MRI-based deep learning and radiomics for prediction of occult cervical lymph node metastasis and prognosis in early-stage oral and oropharyngeal squamous cell carcinoma: a diagnostic study.

INTRODUCTION: The incidence of occult cervical lymph node metastases (OCLNM) is reported to be 20%-30% in early-stage oral cancer and oropharyngeal cancer. There is a lack of an accurate diagnostic method to predict occult lymph node metastasis and to help surgeons make precise treatment decisions. AIM: To construct and evaluate a preoperative diagnostic method to predict occult lymph node metastasis (OCLNM) in early-stage oral and oropharyngeal squamous cell carcinoma (OC and OP SCC) based on deep learning features (DLFs) and radiomics features. METHODS: A total of 319 patients diagnosed with early-stage OC or OP SCC were retrospectively enrolled and divided into training, test and external validation sets. Traditional radiomics features and DLFs were extracted from their MRI images. The least absolute shrinkage and selection operator (LASSO) analysis was employed to identify the most valuable features. Prediction models for OCLNM were developed using radiomics features and DLFs. The effectiveness of the models and their clinical applicability were evaluated using the area under the curve (AUC), decision curve analysis (DCA) and survival analysis. RESULTS: Seventeen prediction models were constructed. The Resnet50 deep learning (DL) model based on the combination of radiomics and DL features achieves the optimal performance, with AUC values of 0.928 (95% CI: 0.881-0.975), 0.878 (95% CI: 0.766-0.990), 0.796 (95% CI: 0.666-0.927) and 0.834 (95% CI: 0.721-0.947) in the training, test, external validation set1 and external validation set2, respectively. Moreover, the Resnet50 model has great prediction value of prognosis in patients with early-stage OC and OP SCC. CONCLUSION: The proposed MRI-based Resnet50 deep learning model demonstrated high capability in diagnosis of OCLNM and prognosis prediction in the early-stage OC and OP SCC. The Resnet50 model could help refine the clinical diagnosis and treatment of the early-stage OC and OP SCC.

Liver transcriptome and physiological analyses preliminarily revealed the adaptation mechanisms of Amur grayling (Thymallus arcticus grubei, Dybowski, 1869) fry for dietary lipid nutrition.

The Amur grayling (Thymallus arcticus grubei Dybowski, 1869), a species of potentially economic and research value, is renowned for its tender meat, exquisite flavor, and high nutritional contents. This study was conducted to investigate the physiological adaptation mechanisms to dietary lipids in Amur grayling fry (with average initial weight 4.64±0.03 g). This study involved a 56-day feeding trial with diets containing varying lipid levels (9.07%, 12.17%, 15.26%, 18.09%, 21.16%, and 24.07%, designated as GL1 through GL6, respectively) to explore the impact of dietary lipids on growth performance, intestinal digestion, liver antioxidative function, and transcriptomic profiles. Results showed that The group receiving 18% dietary lipid exhibited a markedly higher weight gain rate (WGR) and specific growth rate compared to other groups, alongside a reduced feed conversion ratio (FCR), except in comparison to the 15% lipid group. Activities of lipase in pancreatic secretion and amylase in stomach mucosa peaked in the 18% lipid treatment group, indicating enhanced digestive efficiency. The liver of fish in this group also showed increased activities of antioxidative enzymes and higher levels of glutathione and total antioxidative capacity, along with reduced malondialdehyde content compared to the 9% and 24% lipid treatments. Additionally, serum high-density lipoprotein cholesterol levels were highest in the 18% group. Transcriptomic analysis revealed four significant metabolic pathways affected: Cholesterol metabolism, Fat digestion and absorption, PPAR signaling, and Fatty acid degradation, involving key genes such as Lipase, Lipoprotein lipase, Fatty acid-binding protein, and Carnitine palmitoyltransferase I. These findings suggest that the liver of Amur grayling employs adaptive mechanisms to manage excessive dietary lipids. Quadratic regression analysis determined the optimal dietary lipid levels to be 16.62% and 16.52%, based on WGR and FCR, respectively. The optimal dietary lipid level for juvenile Amur grayling appears to be around 18%, as evidenced by improved growth performance, digestive function, balanced serum lipid profile, and enhanced liver antioxidative capacity. Exceeding this lipid threshold triggers both adaptive and potentially detrimental liver responses.

