rat paw edema. This in vivo effect was comparable with that of the nonsteroidal anti-inflammatory drug aspirin. The seed preparation, in a concentration-dependent manner, reduced the mRNA levels of inflammation-related genes such as the inducible forms of cyclooxygenase and nitric-oxide synthase and the proinflammatory cytokine interleukin in lipopolysaccharide-stimulated mouse macrophage cell line RAW 264.7. Activity of the seed preparation was similar to that of the synthetic PEITC. PEITC was the most active of five different forms of isothiocyanate tested for their effects on in vitro proinflammatory gene expression. In vitro activity of the seed preparation was also compared with that of two known anti-inflammatory drugs. We conclude that Barbarea verna seed preparation may function as a potent anti-inflammatory agent, interfering with the transcription of proinflammatory genes.