Slimmable transformer with hybrid axial-attention for medical image segmentation.

The transformer architecture has achieved remarkable success in medical image analysis owing to its powerful capability for capturing long-range dependencies. However, due to the lack of intrinsic inductive bias in modeling visual structural information, the transformer generally requires a large-scale pre-training schedule, limiting the clinical applications over expensive small-scale medical data. To this end, we propose a slimmable transformer to explore intrinsic inductive bias via position information for medical image segmentation. Specifically, we empirically investigate how different position encoding strategies affect the prediction quality of the region of interest (ROI) and observe that ROIs are sensitive to different position encoding strategies. Motivated by this, we present a novel Hybrid Axial-Attention (HAA) that can be equipped with pixel-level spatial structure and relative position information as inductive bias. Moreover, we introduce a gating mechanism to achieve efficient feature selection and further improve the representation quality over small-scale datasets. Experiments on LGG and COVID-19 datasets prove the superiority of our method over the baseline and previous works. Internal workflow visualization with interpretability is conducted to validate our success better; the proposed slimmable transformer has the potential to be further developed into a visual software tool for improving computer-aided lesion diagnosis and treatment planning.

Identification of the HLA-A*11:463 allele in a Chinese patient and his father.

HLA-A*11:463 has one nucleotide change from HLA-A*11:01:01:01 at nucleotide 508 changing Lysine (146) to Glutamine.

Pharmacokinetics, Pharmacodynamics, and Safety of Single Dose HSK7653 Tablets in Chinese Subjects with Normal or Impaired Renal Function.

OBJECTIVE: HSK7653 is a novel, ultralong-acting dipeptidyl peptidase-4 (DPP-4) inhibitor, promising for type 2 diabetes mellitus with a dosing regimen of once every 2 weeks. This trial investigates the pharmacokinetics (PKs), pharmacodynamics (PDs),and safety of HSK7653 in outpatients with normal or impaired renal function. METHODS: This is a multicenter, open-label, nonrandomized, parallel-controlled phase I clinical study that investigates the pharmacokinetic profiles of HSK7653 after a single oral administration in 42 subjects with mild (n = 8), moderate (n = 10), severe renal impairment (n = 10), and end-stage renal disease (without dialysis, n = 5) compared with matched control subjects with normal renal function (n = 9). Safety was evaluated throughout the study, and the pharmacodynamic effects were assessed on the basis of a DPP-4 inhibition rate. RESULTS: HSK7653 exposure levels including the maximum plasma concentration (Cmax), area under the plasma concentration-time curve from zero to last time of quantifiable concentration (AUC0-t), and area under the plasma concentration-time curve from zero to infinity (AUC0-inf) showed no significant differences related to the severity of renal impairment. Renal clearance (CLR) showed a certain downtrend along with the severity of renal impairment. The CLR of the group with severe renal impairment and the group with end-stage renal disease were basically similar. The DPP-4 inhibition rate-time curve graph was similar among the renal function groups. All groups had favorable safety, and no serious adverse events occurred. CONCLUSIONS: HSK7653 is a potent oral DPP-4 inhibitor with a long plasma half-life, supporting a dosing regimen of once every 2 weeks. Impaired renal function does not appear to impact the pharmacokinetic and pharmacodynamic properties of HSK7653 after a single administration in Chinese subjects. HSK7653 is also well tolerated without an increase in adverse events with increasing renal impairment. These results indicate that dose adjustment of HSK7653 may not be required in patients with renal impairment. TRIAL REGISTRATION: ClinicalTrials.gov Identifier: NCT05497297.

Electrochemical nitrogen reduction reaction on anchored SnS2 nanosheets with TM2 dimers.

The design of efficient, high-stability nitrogen fixation catalysts remains a great challenge to achieve electrochemical nitrogen reduction reaction (NRR) under ambient conditions. Herein, the high-throughput first-principles calculations are performed to obtain potential electrochemical NRR catalysts from transition metal (TM) dimers anchored on SnS2 nanosheets. The selected W2/SnS2 behaves as a promising NRR candidate possessing -0.27 V limiting potential and 0.81 eV maximum kinetic potential, and it exhibits the adsorption advantages of *N2 over other small molecules (*H2O, *O, *OH, *H). More importantly, the moderate d orbital valence electron number and electronegativity of TM atom could obtain better NRR activity, and a new descriptor φ considering the effects of coordination environments and adsorbates is proposed to achieve the fast pre-screening among various candidates. This work presents practical insights into the fast screening of TM2/SnS2 candidates for efficient nitrogen fixation and further streamlining the design of electrochemical NRR catalysts.

