Abstract Background and aims To establish an echocardiography-based clinical-radiomics model for predicting the prognosis of acute ischemic stroke (AIS) using three-dimensional echocardiography with a specialized LA-dedicated software. Methods A total of 295 patients were prospectively enrolled and randomly divided into a training set (n=236) and an external validation set (n=59). Two distinct mitral valve regions - leaflets (ROI1) and annulus (ROI2) - were manually segmented on echocardiographic images. After dimensionality reduction and feature selection using the minimum redundancy maximum relevance (mRMR) and least absolute shrinkage and selection operator (LASSO) methods, radiomics models were constructed by logistic regression (LR), random forest (RF), C-support vector classification (CSVC), nu-support vector classification (NuSVC), Adaptive Boosting (AdaBoost), and extreme gradient boosting (XGBoost) algorithms. Clinical features identified by mRMR were incorporated to construct a combined model, with model performance assessed through receiver operating characteristic (ROC) analysis. Results From each ROI, a total of 1595 radiomics features were derived. Following dimensionality reduction and selection processes, 32 and 12 valuable radiomic features were identified from ROI1 and ROI2, respectively. Among the various radiomics models tested, the CSVC model based on ROI2 demonstrated superior predictive efficiency and robustness, achieving an area under the curve (AUC) of 0.893 in five-fold cross-validation and 0.773 in the external validation set. Integration of clinical features further enhanced predictive performance, with the combined CSVC model attaining AUC values of 0.903 and 0.884 in cross-validation and external validation respectively. Conclusions The echocardiography-based radiomics, especially on annulus of the mitral valve holds the potential as a non-invasive tool for prognostic prediction of AIS. Conflict of interest Xia Zhang: nothing to disclose; Jiahui Yan:nothing to disclose; Fengmei Li:nothing to disclose;Chen Gang: nothing to disclose; Yanni Wu: nothing to disclose; Hui Li:nothing to disclose

Federated learning (FL) has emerged as a pivotal paradigm for medical image classification in multi-client healthcare applications by enabling collaborative training without sharing raw data. However, its practical deployment remains challenged by negative transfer caused by heterogeneous client distributions and privacy risks introduced by data-sharing-based domain generalization strategies. Existing approaches, including federated model distillation and data-driven domain generalization methods, either lack explicit mechanisms to alleviate inter-client discrepancies or rely on additional information sharing that may compromise privacy. To address these challenges, we propose a Dual-stage Hierarchical Alignment Federated Network (DHA-Fed), which establishes a principled alignment framework from feature-level to prototype-level. In the first stage, a Local-Interim Model Synergistic Feature-Level Alignment strategy (LMSFA) and an interim model mechanism are introduced to explicitly reduce representation discrepancies across clients, thereby mitigating negative transfer. A Bidirectional Divergence loss is further designed to enforce consistent feature alignment while preserving discriminative capability. In the second stage, a Prototype Relaxation Strategy (PRS) is proposed to relax the model's error bound on unseen domains and foster domain-agnostic feature manifolds through prototype contrastive minimization. Extensive experiments on two real-world clinical tasks, including breast cancer and prostate cancer classification, demonstrate that DHA-Fed consistently outperforms state-of-the-art federated learning and domain generalization methods under heterogeneous and privacy-constrained settings.