Progress of Mitochondrial Function Regulation in Cardiac Regeneration.

Heart failure and myocardial infarction, global health concerns, stem from limited cardiac regeneration post-injury. Myocardial infarction, typically caused by coronary artery blockage, leads to cardiac muscle cell damage, progressing to heart failure. Addressing the adult heart's minimal self-repair capability is crucial, highlighting cardiac regeneration research's importance. Studies reveal a metabolic shift from anaerobic glycolysis to oxidative phosphorylation in neonates as a key factor in impaired cardiac regeneration, with mitochondria being central. The heart's high energy demands rely on a robust mitochondrial network, essential for cellular energy, cardiac health, and regenerative capacity. Mitochondria's influence extends to redox balance regulation, signaling molecule interactions, and apoptosis. Changes in mitochondrial morphology and quantity also impact cardiac cell regeneration. This article reviews mitochondria's multifaceted role in cardiac regeneration, particularly in myocardial infarction and heart failure models. Understanding mitochondrial function in cardiac regeneration aims to enhance myocardial infarction and heart failure treatment methods and insights.

Suitable Cottonseed Protein Concentrate Supplementation in Common Carp (Cyprinus carpio) Serves as an Effective Strategy for Fish Meal Sparing Based on Improvement in Intestinal Antioxidant Capacity, Barrier and Microbiota Composition.

The application of cottonseed protein concentrate (CPC) is an effective strategy to moderate the shortage of fish meal (FM) for the aquafeed industry. However, little attention has been paid to the effects of replacing fishmeal with CPC on cyprinid fish. This study used common carp (Cyprinus carpio) as the biological model and assessed the potential of applying CPC as a substitute for fishmeal in the diet of common carp. The proportion of fish meal substituted with CPC in the six diets was 0% (CPC0), 25% (CPC25), 50% (CPC50), 75% (CPC75), and 100% (CPC100). Each diet was fed to three replicate groups of common carp (4.17 ± 0.02 g) for 56 days. Results revealed that the CPC50 group significantly increased the growth indexes via up-regulating the genes of the GH/IGF axis and the TOR pathway. The intestinal digestive ability was also elevated in the CPC50 group via markedly increasing intestinal villus height, protease and lipase activities in the whole intestine, and the amylase activity of the foregut and midgut. The CPC50 group captured significantly higher activities and gene expressions of antioxidant enzymes and lower malonaldehyde contents via evoking the Nrf2/Keap1 signal pathway. The CPC50 group enhance the intestinal mechanical barrier via up-regulating the gene expressions of tight junction proteins and heighten the intestinal biological barrier by increasing the probiotics (Lactococcus) and decreasing the harmful bacteria (Enterococcus). But excessive substitution levels (75% and 100%) would compromise growth performance, intestinal antioxidant capacity, and immune function. The optimum substitution level was estimated to be 46.47%, 47.72%, and 46.43% using broken-line regression analyses based on mass gain rate, protein efficiency ratio, and feed conversion rate. Overall, the fishmeal in common carp feed could be substituted up to 50% by CPC without negative influence on growth, feed utilization, and or intestinal health.

Causal associations of genetically predicted gut microbiota and blood metabolites with inflammatory states and risk of infections: a Mendelian randomization analysis.