Automated polyp segmentation based on a multi-distance feature dissimilarity-guided fully convolutional network.

Colorectal malignancies often arise from adenomatous polyps, which typically begin as solitary, asymptomatic growths before progressing to malignancy. Colonoscopy is widely recognized as a highly efficacious clinical polyp detection method, offering valuable visual data that facilitates precise identification and subsequent removal of these tumors. Nevertheless, accurately segmenting individual polyps poses a considerable difficulty because polyps exhibit intricate and changeable characteristics, including shape, size, color, quantity and growth context during different stages. The presence of similar contextual structures around polyps significantly hampers the performance of commonly used convolutional neural network (CNN)-based automatic detection models to accurately capture valid polyp features, and these large receptive field CNN models often overlook the details of small polyps, which leads to the occurrence of false detections and missed detections. To tackle these challenges, we introduce a novel approach for automatic polyp segmentation, known as the multi-distance feature dissimilarity-guided fully convolutional network. This approach comprises three essential components, i.e., an encoder-decoder, a multi-distance difference (MDD) module and a hybrid loss (HL) module. Specifically, the MDD module primarily employs a multi-layer feature subtraction (MLFS) strategy to propagate features from the encoder to the decoder, which focuses on extracting information differences between neighboring layers' features at short distances, and both short and long-distance feature differences across layers. Drawing inspiration from pyramids, the MDD module effectively acquires discriminative features from neighboring layers or across layers in a continuous manner, which helps to strengthen feature complementary across different layers. The HL module is responsible for supervising the feature maps extracted at each layer of the network to improve prediction accuracy. Our experimental results on four challenge datasets demonstrate that the proposed approach exhibits superior automatic polyp performance in terms of the six evaluation criteria compared to five current state-of-the-art approaches.

Exploring a frequency-domain attention-guided cascade U-Net: Towards spatially tunable segmentation of vasculature.

Developing fully automatic and highly accurate medical image segmentation methods is critically important for vascular disease diagnosis and treatment planning. Although advances in convolutional neural networks (CNNs) have spawned an array of automatic segmentation models converging to saturated high performance, none have explored whether CNNs can achieve (spatially) tunable segmentation. As a result, we propose multiple attention modules from a frequency-domain perspective to construct a unified CNN architecture for segmenting vasculature with desired (spatial) scales. The proposed CNN architecture is named frequency-domain attention-guided cascaded U-Net (FACU-Net). Specifically, FACU-Net contains two innovative components: (1) a frequency-domain-based channel attention module that adaptively tunes channel-wise feature responses and (2) a frequency-domain-based spatial attention module that enables the deep network to concentrate on foreground regions of interest (ROIs) effectively. Furthermore, we devised a novel frequency-domain-based content attention module to enhance or weaken the high (spatial) frequency information, allowing us to strengthen or eliminate vessels of interest. Extensive experiments using clinical data from patients with intracranial aneurysms (IA) and abdominal aortic aneurysms (AAA) demonstrated that the proposed FACU-Net met its design goal. In addition, we further investigated the association between varying (spatial) frequency components and the desirable vessel size/scale attributes. In summary, our preliminary findings are encouraging, and further developments may lead to deployable image segmentation models that are spatially tunable for clinical applications.

S-Net: a multiple cross aggregation convolutional architecture for automatic segmentation of small/thin structures for cardiovascular applications.