Hemorrhagic transformation (HT) and its associated neurological deterioration remain major concerns that limit decision-making in thrombolytic treatment for acute ischemic stroke (AIS). Several existing scales based on stroke severity, laboratory indicators, and cranial imaging characteristics are used to estimate the risk of HT. In recent years, machine learning techniques applied in radiomics analysis have been widely used to develop clinical decision-support tools. The present study applied radiomics analysis of pre-thrombolysis cranial non-contrast CT (NCCT) to predict HT after intravenous thrombolysis. Among 1255 cases of anterior-circulation AIS received intravenous recombinant tissue plasminogen activator (rtPA) thrombolysis, 132 patients developed HT on the cranial NCCT scan performed within 36 h post-thrombolysis. Using propensity score matching, a control group of 132 patients without HT, matched for age, gender, baseline systolic blood pressure and onset-to-treatment time, was selected. After excluding 10 patients with unsatisfactory image quality, 254 patients were finally enrolled in the radiomics analysis. They were randomly divided into training cohort and external validation cohort at a ratio of 4:1. Radiomic models consisting of six machine learning (ML) including C-Support Vector Classification (CSVC), Nu-Support Vector Classification (Nu-svc), Adaptive Boosting (AdaBoost), Xtreme Gradient Boosting (Xgboost), logical regression (LR) and random forest were constructed after extracting and selecting optimal features from the training cohort. Model performance was evaluated and compared in the validation cohort using receiver operating characteristic curves and the area under the curve (AUC). A total of 1874 radiomic features were extracted from the region of interest. The t-test and Least Absolute Shrinkage and Selection Operator identified 26 most relevant features. Among the above six ML classifiers, LR model demonstrated strong performance in both the cross-validation and validation cohorts. In five-fold cross-validation, the model achieved an AUC of 0.832, with a sensitivity of 0.76 and a specificity of 0.781 in the training cohort. In the external validation cohort, the AUC was 0.814, with a sensitivity of 0.68 and a specificity of 0.846. In this study, the LR classifier demonstrated favorable predictive performance among the six models evaluated, showing high specificity in identifying HT based on radiomic features extracted from pre-treatment cranial NCCT in anterior-circulation AIS. These findings suggest potential for early risk stratification of patients at elevated risk for post-thrombolysis hemorrhagic transformation, which may contribute to more individualized treatment decisions pending further external validation.

Rationale and Objectives: Diagnosis of carotid plaques from head and neck CT angiography (CTA) scans is typically time-consuming and labor-intensive, leading to limited studies and unpleasant results in this area. The objective of this study is to develop a deep-learning-based model for detection and segmentation of carotid plaques using CTA images. Materials and Methods: CTA images from 1061 patients (765 male; 296 female) with 4048 carotid plaques were included and split into a 75% training-validation set and a 25% independent test set. We built a workflow involving three modified deep learning networks: a plain U-Net for coarse artery segmentation, an Attention U-Net for fine artery segmentation, a dual-channel-input ConvNeXt-based U-Net architecture for plaque segmentation, and post-processing to refine predictions and eliminate false positives. The models were trained on the training-validation set using five-fold cross-validation and further evaluated on the independent test set using comprehensive metrics for segmentation and plaque detection. Results: The proposed workflow was evaluated in the independent test set (261 patients with 902 carotid plaques) and achieved a mean dice similarity coefficient (DSC) of 0.91 +/- 0.04 in artery segmentation, and 0.75 +/- 0.14/0.67 +/- 0.15 in plaque segmentation per artery/patient. The model detected 95.5% (861/902) plaques, including 96.6% (423/438), 95.3% (307/322), and 92.3% (131/142) of calcified, mixed, and soft plaques, with less than one (0.63 +/- 0.93) false positive plaque per patient on average. Conclusion: This study developed an automatic detection and segmentation deep learning-based CAP-Net for carotid plaques using CTA, which yielded promising results in identifying and delineating plaques.

Assessing risk factors for intracranial aneurysm (IA) development on images is crucial for early detection of high-risk cases. IAs often form at bifurcations within the circle of Willis (CoW), but manual assessment of these arteries is both time-consuming and susceptible to inconsistencies. Previous studies on imaging markers for IA development lack sufficient evidence for clinical implications, highlighting the need for automated methods to assess CoW morphology. No systematic approach currently exists to identify the best methodological strategies. To address this, we organized a scientific challenge to compare various techniques against a clinical reference standard. Participants were tasked with (1) automated classification of CoW anatomical variants and (2) automated prediction of CoW artery diameters and bifurcation angles. We provided 300 TOF-MRA scans for training and another 300 for testing, all manually annotated. Submissions were evaluated using balanced accuracy, mean absolute error, and Pearson correlation coefficient metrics. This paper provides a detailed analysis of the results from six participating teams. The findings show that various methods may be suitable for automated CoW assessment, but that these need further improvement to meet clinical standards. The challenge remains open for future submissions, offering a benchmark for new techniques.

