Semi-Supervised Graph Structure Learning via Dual Reinforcement of Label and Prior Structure.

Graph neural networks (GNNs) have achieved considerable success in dealing with graph-structured data by the message-passing mechanism. Actually, this mechanism relies on a fundamental assumption that the graph structure along which information propagates is perfect. However, the real-world graphs are inevitably incomplete or noisy, which violates the assumption, thus resulting in limited performance. Therefore, optimizing graph structure for GNNs is indispensable and important. Although current semi-supervised graph structure learning (GSL) methods have achieved a promising performance, the potential of labels and prior graph structure has not been fully exploited yet. Inspired by this, we examine GSL with dual reinforcement of label and prior structure in this article. Specifically, to enhance label utilization, we first propose to construct the prior label-constrained matrices to refine the graph structure by identifying label consistency. Second, to adequately leverage the prior structure to guide GSL, we develop spectral contrastive learning that extracts global properties embedded in the prior graph structure. Moreover, contrastive fusion with prior spatial structure is further adopted, which promotes the learned structure to integrate local spatial information from the prior graph. To extensively evaluate our proposal, we perform sufficient experiments on seven benchmark datasets, where experimental results confirm the effectiveness of our method and the rationality of the learned structure from various aspects.

Parallel Dual-Branch Fusion Network for Epileptic Seizure Prediction.

Epilepsy is a prevalent chronic disorder of the central nervous system. The timely and accurate seizure prediction using the scalp Electroencephalography (EEG) signal can make patients adopt reasonable preventive measures before seizures occur and thus reduce harm to patients. In recent years, deep learning-based methods have made significant progress in solving the problem of epileptic seizure prediction. However, most current methods mainly focus on modeling short- or long-term dependence in EEG, while neglecting to consider both. In this study, we propose a Parallel Dual-Branch Fusion Network (PDBFusNet) which aims to combine the complementary advantages of Convolutional Neural Network (CNN) and Transformer. Specifically, the features of the EEG signal are first extracted using Mel Frequency Cepstral Coefficients (MFCC). Then, the extracted features are delivered into the parallel dual-branches to simultaneously capture the short- and long-term dependencies of EEG signal. Further, regarding the Transformer branch, a novel feature fusion module is developed to enhance the ability of utilizing time, frequency, and channel information. To evaluate our proposal, we perform sufficient experiments on the public epileptic EEG dataset CHB-MIT, where the accuracy, sensitivity, specificity and precision are 95.76%, 95.81%, 95.71% and 95.71%, respectively. PDBFusNet shows superior performance compared to state-of-the-art competitors, which confirms the effectiveness of our proposal.

Deep Survival Analysis With Latent Clustering and Contrastive Learning.

Survival analysis is employed to analyze the time before the event of interest occurs, which is broadly applied in many fields. The existence of censored data with incomplete supervision information about survival outcomes is one key challenge in survival analysis tasks. Although some progress has been made on this issue recently, the present methods generally treat the instances as separate ones while ignoring their potential correlations, thus rendering unsatisfactory performance. In this study, we propose a novel Deep Survival Analysis model with latent Clustering and Contrastive learning (DSACC). Specifically, we jointly optimize representation learning, latent clustering and survival prediction in a unified framework. In this way, the clusters distribution structure in latent representation space is revealed, and meanwhile the structure of the clusters is well incorporated to improve the ability of survival prediction. Besides, by virtue of the learned clusters, we further propose a contrastive loss function, where the uncensored data in each cluster are set as anchors, and the censored data are treated as positive/negative sample pairs according to whether they belong to the same cluster or not. This design enables the censored data to make full use of the supervision information of the uncensored samples. Through extensive experiments on four popular clinical datasets, we demonstrate that our proposed DSACC achieves advanced performance in terms of both C-index (0.6722, 0.6793, 0.6350, and 0.7943) and Integrated Brier Score (IBS) (0.1616, 0.1826, 0.2028, and 0.1120).

Molecular dynamics simulation on the displacement behaviour of crude oil by CO2/CH4 mixtures on a silica surface.