With the success of U-Net or its variants in automatic medical image segmentation, building a fully convolutional network (FCN) based on an encoder-decoder structure has become an effective end-to-end learning approach. However, the intrinsic property of FCNs is that as the encoder deepens, higher-level features are learned, and the receptive field size of the network increases, which results in unsatisfactory performance for detecting low-level small/thin structures such as atrial walls and small arteries. To address this issue, we propose to keep the different encoding layer features at their original sizes to constrain the receptive field from increasing as the network goes deeper. Accordingly, we develop a novel S-shaped multiple cross-aggregation segmentation architecture named S-Net, which has two branches in the encoding stage, i.e., a resampling branch to capture low-level fine-grained details and thin/small structures and a downsampling branch to learn high-level discriminative knowledge. In particular, these two branches learn complementary features by residual cross-aggregation; the fusion of the complementary features from different decoding layers can be effectively accomplished through lateral connections. Meanwhile, we perform supervised prediction at all decoding layers to incorporate coarse-level features with high semantic meaning and fine-level features with high localization capability to detect multi-scale structures, especially for small/thin volumes fully. To validate the effectiveness of our S-Net, we conducted extensive experiments on the segmentation of cardiac wall and intracranial aneurysm (IA) vasculature, and quantitative and qualitative evaluations demonstrated the superior performance of our method for predicting small/thin structures in medical images.

Shikonin exerts an anti-leukemia effect against FLT3-ITD mutated acute myeloid leukemia cells via targeting FLT3 tyrosine kinase and its downstream pathways.

INTRODUCTION: Acute myeloid leukemia (AML) with internal tandem duplication (ITD) mutations in Fms-like tyrosine kinase 3 (FLT3) has an unfavorable prognosis. Recently, using newly emerging inhibitors of FLT3 has led to improved outcomes of patients with FLT3-ITD mutations. However, drug resistance and relapse continue to be significant challenges in the treatment of patients with FLT3-ITD mutations. This study aimed to evaluate the anti-leukemic effects of shikonin (SHK) and its mechanisms of action against AML cells with FLT3-ITD mutations in vitro and in vivo. METHODS: The CCK-8 assay was used to analyze cell viability, and flow cytometry was used to detect cell apoptosis and differentiation. Western blotting and real-time polymerase chain reaction (RT-PCR) were used to examine the expression of certain proteins and genes. Leukemia mouse model was created to evaluate the anti-leukemia effect of SHK against FLT3-ITD mutated leukemia in vivo. RESULTS: After screening a series of leukemia cell lines, those with FLT3-ITD mutations were found to be more sensitive to SHK in terms of proliferation inhibition and apoptosis induction than those without FLT3-ITD mutations. SHK suppresses the expression and phosphorylation of FLT3 receptors and their downstream molecules. Inhibition of the NF-κB/miR-155 pathway is an important mechanism through which SHK kills FLT3-AML cells. Moreover, a low concentration of SHK promotes the differentiation of AML cells with FLT3-ITD mutations. Finally, SHK could significantly inhibit the growth of MV4-11 cells in leukemia bearing mice. CONCLUSION: The findings of this study indicate that SHK is a promising drug for the treatment of FLT3-ITD mutated AML.

Immune Cell Infiltration and Clinical Significance of Angiogenesis-related Genes in Lung Adenocarcinoma.

BACKGROUND/AIM: Angiogenesis is one of the hallmarks of cancer. However, the role of molecular subtypes of angiogenesis-associated genes (AAGs) in the tumor immune microenvironment (TIME) of lung adenocarcinoma (LUAD) remains unclear. MATERIALS AND METHODS: The expression of AAGs in patients with LUAD were studied. Consensus clustering was performed to identify new AAG-associated molecular subgroups. The TIME and immune status of the subgroups were analyzed. Functional enrichment analysis was performed on the differentially expression genes among the clustered subgroups to analyze their relationship with AAGs. Furthermore, a prognostic risk model and clinical nomogram associated with survival time were constructed. Risk scores of drug sensitivity, immune checkpoint molecules, tumor mutational burden, and tumor cell stemness were analyzed. Finally, a series of in vitro experiments were performed to investigate the role of dickkopf WNT signaling pathway inhibitor 1 (DKK1) in LUAD. RESULTS: Two molecular subgroups with significantly different survival rates and TIME were identified. Immune checkpoint scores were higher in the subgroup with a worse prognosis. Moreover, differentially expressed genes were enriched in cell-cycle regulation, protein metabolism, and the immune microenvironment. The risk model and clinical nomogram constructed based on AAGs accurately predicted the prognosis of patients with LUAD. Patients with high-risk scores were less sensitive to chemotherapy but more sensitive to immunotherapy. DKK1 was highly expressed in basal cells and luminal cells. In addition, the knockdown of DKK1 reduced LUAD cell proliferation, invasion, and migration. CONCLUSION: Models based on AAGs can play an important role in predicting LUAD prognosis and immunotherapy effects. We further characterized the angiogenesis of TIME and studied the AAG DKK1. Our findings provide a theoretical basis for antitumor strategies targeting angiogenesis.