The irregular morphology of cerebral aneurysms,especially the presence of a daughter sac,is a crucial risk factor for aneurysm rupture.Clinical assessment of daughter sac relies mainly on image reconstruction by time of flight-magnetic resonance angiography(TOF-MRA)and judgment based on physicians'vision and experience,which limits the efficiency and accuracy of diagnosis.In this paper,we propose an improved parallel multiscale fusion attention network(PMAF-Net)based on 3D ResNet50 for classification.PMAF-Net uses multi-scale convolution and weighted fusion channel and spatial attention weights to enhance the feature extraction capability.The experiment used 291 cases of TOF-MRA data,including 128 cases in the training set,32 cases in the validation set,and 131 cases in the test set.Compared with other classification networks,PMAF-Net performs best on the test set,with the accuracy of 83.97％,recall of 84.48％,precision of 80.33％,and Fl-score of 0.823 5,and the receiver operating characteristic curve(ROC)also reflects the model's optimal classification performance(AUC of 0.900 8).The results show that the network can identify daughter sac type aneurysms more accurately,which is expected to support the assessment and quantification of the risk of aneurysm rupture.

In this paper, we consider an inverse electromagnetic medium scattering problem of reconstructing unknown objects from time-dependent boundary measurements. A novel time-domain direct sampling method is developed for determining the locations of unknown scatterers by using only a single incident source. Notably, our method imposes no restrictions on the waveform of the incident wave. Based on the Fourier-Laplace transform, we first establish the connection between the frequencydomain and the time-domain direct sampling method. Furthermore, we elucidate the mathematical mechanism of the imaging functional through the properties of modified Bessel functions. Theoretical justifications and stability analyses are provided to demonstrate the effectiveness of the proposed method. Finally, several numerical experiments are presented to illustrate the feasibility of our approach.

Vascular segmentation is essential for various medical applications, such as computer-aided diagnosis, treatment planning, and surgical interventions. However, current deep learning-based vascular segmentation methods face two significant challenges: the complex morphological diversity of vascular structures, which results in discontinuous segmentation in small-scale vessels and incomplete preservation of topological integrity, and the adverse effects of noisy labels during network optimization. To address these challenges, we propose a Joint Hierarchical Vascular Morphology Learning and Noise Label Refinement method (JHN-Seg). JHN-Seg introduces a Hierarchical Vascular Morphology-Aware Network (HVMA-Net) that integrates a Multi-Scale Local Morphology-Aware (MLMA) module, employing a multi-pattern convolutional strategy to adaptively capture intricate vascular features across scales. A Global Morphology Preserving (GMP) loss function is incorporated into HVMA-Net to enforce the continuity of small-scale vessels and maintain the integrality of global vascular structures. Furthermore, JHN-Seg introduces an Uncertainty-Aware Distillation (UD) strategy, which incorporates an Uncertainty Label Refinement (ULR) Module for uncertainty-guided noisy label correction by leveraging pixel-wise KL divergence and consistency generated by teacher-student framework. Comprehensive experiments on liver vessel datasets demonstrate that JHN-Seg outperforms other state-of-the-art segmentation methods. The framework's adaptability and performance advancements position it as a transformative solution for vascular segmentation and broad applicability to medical image analysis tasks requiring precise morphological representation and noise-robust learning.

RATIONALE AND OBJECTIVES:Diagnosis of carotid plaques from head and neck CT angiography (CTA) scans is typically time-consuming and labor-intensive, leading to limited studies and unpleasant results in this area. The objective of this study is to develop a deep-learning-based model for detection and segmentation of carotid plaques using CTA images. MATERIALS AND METHODS:CTA images from 1061 patients (765 male; 296 female) with 4048 carotid plaques were included and split into a 75% training-validation set and a 25% independent test set. We built a workflow involving three modified deep learning networks: a plain U-Net for coarse artery segmentation, an Attention U-Net for fine artery segmentation, a dual-channel-input ConvNeXt-based U-Net architecture for plaque segmentation, and post-processing to refine predictions and eliminate false positives. The models were trained on the training-validation set using five-fold cross-validation and further evaluated on the independent test set using comprehensive metrics for segmentation and plaque detection. RESULTS:The proposed workflow was evaluated in the independent test set (261 patients with 902 carotid plaques) and achieved a mean dice similarity coefficient (DSC) of 0.91±0.04 in artery segmentation, and 0.75±0.14/0.67±0.15 in plaque segmentation per artery/patient. The model detected 95.5% (861/902) plaques, including 96.6% (423/438), 95.3% (307/322), and 92.3% (131/142) of calcified, mixed, and soft plaques, with less than one (0.63±0.93) false positive plaque per patient on average. CONCLUSION:This study developed an automatic detection and segmentation deep learning-based CAP-Net for carotid plaques using CTA, which yielded promising results in identifying and delineating plaques.