Background: Inflammation serves as a key pathologic mediator in the progression of infections and various diseases, involving significant alterations in the gut microbiome and metabolism. This study aims to probe into the potential causal relationships between gut microbial taxa and human blood metabolites with various serum inflammatory markers (CRP, SAA1, IL-6, TNF-α, WBC, and GlycA) and the risks of seven common infections (gastrointestinal infections, dysentery, pneumonia, bacterial pneumonia, bronchopneumonia and lung abscess, pneumococcal pneumonia, and urinary tract infections). Methods: Two-sample Mendelian randomization (MR) analysis was performed using inverse variance weighted (IVW), maximum likelihood, MR-Egger, weighted median, and MR-PRESSO. Results: After adding other MR models and sensitivity analyses, genus Roseburia was simultaneously associated adversely with CRP (Beta IVW = -0.040) and SAA1 (Beta IVW = -0.280), and family Bifidobacteriaceae was negatively associated with both CRP (Beta IVW = -0.034) and pneumonia risk (Beta IVW = -0.391). After correction by FDR, only glutaroyl carnitine remained significantly associated with elevated CRP levels (Beta IVW = 0.112). Additionally, threonine (Beta IVW = 0.200) and 1-heptadecanoylglycerophosphocholine (Beta IVW = -0.246) were found to be significantly associated with WBC levels. Three metabolites showed similar causal effects on different inflammatory markers or infectious phenotypes, stearidonate (18:4n3) was negatively related to SAA1 and urinary tract infections, and 5-oxoproline contributed to elevated IL-6 and SAA1 levels. In addition, 7-methylguanine showed a positive correlation with dysentery and bacterial pneumonia. Conclusion: This study provides novel evidence confirming the causal effects of the gut microbiome and the plasma metabolite profile on inflammation and the risk of infection. These potential molecular alterations may aid in the development of new targets for the intervention and management of disorders associated with inflammation and infections.

Cellular nucleic acid binding protein facilitates cardiac repair after myocardial infarction by activating ß-catenin signaling.

The regenerative capacity of the adult mammalian heart is limited, while the neonatal heart is an organ with regenerative and proliferative ability. Activating adult cardiomyocytes (CMs) to re-enter the cell cycle is an effective therapeutic method for ischemic heart disease such as myocardial infarction (MI) and heart failure. Here, we aimed to reveal the role and potential mechanisms of cellular nucleic acid binding protein (CNBP) in cardiac regeneration and repair after heart injury. CNBP is highly expressed within 7 days post-birth while decreases significantly with the loss of regenerative ability. In vitro, overexpression of CNBP promoted CM proliferation and survival, whereas knockdown of CNBP inhibited these processes. In vivo, knockdown of CNBP in CMs robustly hindered myocardial regeneration after apical resection in neonatal mice. In adult MI mice, CM-specific CNBP overexpression in the infarct border zone ameliorated myocardial injury in acute stage and facilitated CM proliferation and functional recovery in the long term. Quantitative proteomic analysis with TMT labeling showed that CNBP overexpression promoted the DNA replication, cell cycle progression, and cell division. Mechanically, CNBP overexpression increased the expression of ß-catenin and its downstream target genes CCND1 and c-myc; Furthermore, Luciferase reporter and Chromatin immunoprecipitation (ChIP) assays showed that CNBP could directly bind to the ß-catenin promoter and promote its transcription. CNBP also upregulated the expression of G1/S-related cell cycle genes CCNE1, CDK2, and CDK4. Collectively, our study reveals the positive role of CNBP in promoting cardiac repair after injury, providing a new therapeutic option for the treatment of MI.

Hybrid unsupervised representation learning and pseudo-label supervised self-distillation for rare disease imaging phenotype classification with dispersion-aware imbalance correction.