Application of case-based learning in psychology teaching: a meta-analysis.

BACKGROUND: Case-based learning (CBL) has been found to be effective for many subjects, but there is currently a lack of evidence regarding its utility in psychology education. The present study investigated whether CBL pedagogy can improve students' academic performance in psychology courses compared to the traditional teaching methods. METHODS: A systematic review and meta-analysis were conducted to investigate the effectiveness of CBL in psychology teaching. Databases including PubMed, Embase, Web of Science, China National Knowledge Infrastructure (CNKI), the VIP database, and Wanfang data were searched to find eligible randomized controlled trials. Pooled effect estimates were calculated using Hedges' g under the random effects model, and a subgroup analysis was carried to investigate the heterogeneity among studies. RESULTS: Fifteen studies with 2172 participants, 1086 in the CBL group and 1086 in the traditional lecture-based teaching group, were included in the meta-analysis. Students in the CBL group scored significantly higher on exams than those in the lecture-based group [Hedges' g = 0.68, 95%CI (0.49, 0.88), p < 0.00]. Relatively high heterogeneity was noted among the included studies. Publication bias was examined by the funnel plot and Egger's test, but did not significantly influence the stability of the results. A subsequent evaluation using the trim-and-fill method confirmed that no single study was skewing the overall results. A qualitative review of the included studies suggested that most students in the CBL group were satisfied with the CBL teaching mode. CONCLUSIONS: This meta-analysis indicated that the CBL pedagogy could be effective in psychology education, and might help increase students' academic scores, while encouraging a more engaging and cooperative learning environment. At present, the application of CBL in psychology education is in its initial stage. Problems related to the curriculum itself, research methodology, and challenges faced by both teachers and learners have confined its practice. Fully tapping into the strengths of CBL in psychology teaching will require additional work and advancing research.

Deep-learning-based image segmentation for image-based computational hemodynamic analysis of abdominal aortic aneurysms: a comparison study.

Computational hemodynamics is increasingly being used to quantify hemodynamic characteristics in and around abdominal aortic aneurysms (AAA) in a patient-specific fashion. However, the time-consuming manual annotation hinders the clinical translation of computational hemodynamic analysis. Thus, we investigate the feasibility of using deep-learning-based image segmentation methods to reduce the time required for manual segmentation. Two of the latest deep-learning-based image segmentation methods, ARU-Net and CACU-Net, were used to test the feasibility of automated computer model creation for computational hemodynamic analysis. Morphological features and hemodynamic metrics of 30 computed tomography angiography (CTA) scans were compared between pre-dictions and manual models. The DICE score for both networks was 0.916, and the correlation value was above 0.95, indicating their ability to generate models comparable to human segmentation. The Bland-Altman analysis shows a good agreement between deep learning and manual segmentation results. Compared with manual (computational hemodynamics) model recreation, the time for automated computer model generation was significantly reduced (from ∼2 h to ∼10 min). Automated image segmentation can significantly reduce time expenses on the recreation of patient-specific AAA models. Moreover, our study showed that both CACU-Net and ARU-Net could accomplish AAA segmentation, and CACU-Net outperformed ARU-Net in terms of accuracy and time-saving.

Characterization of small abdominal aortic aneurysms' growth status using spatial pattern analysis of aneurismal hemodynamics.