Produced gas re-injection is an effective and eco-friendly approach for enhancing oil recovery from shale oil reservoirs. However, the interactions between different gas phase components, and the oil phase and rocks are still unclear during the re-injection process. This study aims to investigate the potential of produced gas re-injection, particularly focusing on the effects of methane (CH4) content in the produced gas on shale oil displacement. Molecular dynamics simulations were employed to analyze the interactions between gas, oil, and matrix phases with different CH4 proportions (0%, 25%, 50%, and 100%), alkanes and under various burial depth. Results show that a 25% CH4 content in the produced gas achieves almost the same displacement effect as pure carbon dioxide (CO2) injection. However, when the CH4 content increases to 50% and 100%, the interaction between gas and quartz becomes insufficient to effectively isolate oil from quartz, causing only expansion and slight dispersion. Interestingly, the presence of CH4 has a synergistic effect on CO2, facilitating the diffusion of CO2 into the oil film. During the gas stripping process, CO2 is the main factor separating oil from quartz, while CH4 mainly contributes to oil expansion. In addition, for crude oil containing a large amount of light alkanes, extracting light components through mixed gas may be more effective than pure CO2. This study offers valuable insights for applications of produced gas re-injection to promote shale oil recovery.

Surgical evidence-based comparison of [68Ga]Ga-FAPI-04 PET and MRI-DWI for assisting debulking surgery in ovarian cancer patients.

PURPOSE: Imaging assessment of abdominopelvic tumor burden is crucial for debulking surgery decision in ovarian cancer patients. This study aims to compare the efficiency of [68Ga]Ga-FAPI-04 FAPI PET and MRI-DWI in the preoperative evaluation and its potential impact to debulking surgery decision. METHODS: Thirty-six patients with suspected/confirmed ovarian cancer were enrolled and underwent integrated [68Ga]Ga-FAPI-04 PET/MRI. Nineteen patients (15 stage III-IV and 4 I-II stage) who underwent debulking surgery were involved in the diagnostic efficiency analysis. The images of [68Ga]Ga-FAPI-04 PET and MRI-DWI were visually analyzed respectively. Immunohistochemistry on FAP was performed in metastatic lesions to investigate the radiological missing of [68Ga]Ga-FAPI-04 PET as well as its different performance in primary debulking surgery (PDS) and interval debulking surgery (IDS) patients. Potential imaging impact on management was also studied in 35 confirmed ovarian cancer patients. RESULTS: [68Ga]Ga-FAPI-04 PET displayed higher sensitivity (76.8% vs.59.9%), higher accuracy (84.9% vs. 80.7%), and lower missing rate (23.2% vs. 40.1%) than MRI-DWI in detecting abdominopelvic metastasis. The diagnostic superiority of [68Ga]Ga-FAPI-04 PET is more obvious in PDS patients but diminished in IDS patients. [68Ga]Ga-FAPI-04 PET outperformed MRI-DWI in 70.8% abdominopelvic regions (17/24), which contained seven key regions that impact the resectability and surgical complexity. MRI-DWI hold advantage in the peritoneal surface of the bladder and the central tendon of the diaphragm. Of the contradictory judgments between the two modalities (14.9%), [68Ga]Ga-FAPI-04 PET correctly identified more lesions, particularly in PDS patients (73.8%). In addition, FAP expression was independent of lesion size and decreased in IDS patients. [68Ga]Ga-FAPI-04 PET changed 42% of surgical planning that was previously based on MRI-DWI. CONCLUSION: [68Ga]Ga-FAPI-04 PET is more efficient in assisting debulking surgery in ovarian cancer patients than MRI-DWI. Integrated [68Ga]Ga-FAPI-04 PET/MR imaging is a potential method for planning debulking surgery in ovarian cancer patients.

Adaptive-weighted deep multi-view clustering with uniform scale representation.