Rare diseases are characterized by low prevalence and are often chronically debilitating or life-threatening. Imaging phenotype classification of rare diseases is challenging due to the severe shortage of training examples. Few-shot learning (FSL) methods tackle this challenge by extracting generalizable prior knowledge from a large base dataset of common diseases and normal controls and transferring the knowledge to rare diseases. Yet, most existing methods require the base dataset to be labeled and do not make full use of the precious examples of rare diseases. In addition, the extremely small size of the training samples may result in inter-class performance imbalance due to insufficient sampling of the true distributions. To this end, we propose in this work a novel hybrid approach to rare disease imaging phenotype classification, featuring three key novelties targeted at the above drawbacks. First, we adopt the unsupervised representation learning (URL) based on self-supervising contrastive loss, whereby to eliminate the overhead in labeling the base dataset. Second, we integrate the URL with pseudo-label supervised classification for effective self-distillation of the knowledge about the rare diseases, composing a hybrid approach taking advantage of both unsupervised and (pseudo-) supervised learning on the base dataset. Third, we use the feature dispersion to assess the intra-class diversity of training samples, to alleviate the inter-class performance imbalance via dispersion-aware correction. Experimental results of imaging phenotype classification of both simulated (skin lesions and cervical smears) and real clinical rare diseases (retinal diseases) show that our hybrid approach substantially outperforms existing FSL methods (including those using a fully supervised base dataset) via effective integration of the URL, pseudo-label driven self-distillation, and dispersion-aware imbalance correction, thus establishing a new state of the art.

S-Nitrosylation of Septin2 Exacerbates Aortic Aneurysm and Dissection by Coupling the TIAM1-RAC1 Axis in Macrophages.

BACKGROUND: S-Nitrosylation (SNO), a prototypic redox-based posttranslational modification, is involved in cardiovascular disease. Aortic aneurysm and dissection are high-risk cardiovascular diseases without an effective cure. The aim of this study was to determine the role of SNO of Septin2 in macrophages in aortic aneurysm and dissection. METHODS: Biotin-switch assay combined with liquid chromatography-tandem mass spectrometry was performed to identify the S-nitrosylated proteins in aortic tissue from both patients undergoing surgery for aortic dissection and Apoe-/- mice infused with angiotensin II. Angiotensin II-induced aortic aneurysm model and ß-aminopropionitrile-induced aortic aneurysm and dissection model were used to determine the role of SNO of Septin2 (SNO-Septin2) in aortic aneurysm and dissection development. RNA-sequencing analysis was performed to recapitulate possible changes in the transcriptome profile of SNO-Septin2 in macrophages in aortic aneurysm and dissection. Liquid chromatography-tandem mass spectrometry and coimmunoprecipitation were used to uncover the TIAM1-RAC1 (Ras-related C3 botulinum toxin substrate 1) axis as the downstream target of SNO-Septin2. Both R-Ketorolac and NSC23766 treatments were used to inhibit the TIAM1-RAC1 axis. RESULTS: Septin2 was identified S-nitrosylated at cysteine 111 (Cys111) in both aortic tissue from patients undergoing surgery for aortic dissection and Apoe-/- mice infused with Angiotensin II. SNO-Septin2 was demonstrated driving the development of aortic aneurysm and dissection. By RNA-sequencing, SNO-Septin2 in macrophages was demonstrated to exacerbate vascular inflammation and extracellular matrix degradation in aortic aneurysm. Next, TIAM1 (T lymphoma invasion and metastasis-inducing protein 1) was identified as a SNO-Septin2 target protein. Mechanistically, compared with unmodified Septin2, SNO-Septin2 reduced its interaction with TIAM1 and activated the TIAM1-RAC1 axis and consequent nuclear factor-κB signaling pathway, resulting in stronger inflammation and extracellular matrix degradation mediated by macrophages. Consistently, both R-Ketorolac and NSC23766 treatments protected against aortic aneurysm and dissection by inhibiting the TIAM1-RAC1 axis. CONCLUSIONS: SNO-Septin2 drives aortic aneurysm and dissection through coupling the TIAM1-RAC1 axis in macrophages and activating the nuclear factor-κB signaling pathway-dependent inflammation and extracellular matrix degradation. Pharmacological blockade of RAC1 by R-Ketorolac or NSC23766 may therefore represent a potential treatment against aortic aneurysm and dissection.

Predicting survival in patients with buccal cancer: A study based on SEER database and external validation in China.