Aneurysm hemodynamics is known for its crucial role in the natural history of abdominal aortic aneurysms (AAA). However, there is a lack of well-developed quantitative assessments for disturbed aneurysmal flow. Therefore, we aimed to develop innovative metrics for quantifying disturbed aneurysm hemodynamics and evaluate their effectiveness in predicting the growth status of AAAs, specifically distinguishing between fast-growing and slowly-growing aneurysms. The growth status of aneurysms was classified as fast (≥ 5 mm/year) or slow (< 5 mm/year) based on serial imaging over time. We conducted computational fluid dynamics (CFD) simulations on 70 patients with computed tomography (CT) angiography findings. By converting hemodynamics data (wall shear stress and velocity) located on unstructured meshes into image-like data, we enabled spatial pattern analysis using Radiomics methods, referred to as "Hemodynamics-informatics" (i.e., using informatics techniques to analyze hemodynamic data). Our best model achieved an AUROC of 0.93 and an accuracy of 87.83%, correctly identifying 82.00% of fast-growing and 90.75% of slowly-growing AAAs. Compared with six classification methods, the models incorporating hemodynamics-informatics exhibited an average improvement of 8.40% in AUROC and 7.95% in total accuracy. These preliminary results indicate that hemodynamics-informatics correlates with AAAs' growth status and aids in assessing their progression.

Intercellular mitochondrial component transfer triggers ischemic cardiac fibrosis.

Myocardial fibrosis is the villain of sudden cardiac death. Myocardial ischemia/reperfusion (MI/R) injury induces cardiomyocyte damage or even death, which in turn stimulates fibroblast activation and fibrosis, but the intercellular communication mechanism remains unknown. Recent studies have shown that small extracellular vesicles (sEVs) significantly contribute to intercellular communication. Whether and how sEV might mediate post-MI/R cardiomyocyte/fibroblasts communication remain unknown. Here, in vivo and in vitro MI/R models were established. We demonstrate that sEVs derived from cardiomyocyte (Myo-sEVs) carry mitochondrial components, which enter fibroblasts to initiate myocardial fibrosis. Based on bioinformatics screening and experimental verification, the activating molecule in Beclin1-regulated autophagy protein 1 (autophagy/beclin-1 regulator 1, Ambra1) was found to be a critical component of these sEV and might be a new marker for Myo-sEVs. Interestingly, release of Ambra1+-Myo-sEVs was caused by secretory rather than canonical autophagy after MI/R injury and thereby escaped degradation. In ischemic and peripheral areas, Ambra1+-Myo-sEVs were internalized by fibroblasts, and the delivered mtDNA components to activate the fibroblast cyclic GMP-AMP synthase (cGAS)-stimulator of interferon genes (STING) pathway to promote fibroblast activation and proliferation. In addition, our data show that Ambra1 is expressed on the EV surface and cardiac-specific Ambra1 down regulation inhibits the Ambra1+-Myo-sEVs release and fibroblast uptake, effectively inhibiting ischemic myocardial fibrosis. This finding newly provides the evidence that myocardial secretory autophagy plays a role in intercellular communication during cardiac fibrosis. Ambra1 is a newly characterized molecule with bioactivity and might be a marker for Myo-sEVs, providing new therapeutic targets for cardiac remodeling.

Radiomic-based Textural Analysis of Intraluminal Thrombus in Aortic Abdominal Aneurysms: A Demonstration of Automated Workflow.

Our main objective is to investigate how the structural information of intraluminal thrombus (ILT) can be used to predict abdominal aortic aneurysms (AAA) growth status through an automated workflow. Fifty-four human subjects with ILT in their AAAs were identified from our database; those AAAs were categorized as slowly- (< 5 mm/year) or fast-growing (≥ 5 mm/year) AAAs. In-house deep-learning image segmentation models were used to generate 3D geometrical AAA models, followed by automated analysis. All features were fed into a support vector machine classifier to predict AAA's growth status.The most accurate prediction model was achieved through four geometrical parameters measuring the extent of ILT, two parameters quantifying the constitution of ILT, antihypertensive medication, and the presence of co-existing coronary artery disease. The predictive model achieved an AUROC of 0.89 and a total accuracy of 83%. When ILT was not considered, our prediction's AUROC decreased to 0.75 (P-value < 0.001).

Frataxin inhibits the sensitivity of the myocardium to ferroptosis by regulating iron homeostasis.

