When surface-enhanced Raman spectroscopy meets complex biofluids: A new representation strategy for reliable and comprehensive characterization.

BACKGROUND: Surface-enhanced Raman spectroscopy (SERS) has gained increasing importance in molecular detection due to its high specificity and sensitivity. Complex biofluids (e.g., cell lysates and serums) typically contain large numbers of different bio-molecules with various concentrations, making it extremely challenging to be reliably and comprehensively characterized via conventional single SERS spectra due to uncontrollable electromagnetic hot spots and irregular molecular motions. The traditional approach of directly reading out the single SERS spectra or calculating the average of multiple spectra is less likely to take advantage of the full information of complex biofluid systems. RESULTS: Herein, we propose to construct a spectral set with unordered multiple SERS spectra as a novel representation strategy to characterize full molecular information of complex biofluids. This new SERS representation not only contains details from each single spectra but captures the temporal/spatial distribution characteristics. To address the ordering-independent property of traditional chemometric methods (e.g., the Euclidean distance and the Pearson correlation coefficient), we introduce Wasserstein distance (WD) to quantitatively and comprehensively assess the quality of spectral sets on biofluids. WD performs its superiority for the quantitative assessment of the spectral sets. Additionally, WD benefits from its independence of the ordering of spectra in a spectral set, which is undesirable for traditional chemometric methods. With experiments on cell lysates and human serums, we successfully achieve the verification for the reproducibility between parallel samples, the uniformity at different positions in the same sample, the repeatability from multiple tests at one location of the same sample, and the cardinality effect of the spectral set. SERS spectral sets also manage to distinguish different classes of human serums and achieve higher accuracy than the traditional prostate-specific antigen in prostate cancer classification. SIGNIFICANCE: The proposed SERS spectral set is a robust representation approach in accessing full information of biological samples compared to relying on a single or averaged spectra in terms of reproducibility, uniformity, repeatability, and cardinality effect. The application of WD further demonstrates the effectiveness and robustness of spectral sets in characterizing complex biofluid samples, which extends and consolidates the role of SERS.

A Generalizable Causal-Invariance-Driven Segmentation Model for Peripancreatic Vessels.

Segmenting peripancreatic vessels in CT, including the superior mesenteric artery (SMA), the coeliac artery (CA), and the partial portal venous system (PPVS), is crucial for preoperative resectability analysis in pancreatic cancer. However, the clinical applicability of vessel segmentation methods is impeded by the low generalizability on multi-center data, mainly attributed to the wide variations in image appearance, namely the spurious correlation factor. Therefore, we propose a causal-invariance-driven generalizable segmentation model for peripancreatic vessels. It incorporates interventions at both image and feature levels to guide the model to capture causal information by enforcing consistency across datasets, thus enhancing the generalization performance. Specifically, firstly, a contrast-driven image intervention strategy is proposed to construct image-level interventions by generating images with various contrast-related appearances and seeking invariant causal features. Secondly, the feature intervention strategy is designed, where various patterns of feature bias across different centers are simulated to pursue invariant prediction. The proposed model achieved high DSC scores (79.69%, 82.62%, and 83.10%) for the three vessels on a cross-validation set containing 134 cases. Its generalizability was further confirmed on three independent test sets of 233 cases. Overall, the proposed method provides an accurate and generalizable segmentation model for peripancreatic vessels and offers a promising paradigm for increasing the generalizability of segmentation models from a causality perspective. Our source codes will be released at https://github.com/SJTUBME-QianLab/PC_VesselSeg.

A causality-inspired generalized model for automated pancreatic cancer diagnosis.