Multi-view clustering has attracted growing attention owing to its powerful capacity of multi-source information integration. Although numerous advanced methods have been proposed in past decades, most of them generally fail to distinguish the unequal importance of multiple views to the clustering task and overlook the scale uniformity of learned latent representation among different views, resulting in blurry physical meaning and suboptimal model performance. To address these issues, in this paper, we propose a joint learning framework, termed Adaptive-weighted deep Multi-view Clustering with Uniform scale representation (AMCU). Specifically, to achieve more reasonable multi-view fusion, we introduce an adaptive weighting strategy, which imposes simplex constraints on heterogeneous views for measuring their varying degrees of contribution to consensus prediction. Such a simple yet effective strategy shows its clear physical meaning for the multi-view clustering task. Furthermore, a novel regularizer is incorporated to learn multiple latent representations sharing approximately the same scale, so that the objective for calculating clustering loss cannot be sensitive to the views and thus the entire model training process can be guaranteed to be more stable as well. Through comprehensive experiments on eight popular real-world datasets, we demonstrate that our proposal performs better than several state-of-the-art single-view and multi-view competitors.

Indirect estimation of pediatric reference interval via density graph deep embedded clustering.

Establishing reference intervals (RIs) for pediatric patients is crucial in clinical decision-making, and there is a critical gap of pediatric RIs in China. However, the direct sampling technique for establishing RIs is resource-intensive and ethically challenging. Indirect estimation methods, such as unsupervised clustering algorithms, have emerged as potential alternatives for predicting reference intervals. This study introduces deep graph clustering methods into indirect estimation of pediatric reference intervals. Specifically, we propose a Density Graph Deep Embedded Clustering (DGDEC) algorithm, which incorporates a density feature extractor to enhance sample representation and provides additional perspectives for distinguishing different levels of health status among populations. Additionally, we construct an adjacency matrix by computing the similarity between samples after feature enhancement. The DGDEC algorithm leverages the adjacency matrix to capture the interrelationships between patients and divides patients into different groups, thereby estimating reference intervals for the potential healthy population. The experimental results demonstrate that when compared to other indirect estimation techniques, our method ensures the predicted pediatric reference intervals in different age and gender groups are closer to the true values while maintaining good generalization performance. Additionally, through ablation experiments, our study confirms that the similarity between patients and the multi-scale density features of samples can effectively describe the potential health status of patients.

Machine learning models developed and internally validated for predicting chronicity in pediatric immune thrombocytopenia.

BACKGROUND: Primary immune thrombocytopenia (ITP) in children is typically self-limiting; however, 20% to 30% of patients may experience prolonged thrombocytopenia lasting over a year. The challenge is predicting chronicity to ensure personalized treatment approaches. OBJECTIVES: To address this issue, we developed and internally validated 4 machine learning (ML) models using demographic and immunologic characteristics to predict the likelihood of chronicity. METHODS: The present study was conducted at Beijing Children's Hospital from June 2018 to December 2021, aiming to develop predictive models for determining the chronicity of pediatric ITP. Four ML models, based on a logistic regression classifier, random forest classifier, eXtreme Gradient Boosting (XGBoost), and support vector machine, were employed. These models used a set of 16 variables, including 14 immunologic and 2 demographic predictors. The performance evaluation criteria included prediction accuracy, precision, recall, F1 score, and area under the receiver operating characteristic curve (AUROC). RESULTS: Data were collected from 662 patients who were randomly assigned to either a training dataset or a testing dataset using a random number generator. Among them, 26.5% had chronic disease. All models performed well, with AUROC values ranging from 0.81 to 0.84, and XGBoost was selected for its highest AUROC score and interpretability in constructing the predictive model. Age, T helper 17, T helper 17-to-regulatory T cell ratio, T helper 1, and double-negative T cells were identified as significant predictors by the XGBoost algorithm. CONCLUSION: We developed a precise predictive model for chronicity in pediatric ITP using ML during the initial phase. The XGBoost model achieved high predictive accuracy by using individual patient clinical parameters and demonstrated commendable interpretability.

Machine learning model for predicting physical activity related bleeding risk in Chinese boys with haemophilia A.