OBJECTIVE: Buccal mucosa cancer (BMC) is one of the most common oral cancers and has poor prognosis. The study aimed to develop and validate nomograms for predicting the 1-, 3-, and 5-year overall survival (OS) and cancer-specific survival (CSS) of BMC patients. METHODS: We collected and reviewed information on BMC patients diagnosed between 2004 and 2019 from the Surveillance Epidemiology and End Results database. Two nomograms were developed and validated to predict the OS and CSS based on predictors identified by univariate and multivariate Cox regression. An extra external validation was further performed using data from Sun Yat-sen Memorial Hospital (SYSMH). RESULTS: A total of 3154 BMC patients included in this study were randomly assigned to training and validation groups in a 2:1 ratio. Independent prognostic predictors were identified, confirmed, and fitted into nomograms for OS and CSS, respectively. The C-indices are 0.767 (Training group OS), 0.801 (Training group CSS), 0.763 (Validation group OS), and 0.781 (Validation group OS), respectively. Moreover, the nomograms exhibited remarkable precision in forecasting and significant clinical significance, as evidenced by receiver operating characteristic (ROC) curves, calibration curves, and decision curve analyses (DCA). The final validation using our data from SYSMH also showed high accuracy and substantial clinical benefits within the nomograms. The C-indices are 0.849 (SYSMH group OS) and 0.916 (SYSMH group CSS). These indexes are better than tumor, node, and metastasis stage based on prediction results. CONCLUSIONS: The nomograms developed with great performance predicted 1-, 3-, and 5-year OS and CSS of BMC patients. Use of the nomograms in clinical practices shall bring significant benefits to BMC patients.

Model-Heterogeneous Semi-Supervised Federated Learning for Medical Image Segmentation.

Medical image segmentation is crucial in clinical diagnosis, helping physicians identify and analyze medical conditions. However, this task is often accompanied by challenges like sensitive data, privacy concerns, and expensive annotations. Current research focuses on personalized collaborative training of medical segmentation systems, ignoring that obtaining segmentation annotations is time-consuming and laborious. Achieving a perfect balance between annotation cost and segmentation performance while ensuring local model personalization has become a valuable direction. Therefore, this study introduces a novel Model-Heterogeneous Semi-Supervised Federated (HSSF) Learning framework. It proposes Regularity Condensation and Regularity Fusion to transfer autonomously selective knowledge to ensure the personalization between sites. In addition, to efficiently utilize unlabeled data and reduce the annotation burden, it proposes a Self-Assessment (SA) module and a Reliable Pseudo-Label Generation (RPG) module. The SA module generates self-assessment confidence in real-time based on model performance, and the RPG module generates reliable pseudo-label based on SA confidence. We evaluate our model separately on the Skin Lesion and Polyp Lesion datasets. The results show that our model performs better than other methods characterized by heterogeneity. Moreover, it exhibits highly commendable performance even in homogeneous designs, most notably in region-based metrics. The full range of resources can be readily accessed through the designated repository located at HSSF(github.com) on the platform of GitHub.

Development and validation of a deep-learning model for the detection of non-displaced femoral neck fractures with anteroposterior and lateral hip radiographs.

Background: Hip fractures, including femoral neck fractures, are a significant cause of morbidity and mortality in the elderly population and are typically diagnosed using plain radiography. However, diagnosing non-displaced femoral neck fractures can be challenging due to their subtle appearance on hip radiographs. Previous deep-learning models have shown low accuracy in identifying these fractures on anteroposterior (AP) radiographs; however, no studies have used lateral radiographs. This study aimed to evaluate the potential of using deep-learning with both AP and lateral hip radiographs to automatically identify non-displaced femoral neck fractures. Methods: We conducted a retrospective analysis of patients with femoral neck fractures at The First Affiliated Hospital of Xiamen University. All the hip radiographs were reviewed, and cases of non-displaced femoral neck fractures were included in the study. Additionally, 439 participants with normal hip radiographs were also included in the study. A vision transformer (Vit) model was developed using 1,536 AP and lateral hip radiograph. The model's performance was compared to the performance of two groups of human observers: an expert group comprising orthopedic surgeons and radiologists, and a non-expert group, including emergency physicians and general practice doctors. We also carried out the external validation using two additional data sets to assess the generalizability of the model. Results: The Vit model showed exceptional performance in detecting non-displaced femoral neck fractures on paired AP and lateral hip radiographs, achieving a binary accuracy of 95.8% [95% confidence interval (CI): 94.9%, 96.8%] and an area under the curve (AUC) of 0.988. Compared to the human observers, the model had a higher accuracy of 96.7% (95% CI: 93.9%, 99.5%) on the paired AP and lateral hip radiographs, while the accuracy of the expert group was 90.5% (95% CI: 85.7%, 95.2%). Further, the model maintained good performance during the external validation, with an AUC of 0.959 on the paired AP and lateral views. Conclusions: Our Vit model showed expert-level performance in identifying non-displaced femoral neck fractures on paired AP and lateral hip radiographs. This model has the potential to enhance diagnosis accuracy and improve patient outcomes by reducing the need for additional examinations and preoperative time.