Pancreatic cancer (PC) is a severely malignant cancer variant with high mortality. Since PC has no obvious symptoms, most PC patients are belatedly diagnosed at advanced disease stages. Recently, artificial intelligence (AI) approaches have demonstrated promising prospects for early diagnosis of pancreatic cancer. However, certain non-causal factors (such as intensity and texture appearance variations, also called confounders) tend to induce spurious correlation with PC diagnosis. This undermines the generalization performance and the clinical applicability of the AI-based PC diagnosis approaches. Therefore, we propose a causal intervention based automated method for pancreatic cancer diagnosis with contrast-enhanced computerized tomography (CT) images, where a confounding effects reduction scheme is developed for alleviating spurious correlations to achieve unbiased learning, thereby improving the generalization performance. Specifically, a continuous image generation strategy was developed to simulate wide variations of intensity differences caused by imaging heterogeneities, where Monte Carlo sampling is added to further enhance the continuity of simulated images. Then, to enhance the pancreatic texture variability, a texture diversification method was introduced in conjunction with gradient-based data augmentation. Finally, a causal intervention strategy was proposed to alleviate the adverse confounding effects by decoupling the causal and non-causal factors and combining them randomly. Extensive experiments showed remarkable diagnosis performance on a cross-validation dataset. Also, promising generalization performance with an average accuracy of 0.87 was attained on three independent test sets of a total of 782 subjects. Therefore, the proposed method shows high clinical feasibility and applicability for pancreatic cancer diagnosis.

CHRNA6 RNA In Situ Hybridization Is a Useful Tool for the Diagnosis of Extraskeletal Myxoid Chondrosarcoma.

Extraskeletal myxoid chondrosarcoma (EMC) is an uncommon mesenchymal neoplasm characteristically composed of uniform-appearing round to spindle-shaped cells with eosinophilic cytoplasm and abundant myxoid extracellular matrix. Although the majority of cases harbor a pathognomonic t(9;22) translocation that fuses EWSR1 with the orphan nuclear receptor NR4A3, there are less common variants that partner NR4A3 with TAF15, TCF12, or TFG. By immunohistochemistry, EMC has features of both cartilaginous and neuroendocrine differentiation, as evidenced by inconsistent expression of S100 protein and synaptophysin or INSM1, respectively, in a subset of cases. Given the limitations of available immunohistochemical stains for the diagnosis of EMC, we analyzed genome-wide gene expression microarray data to identify candidate biomarkers based on differential expression in EMC in comparison with other mesenchymal neoplasms. This analysis pointed to CHRNA6 as the gene with the highest relative expression in EMC (96-fold; P = 8.2 × 10-26) and the only gene with >50-fold increased expression in EMC compared with other tumors. Using RNA chromogenic in situ hybridization, we observed strong and diffuse expression of CHRNA6 in 25 cases of EMC, including both EWSR1-rearranged and TAF15-rearranged variants. All examined cases of histologic mimics were negative for CHRNA6 overexpression; however, limited CHRNA6 expression, not reaching a threshold of >5 puncta or 1 aggregate of chromogen in >25% of cells, was observed in 69 of 685 mimics (10.1%), spanning an array of mesenchymal tumors. Taken together, these findings suggest that, with careful interpretation and the use of appropriate thresholds, CHRNA6 RNA chromogenic in situ hybridization is a potentially useful ancillary histologic tool for the diagnosis of EMC.

Prediction of intravenous immunoglobulin retreatment in children with Kawasaki disease using models combining lymphocyte subset and cytokine profile in an East Asian cohort.

Objectives: For children with Kawasaki disease (KD) at high risk of developing coronary artery lesions and requiring retreatment with intravenous immunoglobulin (IVIG), the availability of accurate prediction models remains limited because of inconsistent variables and unsatisfactory prediction results. We aimed to construct models to predict patient's probability of IVIG retreatment combining children's individual inflammatory characteristics. Methods: Clinical manifestations and laboratory examinations of 266 children with KD were retrospectively analysed to build a development cohort data set (DC) and a validation cohort data set (VC). In the DC, binary logistic regression analyses were performed using R language. Nomograms and receiver operating curves were plotted. The concordance index (C index), net reclassification index, integrated discrimination improvement index and confusion matrix were applied to evaluate and validate the models. Results: Models_5V and _9V were established. Both contained variables including the percentages of CD8+ T cells, CD4+ T cells, CD3+ T cells, levels of interleukin (IL)-2R and CRP. Model_9V additionally included variables for IL-6, TNF-α, NT-proBNP and sex, with a C index of 0.86 (95% CI 0.79-0.92). When model_9V was compared with model_5V, the NRI and IDI were 0.15 (95% CI 0.01-0.30, P < 0.01) and 0.07 (95% CI 0.02-0.12, P < 0.01). In the VC, the sensitivity, specificity and precision of model_9V were 1, 0.875 and 0.667, while those of model_5V were 0.833, 0.875 and 0.625. Conclusion: Model_9V combined cytokine profiles and lymphocyte subsets with clinical characteristics and was superior to model_5V achieving satisfactory predictive power and providing a novel strategy early to identify patients who needed IVIG retreatment.