BACKGROUND: Physical activity is a crucial part of an active lifestyle for haemophiliac children. However, the fear of bleeds has been identified as barriers to participating physical activity for haemophiliac children even with prophylaxis. Lack of evidence and metrics driven by data is key problem. OBJECTIVES: We aim to develop machine learning models based on clinical data with multiple potential factors considered to predict risk of physical activity bleeding for haemophilia children with prophylaxis. METHODS: From this cohort study, we collected information on 98 haemophiliac children with adequate prophylaxis (trough FVIII:C level > 1 %). The involved potential predictor variables include demographic information, treatment information, physical activity, joint evaluation, and pharmacokinetic parameters, etc. We applied CoxPH, Random Survival Forests (RSF) and DeepSurv to construct prediction models for the risk of bleeding during physical activities. All three survival analysis models were internally and externally validated. RESULTS: A total of 98 patients were enrolled in this study. Their median age was 7.9 (5.5, 10.2) years. The CoxPH, RSF and DeepSurv models' discriminative and calibration abilities were all high, and the RSF model had the best performance (Internal validation: C-index, 0.7648 ± 0.0139; Brier Score, 0.1098 ± 0.0015; External validation: C-index, 0.7260 ± 0.0154; Brier Score, 0.0930 ± 0.0018). The prediction curves demonstrated that the developed RSF model can distinguish the risks well between bleeding and non-bleeding patients, as well as patients with different levels of physical activity. Meanwhile, the feature importance analysis confirmed that physical activity bleeding was deduced by comprehensive effects of various factors, and the importance of different factors on bleeding outcome is discrepant. CONCLUSIONS: This study revealed from the mechanism that it is necessary to incorporate multiple factors to accurately predict physical activity related bleeding risk. In clinical practice, the designed machine learning models can provide guidance for children with haemophilia A to positively participate in physical activity.

Clustering Enhanced Multiplex Graph Contrastive Representation Learning.

Multiplex graph representation learning has attracted considerable attention due to its powerful capacity to depict multiple relation types between nodes. Previous methods generally learn representations of each relation-based subgraph and then aggregate them into final representations. Despite the enormous success, they commonly encounter two challenges: 1) the latent community structure is overlooked and 2) consistent and complementary information across relation types remains largely unexplored. To address these issues, we propose a clustering-enhanced multiplex graph contrastive representation learning model (CEMR). In CEMR, by formulating each relation type as a view, we propose a multiview graph clustering framework to discover the potential community structure, which promotes representations to incorporate global semantic correlations. Moreover, under the proposed multiview clustering framework, we develop cross-view contrastive learning and cross-view cosupervision modules to explore consistent and complementary information in different views, respectively. Specifically, the cross-view contrastive learning module equipped with a novel negative pairs selecting mechanism enables the view-specific representations to extract common knowledge across views. The cross-view cosupervision module exploits the high-confidence complementary information in one view to guide low-confidence clustering in other views by contrastive learning. Comprehensive experiments on four datasets confirm the superiority of our CEMR when compared to the state-of-the-art rivals.

Constrained Maximum Cross-Domain Likelihood for Domain Generalization.

As a recent noticeable topic, domain generalization aims to learn a generalizable model on multiple source domains, which is expected to perform well on unseen test domains. Great efforts have been made to learn domain-invariant features by aligning distributions across domains. However, existing works are often designed based on some relaxed conditions which are generally hard to satisfy and fail to realize the desired joint distribution alignment. In this article, we propose a novel domain generalization method, which originates from an intuitive idea that a domain-invariant classifier can be learned by minimizing the Kullback-Leibler (KL)-divergence between posterior distributions from different domains. To enhance the generalizability of the learned classifier, we formalize the optimization objective as an expectation computed on the ground-truth marginal distribution. Nevertheless, it also presents two obvious deficiencies, one of which is the side-effect of entropy increase in KL-divergence and the other is the unavailability of ground-truth marginal distributions. For the former, we introduce a term named maximum in-domain likelihood to maintain the discrimination of the learned domain-invariant representation space. For the latter, we approximate the ground-truth marginal distribution with source domains under a reasonable convex hull assumption. Finally, a constrained maximum cross-domain likelihood (CMCL) optimization problem is deduced, by solving which the joint distributions are naturally aligned. An alternating optimization strategy is carefully designed to approximately solve this optimization problem. Extensive experiments on four standard benchmark datasets, i.e., Digits-DG, PACS, Office-Home, and miniDomainNet, highlight the superior performance of our method.

