Robust Ensemble of Two Different Multimodal Approaches to Segment 3D Ischemic Stroke Segmentation Using Brain Tumor Representation Among Multiple Center Datasets.

Ischemic stroke segmentation at an acute stage is vital in assessing the severity of patients' impairment and guiding therapeutic decision-making for reperfusion. Although many deep learning studies have shown attractive performance in medical segmentation, it is difficult to use these models trained on public data with private hospitals' datasets. Here, we demonstrate an ensemble model that employs two different multimodal approaches for generalization, a more effective way to perform on external datasets. First, after we jointly train a segmentation model on diffusion-weighted imaging (DWI) and apparent diffusion coefficient (ADC) MR modalities, the model is inferred on the DWI images. Second, a channel-wise segmentation model is trained by concatenating the DWI and ADC images as input, and then is inferred using both MR modalities. Before training with ischemic stroke data, we utilized BraTS 2021, a public brain tumor dataset, for transfer learning. An extensive ablation study evaluates which strategy learns better representations for ischemic stroke segmentation. In our study, nnU-Net well-known for robustness is selected as our baseline model. Our proposed method is evaluated on three different datasets: the Asan Medical Center (AMC) I and II, and the 2022 Ischemic Stroke Lesion Segmentation (ISLES). Our experiments are widely validated over a large, multi-center, and multi-scanner dataset with a huge amount of 846 scans. Not only stroke lesion models can benefit from transfer learning using brain tumor data, but combining the MR modalities using different training schemes also highly improves segmentation performance. The method achieved a top-1 ranking in the ongoing ISLES'22 challenge and performed particularly well on lesion-wise metrics of interest to neuroradiologists, achieving a Dice coefficient of 78.69% and a lesion-wise F1 score of 82.46%. Also, the method was relatively robust on the AMC I (Dice, 60.35%; lesion-wise F1, 68.30%) and II (Dice; 74.12%; lesion-wise F1, 67.53%) datasets in different settings. The high segmentation accuracy of our proposed method could improve radiologists' ability to detect ischemic stroke lesions in MRI images. Our model weights and inference code are available on https://github.com/MDOpx/ISLES22-model-inference .

Heterologous complementation systems verify the mosaic distribution of three distinct protoporphyrinogen IX oxidase in the cyanobacterial phylum.

The pathways for synthesizing tetrapyrroles, including heme and chlorophyll, are well-conserved among organisms, despite the divergence of several enzymes in these pathways. Protoporphyrinogen IX oxidase (PPOX), which catalyzes the last common step of the heme and chlorophyll biosynthesis pathways, is encoded by three phylogenetically-unrelated genes, hemY, hemG and hemJ. All three types of homologues are present in the cyanobacterial phylum, showing a mosaic phylogenetic distribution. Moreover, a few cyanobacteria appear to contain two types of PPOX homologues. Among the three types of cyanobacterial PPOX homologues, only a hemJ homologue has been experimentally verified for its functionality. An objective of this study is to provide experimental evidence for the functionality of the cyanobacterial PPOX homologues by using two heterologous complementation systems. First, we introduced hemY and hemJ homologues from Gloeobacter violaceus PCC7421, hemY homologue from Trichodesmium erythraeum, and hemG homologue from Prochlorococcus marinus MIT9515 into a ΔhemG strain of E. coli. hemY homologues from G. violaceus and T. erythraeum, and the hemG homologue of P. marinus complimented the E. coli strain. Subsequently, we attempted to replace the endogenous hemJ gene of the cyanobacterium Synechocystis sp. PCC6803 with the four PPOX homologues mentioned above. Except for hemG from P. marinus, the other PPOX homologues substituted the function of hemJ in Synechocystis. These results show that all four homologues encode functional PPOX. The transformation of Synechocystis with G. violaceus hemY homologue rendered the cells sensitive to an inhibitor of the HemY-type PPOX, acifluorfen, indicating that the hemY homologue is sensitive to this inhibitor, while the wild-type G. violaceus was tolerant to it, most likely due to the presence of HemJ protein. These results provide an additional level of evidence that G. violaceus contains two types of functional PPOX.

In Vitro Enzymatic Activity Assays Implicate the Existence of the Chlorophyll Cycle in Chlorophyll b-Containing Cyanobacteria.

In plants, chlorophyll (Chl) a and b are interconvertible by the action of three enzymes-chlorophyllide a oxygenase, Chl b reductase (CBR) and 7-hydroxymethyl chlorophyll a reductase (HCAR). These reactions are collectively referred to as the Chl cycle. In plants, this cyclic pathway ubiquitously exists and plays essential roles in acclimation to different light conditions at various developmental stages. By contrast, only a limited number of cyanobacteria species produce Chl b, and these include Prochlorococcus, Prochloron, Prochlorothrix and Acaryochloris. In this study, we investigated a possible existence of the Chl cycle in Chl b synthesizing cyanobacteria by testing in vitro enzymatic activities of CBR and HCAR homologs from Prochlorothrix hollandica and Acaryochloris RCC1774. All of these proteins show respective CBR and HCAR activity in vitro, indicating that both cyanobacteria possess the potential to complete the Chl cycle. It is also found that CBR and HCAR orthologs are distributed only in the Chl b-containing cyanobacteria that habitat shallow seas or freshwater, where light conditions change dynamically, whereas they are not found in Prochlorococcus species that usually habitat environments with fixed lighting. Taken together, our results implicate a possibility that the Chl cycle functions for light acclimation in Chl b-containing cyanobacteria.

Role of Helicobacter pylori infection among offspring or siblings of gastric cancer patients.

A positive family history is an increased risk factor for gastric cancer within family members, and one of the possible causes of this is the intrafamilial clustering of Helicobacter pylori infection. Our study examined the prevalence of H. pylori infection, serum antibodies to CagA and VacA and atrophic gastritis and/or intestinal metaplasia in the offspring or siblings of gastric cancer patients. A total of 726 subjects included 300 relatives of 300 separate gastric cancer patients and 426 controls. All subjects underwent upper gastrointestinal endoscopic examination with a rapid urease test. Blood samples were obtained to test for the presence of serum antibodies to the CagA and VacA proteins of H. pylori. The prevalence of H. pylori infection was higher in relatives of cancer patients (75.3%) than in controls (60.1%), and the adjusted odds ratio was 2.1 (95% CI 1.5-2.9). When either siblings or 2 or more family members were gastric cancer patients, the prevalence of H. pylori infection was much higher compared to the prevalence in controls. There was no specific relationship between CagA and VacA, and H. pylori infection. Atrophic gastritis and/or intestinal metaplasia were more frequently found in H. pylori-infected relatives of cancer patients (26.1%) than in H. pylori-infected controls (12.9%). These results strongly support a role for H. pylori infection in familial aggregation of gastric cancer. The prophylactic eradication of H. pylori infection in the offspring or siblings of gastric cancer patients may be clinically beneficial.