A Causality-Aware Graph Convolutional Network Framework for Rigidity Assessment in Parkinsonians.

Rigidity is one of the common motor disorders in Parkinson's disease (PD), which lead to life quality deterioration. The widely-used rating-scale-based approach for rigidity assessment still depends on the availability of experienced neurologists and is limited by rating subjectivity. Given the recent successful applications of quantitative susceptibility mapping (QSM) in auxiliary PD diagnosis, automated assessment of PD rigidity can be essentially achieved through QSM analysis. However, a major challenge is the performance instability due to the confounding factors (e.g., noise and distribution shift) which conceal the truly-causal features. Therefore, we propose a causality-aware graph convolutional network (GCN) framework, where causal feature selection is combined with causal invariance to ensure that causality-informed model decisions are reached. Firstly, a GCN model that integrates causal feature selection is systematically constructed at three graph levels: node, structure, and representation. In this model, a causal diagram is learned to extract a subgraph with truly-causal information. Secondly, a non-causal perturbation strategy is developed along with an invariance constraint to ensure the stability of the assessment results under different distributions, and thus avoid spurious correlations caused by distribution shifts. The superiority of the proposed method is shown by extensive experiments and the clinical value is revealed by the direct relevance of selected brain regions to rigidity in PD. Besides, its extensibility is verified on other two tasks: PD bradykinesia and mental state for Alzheimer's disease. Overall, we provide a clinically-potential tool for automated and stable assessment of PD rigidity. Our source code will be available at https://github.com/SJTUBME-QianLab/Causality-Aware-Rigidity.

Rapid Prototyping of Thermoplastic Microfluidic 3D Cell Culture Devices by Creating Regional Hydrophilicity Discrepancy.

Microfluidic 3D cell culture devices that enable the recapitulation of key aspects of organ structures and functions in vivo represent a promising preclinical platform to improve translational success during drug discovery. Essential to these engineered devices is the spatial patterning of cells from different tissue types within a confined microenvironment. Traditional fabrication strategies lack the scalability, cost-effectiveness, and rapid prototyping capabilities required for industrial applications, especially for processes involving thermoplastic materials. Here, an approach to pattern fluid guides inside microchannels is introduced by establishing differential hydrophilicity using pressure-sensitive adhesives as masks and a subsequent selective coating with a biocompatible polymer. Optimal coating conditions are identified using polyvinylpyrrolidone, which resulted in rapid and consistent hydrogel flow in both the open-chip prototype and the fully bonded device containing additional features for medium perfusion. The suitability of the device for dynamic 3D cell culture is tested by growing human hepatocytes in the device under controlled fluid flow for a 14-day period. Additionally, the study demonstrated the potential of using the device for pharmaceutical high-throughput screening applications, such as predicting drug-induced liver injury. The approach offers a facile strategy of rapid prototyping thermoplastic microfluidic organ chips with varying geometries, microstructures, and substrate materials.

Pitfalls in soft tissue cytopathology.

Fine needle aspiration biopsy (FNAB) is a diagnostic modality for the evaluation of suspicious soft tissue masses. Despite its reasonable sensitivity, specificity and positive predictive value in differentiating benign from malignant neoplasms, the exact subtyping of the primary soft tissue tumours can be challenging. Certain tumours constitute "pitfalls" and add to the diagnostic challenge. This review provides a detailed account of the diagnostic challenges in soft tissue cytopathology, including pitfalls and, more importantly, the ways to overcome these challenges by integrating clinical details, key cytomorphological features and judicious application of ancillary techniques.

A Causality-Driven Graph Convolutional Network for Postural Abnormality Diagnosis in Parkinsonians.