Semisupervised Progressive Representation Learning for Deep Multiview Clustering.

Multiview clustering has become a research hotspot in recent years due to its excellent capability of heterogeneous data fusion. Although a great deal of related works has appeared one after another, most of them generally overlook the potentials of prior knowledge utilization and progressive sample learning, resulting in unsatisfactory clustering performance in real-world applications. To deal with the aforementioned drawbacks, in this article, we propose a semisupervised progressive representation learning approach for deep multiview clustering (namely, SPDMC). Specifically, to make full use of the discriminative information contained in prior knowledge, we design a flexible and unified regularization, which models the sample pairwise relationship by enforcing the learned view-specific representation of must-link (ML) samples (cannot-link (CL) samples) to be similar (dissimilar) with cosine similarity. Moreover, we introduce the self-paced learning (SPL) paradigm and take good care of two characteristics in terms of both complexity and diversity when progressively learning multiview representations, such that the complementarity across multiple views can be squeezed thoroughly. Through comprehensive experiments on eight widely used image datasets, we prove that the proposed approach can perform better than the state-of-the-art opponents.

Consistency and adversarial semi-supervised learning for medical image segmentation.

Medical image segmentation based on deep learning has made enormous progress in recent years. However, the performance of existing methods generally heavily relies on a large amount of labeled data, which are commonly expensive and time-consuming to obtain. To settle above issue, in this paper, a novel semi-supervised medical image segmentation method is proposed, in which the adversarial training mechanism and the collaborative consistency learning strategy are introduced into the mean teacher model. With the adversarial training mechanism, the discriminator can generate confidence maps for unlabeled data, such that more reliable supervised information for the student network is exploited. In the process of adversarial training, we further propose a collaborative consistency learning strategy by which the auxiliary discriminator can assist the primary discriminator in achieving supervised information with higher quality. We extensively evaluate our method on three representative yet challenging medical image segmentation tasks: (1) skin lesion segmentation from dermoscopy images in the International Skin Imaging Collaboration (ISIC) 2017 dataset; (2) optic cup and optic disk (OC/OD) segmentation from fundus images in the Retinal Fundus Glaucoma Challenge (REFUGE) dataset; and (3) tumor segmentation from lower-grade glioma (LGG) tumors images. The experimental results validate the superiority and effectiveness of our proposal when compared with the state-of-the-art semi-supervised medical image segmentation methods.

Affine Subspace Robust Low-Rank Self-Representation: From Matrix to Tensor.

Low-rank self-representation based subspace learning has confirmed its great effectiveness in a broad range of applications. Nevertheless, existing studies mainly focus on exploring the global linear subspace structure, and cannot commendably handle the case where the samples approximately (i.e., the samples contain data errors) lie in several more general affine subspaces. To overcome this drawback, in this paper, we innovatively propose to introduce affine and nonnegative constraints into low-rank self-representation learning. While simple enough, we provide their underlying theoretical insight from a geometric perspective. The union of two constraints geometrically restricts each sample to be expressed as a convex combination of other samples in the same subspace. In this way, when exploring the global affine subspace structure, we can also consider the specific local distribution of data in each subspace. To comprehensively demonstrate the benefits of introducing two constraints, we instantiate three low-rank self-representation methods ranging from single-view low-rank matrix learning to multi-view low-rank tensor learning. We carefully design the solution algorithms to efficiently optimize the proposed three approaches. Extensive experiments are conducted on three typical tasks, including single-view subspace clustering, multi-view subspace clustering, and multi-view semi-supervised classification. The notably superior experimental results powerfully verify the effectiveness of our proposals.

Analysis of clinical diagnosis and treatment of intestinal volvulus.

