Vision Transformer-based Multimodal Feature Fusion Network for Lymphoma Segmentation on PET/CT Images

Background: Diffuse large B-cell lymphoma (DLBCL) segmentation is a challenge in medical image analysis. Traditional segmentation methods for lymphoma struggle with the complex patterns and the presence of DLBCL lesions. Objective: We aim to develop an accurate method for lymphoma segmentation with 18F-Fluorodeoxyglucose positron emission tomography (PET) and computed tomography (CT) images. Methods: Our lymphoma segmentation approach combines a vision transformer with dual encoders, adeptly fusing PET and CT data via multimodal cross-attention fusion (MMCAF) module. In this study, PET and CT data from 165 DLBCL patients were analyzed. A 5-fold cross-validation was employed to evaluate the performance and generalization ability of our method. Ground truths were annotated by experienced nuclear medicine experts. We calculated the total metabolic tumor volume (TMTV) and performed a statistical analysis on our results. Results: The proposed method exhibited accurate performance in DLBCL lesion segmentation, achieving a Dice similarity coefficient of 0.9173$\pm$0.0071, a Hausdorff distance of 2.71$\pm$0.25mm, a sensitivity of 0.9462$\pm$0.0223, and a specificity of 0.9986$\pm$0.0008. Additionally, a Pearson correlation coefficient of 0.9030$\pm$0.0179 and an R-square of 0.8586$\pm$0.0173 were observed in TMTV when measured on manual annotation compared to our segmentation results. Conclusion: This study highlights the advantages of MMCAF and vision transformer for lymphoma segmentation using PET and CT, offering great promise for computer-aided lymphoma diagnosis and treatment.

A Robust Deep Learning Method with Uncertainty Estimation for the Pathological Classification of Renal Cell Carcinoma based on CT Images

Objectives To develop and validate a deep learning-based diagnostic model incorporating uncertainty estimation so as to facilitate radiologists in the preoperative differentiation of the pathological subtypes of renal cell carcinoma (RCC) based on CT images. Methods Data from 668 consecutive patients, pathologically proven RCC, were retrospectively collected from Center 1. By using five-fold cross-validation, a deep learning model incorporating uncertainty estimation was developed to classify RCC subtypes into clear cell RCC (ccRCC), papillary RCC (pRCC), and chromophobe RCC (chRCC). An external validation set of 78 patients from Center 2 further evaluated the model's performance. Results In the five-fold cross-validation, the model's area under the receiver operating characteristic curve (AUC) for the classification of ccRCC, pRCC, and chRCC was 0.868 (95% CI: 0.826-0.923), 0.846 (95% CI: 0.812-0.886), and 0.839 (95% CI: 0.802-0.88), respectively. In the external validation set, the AUCs were 0.856 (95% CI: 0.838-0.882), 0.787 (95% CI: 0.757-0.818), and 0.793 (95% CI: 0.758-0.831) for ccRCC, pRCC, and chRCC, respectively. Conclusions The developed deep learning model demonstrated robust performance in predicting the pathological subtypes of RCC, while the incorporated uncertainty emphasized the importance of understanding model confidence, which is crucial for assisting clinical decision-making for patients with renal tumors. Clinical relevance statement Our deep learning approach, integrated with uncertainty estimation, offers clinicians a dual advantage: accurate RCC subtype predictions complemented by diagnostic confidence references, promoting informed decision-making for patients with RCC.

MLA-BIN: Model-level Attention and Batch-instance Style Normalization for Domain Generalization of Federated Learning on Medical Image Segmentation

The privacy protection mechanism of federated learning (FL) offers an effective solution for cross-center medical collaboration and data sharing. In multi-site medical image segmentation, each medical site serves as a client of FL, and its data naturally forms a domain. FL supplies the possibility to improve the performance of seen domains model. However, there is a problem of domain generalization (DG) in the actual de-ployment, that is, the performance of the model trained by FL in unseen domains will decrease. Hence, MLA-BIN is proposed to solve the DG of FL in this study. Specifically, the model-level attention module (MLA) and batch-instance style normalization (BIN) block were designed. The MLA represents the unseen domain as a linear combination of seen domain models. The atten-tion mechanism is introduced for the weighting coefficient to obtain the optimal coefficient ac-cording to the similarity of inter-domain data features. MLA enables the global model to gen-eralize to unseen domain. In the BIN block, batch normalization (BN) and instance normalization (IN) are combined to perform the shallow layers of the segmentation network for style normali-zation, solving the influence of inter-domain image style differences on DG. The extensive experimental results of two medical image seg-mentation tasks demonstrate that the proposed MLA-BIN outperforms state-of-the-art methods.