Abnormal posture is a common movement disorder in the progress of Parkinson's disease (PD), and this abnormality can increase the risk of falls or even disabilities. The conventional assessment approach depends on the judgment of well-trained experts via canonical scales. However, this approach requires extensive clinical expertise and is highly subjective. Considering the potential of quantitative susceptibility mapping (QSM) in PD diagnosis, this study explored the QSM-based method for the automated classification between PD patients with and without postural abnormalities. Nevertheless, a major challenge is that unstable non-causal features typically lead to less reliable performance. Therefore, we propose a causality-driven graph-convolutional-network framework based on multi-instance learning, where performance stability is enhanced through the invariant prediction principle and causal interventions. Specifically, we adopt an intervention strategy that combines a non-causal intervenor with causal prediction. A stability constraint is proposed to ensure robust integrated prediction under different interventions. Moreover, an intra-class homogeneity constraint is enforced for each individually-learned causality scoring module to promote the extraction of group-level general features, and hence achieve a balance between subject-specific and group-level features. The proposed method demonstrated promising performance through extensive experiments on a real clinical dataset. Also, the features extracted by our method coincide with those reported in previous medical studies on PD posture abnormalities. In general, our work provides a clinically-valuable approach for automated, objective, and reliable diagnosis of postural abnormalities in Parkinsonians. Our source code is publicly available at https://github.com/SJTUBME-QianLab/CausalGCN-PDPA.

Generalizable Pancreas Segmentation via a Dual Self-Supervised Learning Framework.

Recently, numerous pancreas segmentation methods have achieved promising performance on local single-source datasets. However, these methods don't adequately account for generalizability issues, and hence typically show limited performance and low stability on test data from other sources. Considering the limited availability of distinct data sources, we seek to improve the generalization performance of a pancreas segmentation model trained with a single-source dataset, i.e., the single-source generalization task. In particular, we propose a dual self-supervised learning model that incorporates both global and local anatomical contexts. Our model aims to fully exploit the anatomical features of the intra-pancreatic and extra-pancreatic regions, and hence enhance the characterization of the high-uncertainty regions for more robust generalization. Specifically, we first construct a global-feature contrastive self-supervised learning module that is guided by the pancreatic spatial structure. This module obtains complete and consistent pancreatic features through promoting intra-class cohesion, and also extracts more discriminative features for differentiating between pancreatic and non-pancreatic tissues through maximizing inter-class separation. It mitigates the influence of surrounding tissue on the segmentation outcomes in high-uncertainty regions. Subsequently, a local-image-restoration self-supervised learning module is introduced to further enhance the characterization of the high-uncertainty regions. In this module, informative anatomical contexts are actually learned to recover randomly-corrupted appearance patterns in those regions. The effectiveness of our method is demonstrated with state-of-the-art performance and comprehensive ablation analysis on three pancreas datasets (467 cases). The results demonstrate a great potential in providing a stable support for the diagnosis and treatment of pancreatic diseases.

A deep-learning radiomics-based lymph node metastasis predictive model for pancreatic cancer: a diagnostic study.

OBJECTIVES: Preoperative lymph node (LN) status is essential in formulating the treatment strategy among pancreatic cancer patients. However, it is still challenging to evaluate the preoperative LN status precisely now. METHODS: A multivariate model was established based on the multiview-guided two-stream convolution network (MTCN) radiomics algorithms, which focused on primary tumor and peri-tumor features. Regarding discriminative ability, survival fitting, and model accuracy, different models were compared. RESULTS: Three hundred and sixty-three pancreatic cancer patients were divided in to train and test cohorts by 7:3. The modified MTCN (MTCN+) model was established based on age, CA125, MTCN scores, and radiologist judgement. The MTCN+ model outperformed the MTCN model and the artificial model in discriminative ability and model accuracy. [Train cohort area under curve (AUC): 0.823 vs. 0.793 vs. 0.592; train cohort accuracy (ACC): 76.1 vs. 74.4 vs. 56.7%; test cohort AUC: 0.815 vs. 0.749 vs. 0.640; test cohort ACC: 76.1 vs. 70.6 vs. 63.3%; external validation AUC: 0.854 vs. 0.792 vs. 0.542; external validation ACC: 71.4 vs. 67.9 vs. 53.5%]. The survivorship curves fitted well between actual LN status and predicted LN status regarding disease free survival and overall survival. Nevertheless, the MTCN+ model performed poorly in assessing the LN metastatic burden among the LN positive population. Notably, among the patients with small primary tumors, the MTCN+ model performed steadily as well (AUC: 0.823, ACC: 79.5%). CONCLUSIONS: A novel MTCN+ preoperative LN status predictive model was established and outperformed the artificial judgement and deep-learning radiomics judgement. Around 40% misdiagnosed patients judged by radiologists could be corrected. And the model could help precisely predict the survival prognosis.