Segmentation of multiple targets of varying sizes within medical images is of significant importance for the diagnosis of disease and pathological research. Transformer-based methods are emerging in the medical image segmentation, leveraging the powerful yet computationally intensive self-attention mechanism. A variety of attention mechanisms have been proposed to reduce computation at the cost of accuracy loss, utilizing handcrafted patterns within local or artificially defined receptive fields. Furthermore, the common region-based loss functions are insufficient for guiding the transformer to focus on tissue regions, resulting in their unsuitability for the segmentation of tissues with intricate boundaries. This paper presents the development of a bi-level sparse attention network and a narrow band (NB) loss function for the accurate and efficient multi-target segmentation of medical images. In particular, we introduce a bi-level sparse attention module (BSAM) and formulate a segmentation network based on this module. The BSAM consists of coarse-grained patch-level attention and fine-grained pixel-level attention, which captures fine-grained contextual features in adaptive receptive fields learned by patch-level attention. This results in enhanced segmentation accuracy while simultaneously reducing computational complexity. The proposed narrow-band (NB) loss function constructs a target region in close proximity to the tissue boundary. The network is thus guided to perform boundary-aware segmentation, thereby simultaneously alleviating the issues of over-segmentation and under-segmentation. A series of comprehensive experiments on whole brains, brain tumors and abdominal organs, demonstrate that our method outperforms other state-of-the-art segmentation methods. Furthermore, the BSAM and NB loss can be applied flexibly to a variety of network frameworks.

BACKGROUND:Cerebral aneurysms are a type of cerebrovascular disease that poses a severe threat to life and health. Early screening using Time-of-Flight Magnetic Resonance Angiography (TOF-MRA) can effectively reduce the risk of rupture. Despite the importance of early detection, manual image screening remains a laborious and inefficient process. The current thrust of research in computer-aided detection (CAD) methods is to refine neural networks to improve diagnostic accuracy. In our preliminary work, we discovered that utilizing arterial contour as external knowledge guidance can substantially enhance the detection capabilities of existing networks, thus providing a new perspective for optimizing aneurysm detection techniques. METHODS:In this paper, we introduce an innovative approach to building a cerebral aneurysm detection model that employs artery fragment as external guidance data. We propose a hypothesis regarding the optimal distribution pattern of knowledge-guided data based on the brain artery volume of interest (VOI), and based on this, we have developed an end-to-end fully automatic and data-adaptive artery fragment generation method tailored for both training and testing data. Utilizing a multicenter dataset, we tested the performance enhancement capabilities of this method for two commonly used vascular networks, SE-3D UNet and VNet. Furthermore, we conducted a comparative analysis with other guidance methods using the best-performing model to elucidate the mechanisms behind the improved guidance efficacy of our approach. RESULTS:This study amassed a total of 500 cases of 3.0T TOF-MRA data from 13 devices across 6 hospitals, with 400 cases designated as the training set and 100 cases as the test set, while data from one device was exclusively used for testing. The proposed method showed significant improvements for both SE-3D UNet and VNet. Specifically, SE 3D UNET saw a 13.89 % increase in sensitivity while maintaining a false positives per case (FPs/case) of 0.63. For VNet, the FPs/case was reduced by 20 %, with a slight improvement in sensitivity. Compared to other guidance methods, our approach achieved optimal levels in various metrics and exhibited stronger robustness on unfamiliar datasets. CONCLUSIONS:This study presents an artery fragment-guided approach that enhances the detection of cerebral aneurysms in TOF-MRA imaging. It not only outperforms our previous work but also excels when compared to alternative guidance methods. This approach offers a compelling knowledge-guided strategy for cerebral aneurysm detection.