Integrated Single-cell and Bulk RNA Sequencing Analysis Cross Talk between Ferroptosis-related Genes and Prognosis in Oral Cavity Squamous Cell Carcinoma.

BACKGROUND: Ferroptosis is a new type of programmed apoptosis and plays an important role in tumour inhibition and immunotherapy. OBJECTIVE: In this study, we aimed to explore the potential role of ferroptosis-related genes (FRGs) and the potential therapeutic targets in oral cavity squamous cell carcinoma (OCSCC). METHODS: The transcription data of OCSCC samples were obtained from the Cancer Genome Atlas (TCGA) database as a training dataset. The prognostic FRGs were extracted by univariate Cox regression analysis. Then, we constructed a prognostic model using the least absolute shrinkage and selection operator (LASSO) and Cox analysis to determine the independent prognosis FRGs. Based on this model, risk scores were calculated for the OCSCC samples. The model's capability was further evaluated by the receiver operating characteristic curve (ROC). Then, we used the GSE41613 dataset as an external validation cohort to confirm the model's predictive capability. Next, the immune infiltration and somatic mutation analysis were applied. Lastly, single-cell transcriptomic analysis was used to identify the key cells. RESULTS: A total of 12 prognostic FRGs were identified. Eventually, 6 FRGs were screened as independent predictors and a prognostic model was constructed in the training dataset, which significantly stratified OCSCC samples into high-risk and low-risk groups based on overall survival. The external validation of the model using the GSE41613 dataset demonstrated a satisfactory predictive capability for the prognosis of OCSCC. Further analysis revealed that patients in the highrisk group had distinct immune infiltration and somatic mutation patterns from low-risk patients. Mast cell infiltrations were identified as prognostic immune cells and played a role in OCSCC partly through ferroptosis. CONCLUSION: We successfully constructed a novel 6 FRGs model and identified a prognostic immune cell, which can serve to predict clinical prognoses for OCSCC. Ferroptosis may be a new direction for immunotherapy of OCSCC.

FedDP: Dual Personalization in Federated Medical Image Segmentation.

Personalized federated learning (PFL) addresses the data heterogeneity challenge faced by general federated learning (GFL). Rather than learning a single global model, with PFL a collection of models are adapted to the unique feature distribution of each site. However, current PFL methods rarely consider self-attention networks which can handle data heterogeneity by long-range dependency modeling and they do not utilize prediction inconsistencies in local models as an indicator of site uniqueness. In this paper, we propose FedDP, a novel fed erated learning scheme with d ual p ersonalization, which improves model personalization from both feature and prediction aspects to boost image segmentation results. We leverage long-range dependencies by designing a local query (LQ) that decouples the query embedding layer out of each local model, whose parameters are trained privately to better adapt to the respective feature distribution of the site. We then propose inconsistency-guided calibration (IGC), which exploits the inter-site prediction inconsistencies to accommodate the model learning concentration. By encouraging a model to penalize pixels with larger inconsistencies, we better tailor prediction-level patterns to each local site. Experimentally, we compare FedDP with the state-of-the-art PFL methods on two popular medical image segmentation tasks with different modalities, where our results consistently outperform others on both tasks. Our code and models are available at https://github.com/jcwang123/PFL-Seg-Trans.