Causality-Driven Graph Neural Network for Early Diagnosis of Pancreatic Cancer in Non-Contrast Computerized Tomography.

Pancreatic cancer is the emperor of all cancer maladies, mainly because there are no characteristic symptoms in the early stages, resulting in the absence of effective screening and early diagnosis methods in clinical practice. Non-contrast computerized tomography (CT) is widely used in routine check-ups and clinical examinations. Therefore, based on the accessibility of non-contrast CT, an automated early diagnosismethod for pancreatic cancer is proposed. Among this, we develop a novel causalitydriven graph neural network to solve the challenges of stability and generalization of early diagnosis, that is, the proposed method achieves stable performance for datasets from different hospitals, which highlights its clinical significance. Specifically, a multiple-instance-learning framework is designed to extract fine-grained pancreatic tumor features. Afterwards, to ensure the integrity and stability of the tumor features, we construct an adaptivemetric graph neural network that effectively encodes prior relationships of spatial proximity and feature similarity for multiple instances, and hence adaptively fuses the tumor features. Besides, a causal contrastivemechanism is developed to decouple the causality-driven and non-causal components of the discriminative features, suppress the non-causal ones, and hence improve the model stability and generalization. Extensive experiments demonstrated that the proposed method achieved the promising early diagnosis performance, and its stability and generalizability were independently verified on amulti-center dataset. Thus, the proposed method provides a valuable clinical tool for the early diagnosis of pancreatic cancer. Our source codes will be released at https://github.com/SJTUBME-QianLab/ CGNN-PC-Early-Diagnosis.

Development and validation of an MRI-radiomics nomogram for the prognosis of pancreatic ductal adenocarcinoma.

Objective: To develop and validate an MRI-radiomics nomogram for the prognosis of pancreatic ductal adenocarcinoma (PDAC). Background: "Radiomics" enables the investigation of huge amounts of radiological features in parallel by extracting high-throughput imaging data. MRI provides better tissue contrast with no ionizing radiation for PDAC. Methods: There were 78 PDAC patients enrolled in this study. In total, there were 386 radiomics features extracted from MRI scan, which were screened by the least absolute shrinkage and selection operator algorithm to develop a risk score. Cox multivariate regression analysis was applied to develop the radiomics-based nomogram. The performance was assessed by discrimination and calibration. Results: The radiomics-based risk-score was significantly associated with PDAC overall survival (OS) (P < 0.05). With respect to survival prediction, integrating the risk score, clinical data and TNM information into the nomogram exhibited better performance than the TNM staging system, radiomics model and clinical model. In addition, the nomogram showed fine discrimination and calibration. Conclusions: The radiomics nomogram incorporating the radiomics data, clinical data and TNM information exhibited precise survival prediction for PDAC, which may help accelerate personalized precision treatment. Clinical trial registration: clinicaltrials.gov, identifier NCT05313854.

Generalized pancreatic cancer diagnosis via multiple instance learning and anatomically-guided shape normalization.