Background:Videofluoroscopic swallowing study (VFSS) employs quantitative analysis methods, which are valued in dysphagia diagnosis for their objectivity and precision. Nonetheless, conventional methods are laborious and time-consuming. Despite advancements in automatic tracking methods, existing software still requires substantial manual intervention and offers a restricted set of quantitative metrics. This study aimed to develop an intelligent VFSS quantitative analysis tool that assists clinicians in dysphagia diagnosis by enabling automated tracking and providing a comprehensive set of three kinematic and seven temporal parameters. Methods:This software utilizes a feature-based target tracking and contour extraction algorithm, which enables accurate and automated detection of hyoid bone displacement, upper esophageal sphincter (UES) opening amplitude and pharyngeal contraction ratio. This study analyzed 82 VFSS samples from Suzhou Municipal Hospital, comprising 40 from 18 dysphagia patients with varied etiologies and 42 from 14 healthy controls. Agreement between automated and manual tracking for the three kinematic parameters was evaluated using Pearson correlation coefficients and relative errors (%). Results:In both patient and control groups, the results showed strong correlations (Pearson's r ranging from 0.947 to 0.995, P value <0.001) between automatic and manual methods across three kinematic parameters. The relative errors of three parameters of two-dimensional range were 5.30±3.79, 3.72±1.93, and 5.22±3.25 in dysphagic patients and 4.62±3.46, 3.07±2.02, and 5.43±3.69 in controls. Comparative analysis demonstrated significantly reduced values in dysphagia patients versus healthy controls across three parameters, characterizing compromised swallowing biomechanics in the dysphagia population. While manual analysis typically requires about one hour per case, the proposed platform completes automatic quantitative analysis within 3-4 minutes per sample. Conclusions:The developed software provides an efficient and user-friendly platform that streamlines dysphagia diagnosis through the automatic tracking and assessment of essential parameters, thereby enhancing the diagnostic accuracy and workflow efficiency.

BACKGROUND:Chronic obstructive pulmonary disease is a common respiratory disease. The severity of acute exacerbation of chronic obstructive pulmonary disease is related to disease progression and risk of death. However, the existing grading standards mainly depend on indicators, such as respiratory rate, whether to apply assisted respiratory muscles, and changes in consciousness state, and only reflect the subjective judgment. Imaging omics can extract muscle characteristic data for more complex analysis, which helps to provide a more objective and accurate method to assess the severity of disease for clinic. OBJECTIVES:The purpose of this study is to construct a severity prediction model based on the combination of chest CT muscle imaging features and clinical data in hospitalized patients with AECOPD. METHODS:234 hospitalized patients with AECOPD were retrospectively included, divided into 79 grade I, 74 grade II, and 81 grade III. Clinical data and chest CT images were collected. Construction of clinical feature model combined with muscle imaging omics model based on Python machine learning platform. RESULTS:The number of hospitalizations for acute exacerbation, disease course, risk of acute exacerbation in stable stage, white blood cell count, neutrophil count, creatinine, and N-terminal B-type natriuretic peptide precursor were statistically different among hospitalized patients with AECOPD in the last year (all P < 0.05). The best model to predict the severity of AECOPD by cascade probability combination method is Xgboost model with AUC of 0.890. CONCLUSIONS:The disease grading prediction model of AECOPD inpatients constructed based on clinical data and muscle imaging omics characteristics has good performance, and has great potential in assisting clinicians to more accurately stratify the risk of AECOPD inpatients.