RECIST-Induced Reliable Learning: Geometry-Driven Label Propagation for Universal Lesion Segmentation.

Automatic universal lesion segmentation (ULS) from Computed Tomography (CT) images can ease the burden of radiologists and provide a more accurate assessment than the current Response Evaluation Criteria In Solid Tumors (RECIST) guideline measurement. However, this task is underdeveloped due to the absence of large-scale pixel-wise labeled data. This paper presents a weakly-supervised learning framework to utilize the large-scale existing lesion databases in hospital Picture Archiving and Communication Systems (PACS) for ULS. Unlike previous methods to construct pseudo surrogate masks for fully supervised training through shallow interactive segmentation techniques, we propose to unearth the implicit information from RECIST annotations and thus design a unified RECIST-induced reliable learning (RiRL) framework. Particularly, we introduce a novel label generation procedure and an on-the-fly soft label propagation strategy to avoid noisy training and poor generalization problems. The former, named RECIST-induced geometric labeling, uses clinical characteristics of RECIST to preliminarily and reliably propagate the label. With the labeling process, a trimap divides the lesion slices into three regions, including certain foreground, background, and unclear regions, which consequently enables a strong and reliable supervision signal on a wide region. A topological knowledge-driven graph is built to conduct the on-the-fly label propagation for the optimal segmentation boundary to further optimize the segmentation boundary. Experimental results on a public benchmark dataset demonstrate that the proposed method surpasses the SOTA RECIST-based ULS methods by a large margin. Our approach surpasses SOTA approaches over 2.0%, 1.5%, 1.4%, and 1.6% Dice with ResNet101, ResNet50, HRNet, and ResNest50 backbones.

Simultaneous alignment and surface regression using hybrid 2D-3D networks for 3D coherent layer segmentation of retinal OCT images with full and sparse annotations.

Layer segmentation is important to quantitative analysis of retinal optical coherence tomography (OCT). Recently, deep learning based methods have been developed to automate this task and yield remarkable performance. However, due to the large spatial gap and potential mismatch between the B-scans of an OCT volume, all of them were based on 2D segmentation of individual B-scans, which may lose the continuity and diagnostic information of the retinal layers in 3D space. Besides, most of these methods required dense annotation of the OCT volumes, which is labor-intensive and expertise-demanding. This work presents a novel framework based on hybrid 2D-3D convolutional neural networks (CNNs) to obtain continuous 3D retinal layer surfaces from OCT volumes, which works well with both full and sparse annotations. The 2D features of individual B-scans are extracted by an encoder consisting of 2D convolutions. These 2D features are then used to produce the alignment displacement vectors and layer segmentation by two 3D decoders coupled via a spatial transformer module. Two losses are proposed to utilize the retinal layers' natural property of being smooth for B-scan alignment and layer segmentation, respectively, and are the key to the semi-supervised learning with sparse annotation. The entire framework is trained end-to-end. To the best of our knowledge, this is the first work that attempts 3D retinal layer segmentation in volumetric OCT images based on CNNs. Experiments on a synthetic dataset and three public clinical datasets show that our framework can effectively align the B-scans for potential motion correction, and achieves superior performance to state-of-the-art 2D deep learning methods in terms of both layer segmentation accuracy and cross-B-scan 3D continuity in both fully and semi-supervised settings, thus offering more clinical values than previous works.

Severe thrombocytopenia induced by tirofiban after percutaneous coronary intervention: a case report.