Pancreatic cancer is a highly malignant cancer type with a high mortality rate. As no obvious symptoms are associated with this cancer type, most of the diagnoses are made when the patients are already in a late stage. In this work, we propose an automated method for effective early diagnosis of pancreatic cancer based on multiple instance learning with contrast-enhanced CT images. In this method, diagnosis stability and generalizability were improved through shape normalization based on anatomical structures as well as instance-level contrastive learning. Specifically, anatomically-guided shape normalization were developed to reconstruct the pancreatic regions of interest by spatial transformations, account for larger tumor parts in these regions, and hence enhance the extraction of pancreatic features. Moreover, instance-level contrastive learning was employed to aggregate different types of tumor features within the multiple instance learning framework. This learning approach can maintain the tumor feature integrity and enhance the diagnosis stability. Finally, a balance-adjustment strategy was designed to alleviate the class imbalance problem caused by the scarcity of tumor samples. Extensive experimental results demonstrated remarkable performance of our method when conducted cross-validation on an in-house dataset with 310 patients and independent test on two unseen datasets (a private test set with 316 and a publicly-available test set with 281). The proposed strategies also led to significant improvements in generalizability. Besides, the clinical significance of the proposed method was further verified through two independent test results in which tumors smaller than 2 cm in diameter were identified at accuracies of 80.9% and 90.1%, respectively. Overall, our method provides a potentially successful tool for early diagnosis of pancreatic cancer. Our source codes will be released at https://github.com/SJTUBME-QianLab/MIL_PAdiagnosis.

A dual-transformation with contrastive learning framework for lymph node metastasis prediction in pancreatic cancer.

Pancreatic cancer is a malignant tumor, and its high recurrence rate after surgery is related to the lymph node metastasis status. In clinical practice, a preoperative imaging prediction method is necessary for prognosis assessment and treatment decision; however, there are two major challenges: insufficient data and difficulty in discriminative feature extraction. This paper proposed a deep learning model to predict lymph node metastasis in pancreatic cancer using multiphase CT, where a dual-transformation with contrastive learning framework is developed to overcome the challenges in fine-grained prediction with small sample sizes. Specifically, we designed a novel dynamic surface projection method to transform 3D data into 2D images for effectively using the 3D information, preserving the spatial correlation of the original texture information and reducing computational resources. Then, this dynamic surface projection was combined with the spiral transformation to establish a dual-transformation method for enhancing the diversity and complementarity of the dataset. A dual-transformation-based data augmentation method was also developed to produce numerous 2D-transformed images to alleviate the effect of insufficient samples. Finally, the dual-transformation-guided contrastive learning scheme based on intra-space-transformation consistency and inter-class specificity was designed to mine additional supervised information, thereby extracting more discriminative features. Extensive experiments have shown the promising performance of the proposed model for predicting lymph node metastasis in pancreatic cancer. Our dual-transformation with contrastive learning scheme was further confirmed on an external public dataset, representing a potential paradigm for the fine-grained classification of oncological images with small sample sizes. The code will be released at https://github.com/SJTUBME-QianLab/Dual-transformation.

Group-shrinkage feature selection with a spatial network for mining DNA methylation data.

Identifying disease-related biomarkers from high-dimensional DNA methylation data helps in reducing early screening costs and inferring pathogenesis mechanisms. Good discovery results have been achieved through spatial correlation methods of methylation sites, group-based regularization, and network constraints. However, these methods still have some key limitations as they cannot exclude isolated differential sites and only consider adjacent site ordering. Therefore, we propose a group-shrinkage feature selection algorithm to encourage the selection of clustered sites and discourage the selection of isolated differential sites. Specifically, a network-guided group-shrinkage strategy is developed to penalize weakly-correlated isolated methylation sites through a network structure constraint. The spatial network is constructed based on spatial correlation information of DNA methylation sites, where this information accounts for the uneven site distribution. The experimental simulations and applications demonstrated that the proposed method outperforms the advanced regularization methods, especially in rejecting isolated methylation sites; hence this study provides an efficient and clinical-valuable method for biomarker candidate discovery in DNA methylation data. Additionally, the proposed method exhibits enhanced reliability due to introducing biological prior knowledge into a regularization-based feature selection framework and could promote more research in the integration between biological prior knowledge and classical feature selection methods, thus facilitating their clinical application. Our source codes will be released at https://github.com/SJTUBME-QianLab/Group-shrinkage-Spatial-Network once this manuscript is accepted for publication.

Diagnosis of Glioblastoma Multiforme Progression via Interpretable Structure-Constrained Graph Neural Networks.