A novel method using machine learning to integrate features from lung and epicardial adipose tissue for detecting the severity of COVID-19 infection

Objectives: To investigate the value of radiomics features of epicardial adipose tissue (EAT) combined with lung for detecting the severity of Coronavirus Disease 2019 (COVID-19) infection. Methods: The retrospective study included data from 515 COVID-19 patients (Cohort1: 415, cohort2: 100) from the two centers between January 2020 and July 2020. A deep learning method was developed to extract the myocardium and visceral pericardium from chest CTs, and then a threshold was applied for automatic EAT extraction. Lung segmentation was achieved according to a published method. Radiomics features of both EAT and lung were extracted for the severity prediction. In a derivation cohort (290, cohort1), univariate analysis and Pearson correlation analysis were used to identify predictors of the severity of COVID-19. A generalized linear regression model for detecting the severity of COVID-19 was built in a derivation cohort and evaluated in internal (125, cohort1) and external (100, cohort2) validation cohorts. Results: For EAT extraction, the Dice similarity coefficients (DSC) of the two centers were 0.972 (0.011) and 0.968 (0.005), respectively. For severity detection, the AUC, net reclassification improvement (NRI), and integrated discrimination improvement (IDI) of the model with radiomics features of both lung and EAT increased by 0.09 (p<0.001), 22.4%, and 17.0%, respectively, compared with the model with lung radiomics features, in the internal validation cohort. The AUC, NRI, and IDI increased by 0.04 (p<0.001), 11.1%, and 8.0%, respectively, in the external validation cohort. Conclusion: Radiomics features of EAT combined with lung have incremental value in detecting the severity of COVID-19.

A new method incorporating deep learning with shape priors for left ventricular segmentation in myocardial perfusion SPECT images

Background: The assessment of left ventricular (LV) function by myocardial perfusion SPECT (MPS) relies on accurate myocardial segmentation. The purpose of this paper is to develop and validate a new method incorporating deep learning with shape priors to accurately extract the LV myocardium for automatic measurement of LV functional parameters. Methods: A segmentation architecture that integrates a three-dimensional (3D) V-Net with a shape deformation module was developed. Using the shape priors generated by a dynamic programming (DP) algorithm, the model output was then constrained and guided during the model training for quick convergence and improved performance. A stratified 5-fold cross-validation was used to train and validate our models. Results: Results of our proposed method agree well with those from the ground truth. Our proposed model achieved a Dice similarity coefficient (DSC) of 0.9573(0.0244), 0.9821(0.0137), and 0.9903(0.0041), a Hausdorff distances (HD) of 6.7529(2.7334) mm, 7.2507(3.1952) mm, and 7.6121(3.0134) mm in extracting the endocardium, myocardium, and epicardium, respectively. Conclusion: Our proposed method achieved a high accuracy in extracting LV myocardial contours and assessing LV function.

Investigating Transfer Learning in Multilingual Pre-trained Language Models through Chinese Natural Language Inference

Multilingual transformers (XLM, mT5) have been shown to have remarkable transfer skills in zero-shot settings. Most transfer studies, however, rely on automatically translated resources (XNLI, XQuAD), making it hard to discern the particular linguistic knowledge that is being transferred, and the role of expert annotated monolingual datasets when developing task-specific models. We investigate the cross-lingual transfer abilities of XLM-R for Chinese and English natural language inference (NLI), with a focus on the recent large-scale Chinese dataset OCNLI. To better understand linguistic transfer, we created 4 categories of challenge and adversarial tasks (totaling 17 new datasets) for Chinese that build on several well-known resources for English (e.g., HANS, NLI stress-tests). We find that cross-lingual models trained on English NLI do transfer well across our Chinese tasks (e.g., in 3/4 of our challenge categories, they perform as well/better than the best monolingual models, even on 3/5 uniquely Chinese linguistic phenomena such as idioms, pro drop). These results, however, come with important caveats: cross-lingual models often perform best when trained on a mixture of English and high-quality monolingual NLI data (OCNLI), and are often hindered by automatically translated resources (XNLI-zh). For many phenomena, all models continue to struggle, highlighting the need for our new diagnostics to help benchmark Chinese and cross-lingual models. All new datasets/code are released at https://github.com/huhailinguist/ChineseNLIProbing.

CLUE: A Chinese Language Understanding Evaluation Benchmark

We introduce CLUE, a Chinese Language Understanding Evaluation benchmark. It contains eight different tasks, including single-sentence classification, sentence pair classification, and machine reading comprehension. We evaluate CLUE on a number of existing full-network pre-trained models for Chinese. We also include a small hand-crafted diagnostic test set designed to probe specific linguistic phenomena using different models, some of which are unique to Chinese. Along with CLUE, we release a large clean crawled raw text corpus that can be used for model pre-training. We release CLUE, baselines and pre-training dataset on Github.