BACKGROUND: The aim of this study is to investigate the clinical characteristics and treatment experience of intestinal volvulus, and to analyze the incidence of adverse events and related risk factors of intestinal volvulus. METHODS: Thirty patients with intestinal volvulus admitted to the Digestive Emergency Department of Xijing Hospital from January 2015 to December 2020 were selected. The clinical manifestations, laboratory tests, treatment and prognosis were retrospectively analyzed. RESULTS: A total of 30 patients with volvulus were enrolled in this study, including 23 males (76.7%), with a median age of 52 years (33-66 years). The main clinical manifestations were abdominal pain in 30 cases (100%), nausea and vomiting in 20 cases (67.7%), cessation of exhaust and defecation in 24 cases (80%), and fever in 11 cases (36.7%). The positions of intestinal volvulus were jejunum in 11 cases (36.7%), ileum and ileocecal in 10 cases (33.3%), sigmoid colon in 9 cases (30%). All 30 patients received surgical treatment. Among the 30 patients underwent surgery, 11 patients developed intestinal necrosis. We found that the longer the disease duration (> 24 h), the higher the incidence of intestinal necrosis, and the higher the incidence of ascites, white blood cell count and neutrophil ratio in the intestinal necrosis group were significantly higher than those in the non-intestinal necrosis group (p < 0.05). After treatment, 1 patient died of septic shock after operation, and 2 patients with recurrent volvulus were followed up within 1 year. The overall cure rate was 90%, the mortality rate was 3.3%, and the recurrence rate was 6.6%. CONCLUSION: Laboratory examination, abdominal CT and dual-source CT are very important for the diagnosis of volvulus in patients with abdominal pain as the main symptom. Increased white blood cell count, neutrophil ratio, ascites and long course of disease are important for predicting intestinal volvulus accompanied by intestinal necrosis. Early diagnosis and timely intervention can save lives and prevent serious complications.

Case report: Long response to PD-1 blockade after failure of trastuzumab plus chemotherapy in advanced Epstein-Barr virus-associated gastric cancer.

Background: Trastuzumab-containing chemotherapy is the first-line treatment for advanced gastric cancer (GC) with HER2 positive. Although PD-1 inhibitors significantly improved the outcome of GC patient's refractory to previous chemotherapy regimens, few studies explore the role of anti-PD-1 therapy overcomes resistance to trastuzumab plus chemotherapy in advanced Epstein-Barr Virus-associated gastric cancer (EBVaGC) with PD-L1 and HER2 positive. Case Presentation: We report a case of advanced EBVaGC in a 45-year-old man presenting with fatigue, dysphagia, and weight loss for several months. Initial endoscopy revealed a large tumor at the gastroesophageal junction. Computed tomography revealed GC accompanied by multiple lymph nodes and hepatic and pulmonary metastases. The immunohistochemistry indicated that HER-2 and PD-L1 were overexpressed, and tumor cells were positive for EBV-encoded small RNA (EBER) by in situ hybridization. Trastuzumab plus DCS was started as first-line chemotherapy with a PFS of 4 months and shifted to trastuzumab plus FOLFIRI or gemcitabine as second-/third-line therapy. After five-cycle nivolumab monotherapy, the patient received partial response and was treated with total radical gastrectomy plus sequential radiotherapy. He continued the postoperative immunotherapy over 30 cycles with a PFS of 28 months. Due to a new abdominal lymph node metastasis confirmed by PET-CT, he received toripalimab as the next-line treatment and achieved complete remission as the best objective response. Summary: We presented an advanced HER2-positive EBVaGC patient with PD-L1 high expression, refractory to trastuzumab plus chemotherapy, and had a durable clinical benefit sequence with a single dose of the PD-1 inhibitor.

Neoadjuvant study of niraparib in patients with HER2-negative, BRCA-mutated, resectable breast cancer.

This single-arm pilot study (NCT03329937) evaluated neoadjuvant niraparib antitumor activity and safety in patients with localized HER2-negative, BRCA-mutated breast cancer. Twenty-one patients received niraparib 200 mg once daily in 28-day cycles. After 2 cycles, tumor response (≥30% reduction from baseline) by MRI was 90.5% and 40.0% (6 of 15) of patients who received only niraparib (2-6 cycles) had pathological complete response; no new safety signals were identified. High niraparib intratumoral concentration was observed.