Glioblastoma multiforme (GBM) is the most common type of brain tumors with high recurrence and mortality rates. After chemotherapy treatment, GBM patients still show a high rate of differentiating pseudoprogression (PsP), which is often confused as true tumor progression (TTP) due to high phenotypical similarities. Thus, it is crucial to construct an automated diagnosis model for differentiating between these two types of glioma progression. However, attaining this goal is impeded by the limited data availability and the high demand for interpretability in clinical settings. In this work, we propose an interpretable structure-constrained graph neural network (ISGNN) with enhanced features to automatically discriminate between PsP and TTP. This network employs a metric-based meta-learning strategy to aggregate class-specific graph nodes, focus on meta-tasks associated with various small graphs, thus improving the classification performance on small-scale datasets. Specifically, a node feature enhancement module is proposed to account for the relative importance of node features and enhance their distinguishability through inductive learning. A graph generation constraint module enables learning reasonable graph structures to improve the efficiency of information diffusion while avoiding propagation errors. Furthermore, model interpretability can be naturally enhanced based on the learned node features and graph structures that are closely related to the classification results. Comprehensive experimental evaluation of our method demonstrated excellent interpretable results in the diagnosis of glioma progression. In general, our work provides a novel systematic GNN approach for dealing with data scarcity and enhancing decision interpretability. Our source codes will be released at https://github.com/SJTUBME-QianLab/GBM-GNN.

Generalizable Pancreas Segmentation Modeling in CT Imaging via Meta-Learning and Latent-Space Feature Flow Generation.

Accurate pancreas segmentation is highly crucial for diagnosing and treating pancreatic diseases. Although CNN has demonstrated promising outcomes, the performance on unseen data can be significantly compromised by the wide appearance-style variations induced by different imaging factors. Thus, we propose a generalizable pancreas segmentation model based on a meta-learning strategy and latent-space feature flow generation method. Our approach enhances the generalizability by systematically reducing the interference from the cluttered background and appearance-style discrepancies through a coarse-to-fine workflow. Specifically, the integrity-preserving coarse segmentation module is designed to adaptively balance the pancreas coverage and segmentation accuracy with the meta-learning strategy for filtering out background clutter. It also enhances the generalization of the coarse model to reasonably-accurate ROIs thereby promoting the stability of fine segmentation. Subsequently, the appearance-style feature flow generation method is developed to generate a series of progressively-varying style-related intermediate representations between two latent spaces. This feature flow effectively models the distribution variations caused by appearance-style discrepancies, and thus enhances the adaptability of the fine model. Our method achieves superior performance on three pancreas datasets and outperforms state-of-the-art generalization methods. Besides, it can be easily integrated into other workflows, leading to a potential paradigm for enhancing generalization performance.

A tree-structure-guided graph convolutional network with contrastive learning for the assessment of parkinsonian hand movements.

Bradykinesia is one of the core motor symptoms of Parkinson's disease (PD). Neurologists typically perform face-to-face bradykinesia assessment in PD patients according to the Movement Disorder Society-sponsored revision of the Unified Parkinson's Disease Rating Scale (MDS-UPDRS). As this human-expert assessment lacks objectivity and consistency, an automated and objective assessment scheme for bradykinesia is critically needed. In this paper, we propose a tree-structure-guided graph convolutional network with contrastive learning scheme to solve the challenge of difficulty in fine-grained feature extraction and insufficient model stability, finally achieving the video-based automated assessment of Parkinsonian hand movements, which represent a vital MDS-UPDRS component for examining upper-limb bradykinesia. Specifically, a tri-directional skeleton tree scheme is proposed to achieve effective fine-grained modeling of spatial hand dependencies. In this scheme, hand skeletons are extracted from videos, and then the spatial structures of these skeletons are constructed through depth-first tree traversal. Afterwards, a tree max-pooling module is employed to establish remote exchange between outer and inner nodes, hierarchically gather the most salient motion features, and hence achieve fine-grained mining. Finally, a group-sparsity-induced momentum contrast is also developed to learn similar motion patterns under different interference through contrastive learning. This can promote stable learning of discriminative spatial-temporal features with invariant motion semantics. Comprehensive experiments on a large clinical video dataset reveal that our method achieves competitive results, and outperforms other sensor-based and RGB-depth methods. The proposed method leads to accurate assessment of PD bradykinesia through videos collected by low-cost consumer cameras of limited capabilities. Hence, our work provides a convenient tool for PD telemedicine applications with modest hardware requirements.