RATIONALE: Myocardial ischemia/reperfusion (I/R) injury is characterized by cell death via various cellular mechanisms upon reperfusion. As a new type of cell death, ferroptosis provides new opportunities to reduce myocardial cell death. Ferroptosis is known to be more active during reperfusion than ischemia. However, the mechanisms regulating ferroptosis during ischemia and reperfusion remain largely unknown. METHODS: The contribution of ferroptosis in ischemic and reperfused myocardium were detected by administered of Fer-1, a ferroptosis inhibitor to C57BL/6 mice, followed by left anterior descending (LAD) ligation surgery. Ferroptosis was evaluated by measurement of cell viability, ptgs2 mRNA level, iron production, malondialdehyde (MDA) and 4-hydroxynonenal (4-HNE) levels. H9C2 cells were exposed to hypoxia/reoxygenation to mimic in vivo I/R. We used LC-MS/MS to identify potential E3 ligases that interacted with frataxin in heart tissue. Cardiac-specific overexpression of frataxin in whole heart was achieved by intracardiac injection of frataxin, carried by adeno-associated virus serotype 9 (AAV9) containing cardiac troponin T (cTnT) promoter. RESULTS: We showed that regulators of iron metabolism, especially iron regulatory protein activity, were increased in the ischemic myocardium or hypoxia cardiomyocytes. In addition, we found that frataxin, which is involved in iron metabolism, is differentially expressed in the ischemic and reperfused myocardium and involved in the regulation of cardiomyocytes ferroptosis. Furthermore, we identified an E3 ligase, NHL repeat-containing 1 (NHLRC1), that mediates frataxin ubiquitination degradation. Cardiac-specific overexpression of frataxin ameliorated myocardial I/R injury through ferroptosis inhibition. CONCLUSIONS: Through a multi-level study from molecule to animal model, these findings uncover the key role of frataxin in inhibiting cardiomyocyte ferroptosis and provide new strategies and perspectives for the treatment of myocardial I/R injury.

Immunotherapy: A promising novel endometriosis therapy.

Endometriosis is a common disease of the female reproductive system and has malignant features. Although endometriosis by itself is a benign disease, its erosive growth characteristics lead to severe pelvic pain and female infertility. Unfortunately, several aspects of the pathogenesis of endometriosis are still unclear. Furthermore, the clinical therapeutic methods are unsatisfactory. The recurrence rate of endometriosis is high. Accumulating evidence suggests that the onset and development of endometriosis are closely related to the abnormal function of the female autoimmune system, especially the function of some immune cells such as the aggregation of neutrophils, abnormal differentiation of macrophages, decreased cytotoxicity of NK cells, and abnormal function of T- and B-cell lines. Therefore, immunotherapy is probably a novel therapeutic strategy for endometriosis besides surgery and hormone therapy. However, information regarding the clinical application of immunotherapy in the treatment of endometriosis is very limited. This article aimed to review the effects of existing immunomodulators on the development of endometriosis, including immune cell regulators and immune factor regulators. These immunomodulators clinically or experimentally inhibit the pathogenesis and development of endometriosis lesions by acting on the immune cells, immune factors, or immune-related signaling pathways. Thus, immunotherapy is probably a novel and effective clinical treatment choice for endometriosis. Experimental studies of the detailed mechanism of immunotherapy and large-scale clinical studies about the effectiveness and safety of this promising therapeutic method are required in the future.

Augmenting Prediction of Intracranial Aneurysms' Risk Status Using Velocity-Informatics: Initial Experience.

Our primary goal here is to demonstrate that innovative analytics of aneurismal velocities, named velocity-informatics, enhances intracranial aneurysm (IA) rupture status prediction. 3D computer models were generated using imaging data from 112 subjects harboring anterior IAs (4-25 mm; 44 ruptured and 68 unruptured). Computational fluid dynamics simulations and geometrical analyses were performed. Then, computed 3D velocity vector fields within the IA dome were processed for velocity-informatics. Four machine learning methods (support vector machine, random forest, generalized linear model, and GLM with Lasso or elastic net regularization) were employed to assess the merits of the proposed velocity-informatics. All 4 ML methods consistently showed that, with velocity-informatics metrics, the area under the curve and prediction accuracy both improved by approximately 0.03. Overall, with velocity-informatics, the support vector machine's prediction was most promising: an AUC of 0.86 and total accuracy of 77%, with 60% and 88% of ruptured and unruptured IAs being correctly identified, respectively.