BACKGROUND: Tirofiban is a nonpeptide glycoprotein IIb/IIIa receptor antagonist used widely in patients subjected to percutaneous coronary intervention. While the usage of tirofiban sets an important clinical benefit, severe thrombocytopenia can occur with use of this agent. CASE PRESENTATION: A 76-year-old Chinese man was admitted with 1-month history of sudden onset of chest tightness. He was diagnosed as having subacute inferior myocardial infarction, and percutaneous coronary intervention was performed. After the procedure, patient received tirofiban at 0.15 µg/kg/minute for 4 h. A blood sample was obtained for a complete blood count; severe thrombocytopenia was reported according to routine orders at our hospital. All antiplatelet drugs including tirofiban, aspirin, and clopidogrel were immediately discontinued. The patient received platelet transfusions and was treated with immunoglobulin G. Two days later, the patient's platelet count had increased to 75 × 109/L. There was a significant improvement after day 5, and the platelet count was 112 × 109/L. Seven days after the acute thrombocytopenia, he was discharged with normal platelet count. CONCLUSIONS: Clinicians should be particularly aware of tirofiban-induced thrombocytopenia in routine practice.

SUnet: A multi-organ segmentation network based on multiple attention.

Organ segmentation in abdominal or thoracic computed tomography (CT) images plays a crucial role in medical diagnosis as it enables doctors to locate and evaluate organ abnormalities quickly, thereby guiding surgical planning, and aiding treatment decision-making. This paper proposes a novel and efficient medical image segmentation method called SUnet for multi-organ segmentation in the abdomen and thorax. SUnet is a fully attention-based neural network. Firstly, an efficient spatial reduction attention (ESRA) module is introduced not only to extract image features better, but also to reduce overall model parameters, and to alleviate overfitting. Secondly, SUnet's multiple attention-based feature fusion module enables effective cross-scale feature integration. Additionally, an enhanced attention gate (EAG) module is considered by using grouped convolution and residual connections, providing richer semantic features. We evaluate the performance of the proposed model on synapse multiple organ segmentation dataset and automated cardiac diagnostic challenge dataset. SUnet achieves an average Dice of 84.29% and 92.25% on these two datasets, respectively, outperforming other models of similar complexity and size, and achieving state-of-the-art results.

SIRT3-dependent mitochondrial redox homeostasis mitigates CHK1 inhibition combined with gemcitabine treatment induced cardiotoxicity in hiPSC-CMs and mice.

Administration of CHK1-targeted anticancer therapies is associated with an increased cumulative risk of cardiac complications, which is further amplified when combined with gemcitabine. However, the underlying mechanisms remain elusive. In this study, we generated hiPSC-CMs and murine models to elucidate the mechanisms underlying CHK1 inhibition combined with gemcitabine-induced cardiotoxicity and identify potential targets for cardioprotection. Mice were intraperitoneally injected with 25 mg/kg CHK1 inhibitor AZD7762 and 20 mg/kg gemcitabine for 3 weeks. hiPSC-CMs and NMCMs were incubated with 0.5 uM AZD7762 and 0.1 uM gemcitabine for 24 h. Both pharmacological inhibition or genetic deletion of CHK1 and administration of gemcitabine induced mtROS overproduction and pyroptosis in cardiomyocytes by disrupting mitochondrial respiration, ultimately causing heart atrophy and cardiac dysfunction in mice. These toxic effects were further exacerbated with combination administration. Using mitochondria-targeting sequence-directed vectors to overexpress CHK1 in cardiomyocyte (CM) mitochondria, we identified the localization of CHK1 in CM mitochondria and its crucial role in maintaining mitochondrial redox homeostasis for the first time. Mitochondrial CHK1 function loss mediated the cardiotoxicity induced by AZD7762 and CHK1-knockout. Mechanistically, mitochondrial CHK1 directly phosphorylates SIRT3 and promotes its expression within mitochondria. On the contrary, both AZD7762 or CHK1-knockout and gemcitabine decreased mitochondrial SIRT3 abundance, thus resulting in respiration dysfunction. Further hiPSC-CMs and mice experiments demonstrated that SIRT3 overexpression maintained mitochondrial function while alleviating CM pyroptosis, and thereby improving mice cardiac function. In summary, our results suggest that targeting SIRT3 could represent a novel therapeutic approach for clinical prevention and treatment of cardiotoxicity induced by CHK1 inhibition and gemcitabine.