Background:Early alterations in cerebrospinal fluid (CSF) may play a critical role in the progression of acute ischemic stroke (AIS). This study aimed to develop a CSF-based clinical-radiomics model for predicting the functional outcomes in AIS patients following intravenous thrombolysis (IVT). Methods:This study included 308 AIS patients within 6 h of onset, divided into a training set (n=246) and a hold-out test set (n=62). Functional outcome was assessed at discharge using the modified Rankin Scale (mRS), dichotomized into good (mRS ≤2) and poor (mRS >2) outcomes. CSF regions were automatically segmented on non-contrast computed tomography images, and radiomics features were extracted. After feature selection was sequentially performed using minimum redundancy maximum relevance algorithm followed by least absolute shrinkage and selection operator, the radiomics signature model was constructed. Clinical features were selected through univariate and multivariate logistic regressions and subsequently integrated with radiomics features to develop combined models. Three machine learning classifiers, including Nu Support Vector Classification (NuSVC), logistic regression, and random forest, were trained and tested for prognostic prediction. Model performance was evaluated using receiver operating characteristic curve analysis, decision curve analysis (DCA), and calibration curves, with internal validation via five-fold cross-validation. Results:Among the 308 patients included, 155 patients had a good outcome and 153 had a poor outcome. A total of 1,874 radiomics features were extracted, among which 21 were selected for the radiomics model. Three clinical features were also identified for the clinical model. The combined clinical-radiomics model significantly outperformed single-modality models across all classifiers. The NuSVC-based combined model achieved the best performance, with an area under the curve (AUC) of 0.870 [95% confidence interval (CI): 0.850-0.889] in cross-validation and 0.893 (95% CI: 0.817-0.968) in the test cohort. DCA further confirmed the superior clinical utility of the combined model. Conclusions:This study demonstrates the value of CSF radiomics signatures in predicting functional outcomes in AIS patients after IVT. The CSF-based combined model exhibited strong prognostic performance and clinical utility, highlighting its potential for supporting treatment decision-making.

Developing robust methods for evaluating protein-ligand interactions has been a long-standing problem. Data-driven methods may memorize ligand and protein training data rather than learning protein-ligand interactions. Here we show a scoring approach called EquiScore, which utilizes a heterogeneous graph neural network to integrate physical prior knowledge and characterize protein-ligand interactions in equivariant geometric space. EquiScore is trained based on a new dataset constructed with multiple data augmentation strategies and a stringent redundancy-removal scheme. On two large external test sets, EquiScore consistently achieved top-ranking performance compared to 21 other methods. When EquiScore is used alongside different docking methods, it can effectively enhance the screening ability of these docking methods. EquiScore also showed good performance on the activity-ranking task of a series of structural analogues, indicating its potential to guide lead compound optimization. Finally, we investigated different levels of interpretability of EquiScore, which may provide more insights into structure-based drug design. Machine learning can improve scoring methods to evaluate protein-ligand interactions, but achieving good generalization is an outstanding challenge. Cao et al. introduce EquiScore, which is based on a graph neural network that integrates physical knowledge and is shown to have robust capabilities when applied to unseen protein targets.

Background2D CT image-guided radiofrequency ablation (RFA) is an exciting minimally invasive treatment that can destroy liver tumors without removing them. However, CT images can only provide limited static information, and the tumor will move with the patient's respiratory movement. Therefore, how to accurately locate tumors under free conditions is an urgent problem to be solved at present.PurposeThe purpose of this study is to propose a respiratory correlation prediction model for mixed reality surgical assistance system, Riemannian and Multivariate Feature Enhanced Temporal Convolutional Network (R-MFE-TCN), and to achieve accurate respiratory correlation prediction.MethodsThe model adopts a respiration-oriented Riemannian information enhancement strategy to expand the diversity of the dataset. A new Multivariate Feature Enhancement module (MFE) is proposed to retain respiratory data information, so that the network can fully explore the correlation of internal and external data information, the dual-channel is used to retain multivariate respiratory feature, and the Multi-headed Self-attention obtains respiratory peak-to-valley value periodic information. This information significantly improves the prediction performance of the network. At the same time, the PSO algorithm is used for hyperparameter optimization. In the experiment, a total of seven patients' internal and external respiratory motion trajectories were obtained from the dataset, and the first six patients were selected as the training set. The respiratory signal collection frequency was 21 Hz.ResultsA large number of experiments on the dataset prove the good performance of this method, which improves the prediction accuracy while also having strong robustness. This method can reduce the delay deviation under long window prediction and achieve good performance. In the case of 400 ms, the average RMSE and MAE are 0.0453 and 0.0361 mm, respectively, which is better than other research methods.ConclusionThe R-MFE-TCN can be extended to respiratory correlation prediction in different clinical situations, meeting the accuracy requirements for respiratory delay prediction in surgical assistance.