Mitochondrial transplantation as a novel therapeutic strategy for cardiovascular diseases.

Cardiovascular disease (CVD) is the leading cause of noncommunicable disease-related death worldwide, and effective therapeutic strategies against CVD are urgently needed. Mitochondria dysfunction involves in the onset and development of CVD. Nowadays, mitochondrial transplantation, an alternative treatment aimed at increasing mitochondrial number and improving mitochondrial function, has been emerged with great therapeutic potential. Substantial evidence indicates that mitochondrial transplantation improves cardiac function and outcomes in patients with CVD. Therefore, mitochondrial transplantation has profound implications in the prevention and treatment of CVD. Here, we review the mitochondrial abnormalities that occur in CVD and summarize the therapeutic strategies of mitochondrial transplantation for CVD.

Automatic segmentation of abdominal aortic aneurysms from CT angiography using a context-aware cascaded U-Net.

We delineate abdominal aortic aneurysms, including lumen and intraluminal thrombosis (ILT), from contrast-enhanced computed tomography angiography (CTA) data in 70 patients with complete automation. A novel context-aware cascaded U-Net configuration enables automated image segmentation. Notably, auto-context structure, in conjunction with dilated convolutions, anisotropic context module, hierarchical supervision, and a multi-class loss function, are proposed to improve the delineation of ILT in an unbalanced, low-contrast multi-class labeling problem. A quantitative analysis shows that the automated image segmentation produces comparable results with trained human users (e.g., DICE scores of 0.945 and 0.804 for lumen and ILT, respectively). Resultant morphological metrics (e.g., volume, surface area, etc.) are highly correlated to those parameters generated by trained human users. In conclusion, the proposed automated multi-class image segmentation tool has the potential to be further developed as a translational software tool that can be used to improve the clinical management of AAAs.

An attention residual u-net with differential preprocessing and geometric postprocessing: Learning how to segment vasculature including intracranial aneurysms.

OBJECTIVE: Intracranial aneurysms (IA) are lethal, with high morbidity and mortality rates. Reliable, rapid, and accurate segmentation of IAs and their adjacent vasculature from medical imaging data is important to improve the clinical management of patients with IAs. However, due to the blurred boundaries and complex structure of IAs and overlapping with brain tissue or other cerebral arteries, image segmentation of IAs remains challenging. This study aimed to develop an attention residual U-Net (ARU-Net) architecture with differential preprocessing and geometric postprocessing for automatic segmentation of IAs and their adjacent arteries in conjunction with 3D rotational angiography (3DRA) images. METHODS: The proposed ARU-Net followed the classic U-Net framework with the following key enhancements. First, we preprocessed the 3DRA images based on boundary enhancement to capture more contour information and enhance the presence of small vessels. Second, we introduced the long skip connections of the attention gate at each layer of the fully convolutional decoder-encoder structure to emphasize the field of view (FOV) for IAs. Third, residual-based short skip connections were also embedded in each layer to implement in-depth supervision to help the network converge. Fourth, we devised a multiscale supervision strategy for independent prediction at different levels of the decoding path, integrating multiscale semantic information to facilitate the segmentation of small vessels. Fifth, the 3D conditional random field (3DCRF) and 3D connected component optimization (3DCCO) were exploited as postprocessing to optimize the segmentation results. RESULTS: Comprehensive experimental assessments validated the effectiveness of our ARU-Net. The proposed ARU-Net model achieved comparable or superior performance to the state-of-the-art methods through quantitative and qualitative evaluations. Notably, we found that ARU-Net improved the identification of arteries connecting to an IA, including small arteries that were hard to recognize by other methods. Consequently, IA geometries segmented by the proposed ARU-Net model yielded superior performance during subsequent computational hemodynamic studies (also known as "patient-specific" computational fluid dynamics [CFD] simulations). Furthermore, in an ablation study, the five key enhancements mentioned above were confirmed. CONCLUSIONS: The proposed ARU-Net model can automatically segment the IAs in 3DRA images with relatively high accuracy and potentially has significant value for clinical computational hemodynamic analysis.