Regulatory T (Treg) cells are critical for immune tolerance but also form a barrier to antitumor immunity. As therapeutic strategies involving Treg cell depletion are limited by concurrent autoimmune disorders, identification of intratumoral Treg cell-specific regulatory mechanisms is needed for selective targeting. Epigenetic modulators can be targeted with small compounds, but intratumoral Treg cell-specific epigenetic regulators have been unexplored. Here, we show that JMJD1C, a histone demethylase upregulated by cytokines in the tumor microenvironment, is essential for tumor Treg cell fitness but dispensable for systemic immune homeostasis. JMJD1C deletion enhanced AKT signals in a manner dependent on histone H3 lysine 9 dimethylation (H3K9me2) demethylase and STAT3 signals independently of H3K9me2 demethylase, leading to robust interferon-γ production and tumor Treg cell fragility. We have also developed an oral JMJD1C inhibitor that suppresses tumor growth by targeting intratumoral Treg cells. Overall, this study identifies JMJD1C as an epigenetic hub that can integrate signals to establish tumor Treg cell fitness, and we present a specific JMJD1C inhibitor that can target tumor Treg cells without affecting systemic immune homeostasis.

PURPOSE:Multimodal registration is a key task in medical image analysis. Due to the large differences of multimodal images in intensity scale and texture pattern, it is a great challenge to design distinctive similarity metrics to guide deep learning-based multimodal image registration. Besides, since the limitation of the small receptive field, existing deep learning-based methods are mainly suitable for small deformation, but helpless for large deformation. To address the above issues, we present an unsupervised multimodal image registration method based on the multiscale integrated spatial-weight module and dual similarity guidance. METHODS:In this method, a U-shape network with our multiscale integrated spatial-weight module is embedded into a multi-resolution image registration architecture to achieve end-to-end large deformation registration, where the spatial-weight module can effectively highlight the regions with large deformation and aggregate discriminative features, and the multi-resolution architecture further helps to solve the optimization problem of the network in a coarse-to-fine pattern. Furthermore, we introduce a special loss function based on dual similarity, which represents both global gray-scale similarity and local feature similarity, to optimize the unsupervised multimodal registration network. RESULTS:We verified the effectiveness of the proposed method on liver CT-MR images. Experimental results indicate that the proposed method achieves the optimal DSC value and TRE value of 92.70 ± 1.75(%) and 6.52 ± 2.94(mm), compared with other state-of-the-art registration algorithms. CONCLUSION:The proposed method can accurately estimate the large deformation field by aggregating multiscale features, and achieve higher registration accuracy and fast registration speed. Comparative experiments also demonstrate the effectiveness and generalization ability of the algorithm.

Structure-based lead optimization is an open challenge in drug discovery, which is still largely driven by hypotheses and depends on the experience of medicinal chemists. We here propose a pairwise binding comparison network (PBCNet) based on physics-informed graph attention mechanism, specifically tailored for ranking relative binding affinity among congeneric ligands. Benchmarking on two held-out sets (provided by Schrödinger, Inc. and Merck KGaA) containing over 460 ligands and 16 targets, PBCNet demonstrated significant advantages in terms of both prediction accuracy and computational efficiency. Equipped with a fine-tuning operation, the performance of PBCNet reaches that of Schrödinger's FEP+, which is much more computationally intensive and requires significant expert intervention. A further simulation-based experiment showed that active learning-optimized PBCNet may accelerate lead optimization campaigns by 30%. Finally, for the convenience of users, a web service (https://pbcnet.alphama.com.cn/index) for PBCNet is established to facilitate complex relative binding affinity prediction through an easy-to-operate graphical interface.

ABSTRACTDeveloping robust methods for evaluating protein-ligand interactions has been a long-standing problem. Here, we propose a novel approach called EquiScore, which utilizes an equivariant heterogeneous graph neural network to integrate physical prior knowledge and characterize protein-ligand interactions in equivariant geometric space. To improve generalization performance, we constructed a dataset called PDBscreen and designed multiple data augmentation strategies suitable for training scoring methods. We also analyzed potential risks of data leakage in commonly used data-driven modeling processes and proposed a more stringent redundancy removal scheme to alleviate this problem. On two large external test sets, EquiScore outperformed 21 methods across a range of screening performance metrics, and this performance was insensitive to binding pose generation methods. EquiScore also showed good performance on the activity ranking task of a series of structural analogs, indicating its potential to guide lead compound optimization. Finally, we investigated different levels of interpretability of EquiScore, which may provide more insights into structure-based drug design.