Alterations in structural integrity of superior longitudinal fasciculus III associated with cognitive performance in cerebral small vessel disease.

BACKGROUND: This study aimed to investigate the alterations in structural integrity of superior longitudinal fasciculus subcomponents with increasing white matter hyperintensity severity as well as the relationship to cognitive performance in cerebral small vessel disease. METHODS: 110 cerebral small vessel disease study participants with white matter hyperintensities were recruited. According to Fazekas grade scale, white matter hyperintensities of each subject were graded. All subjects were divided into two groups. The probabilistic fiber tracking method was used for analyzing microstructure characteristics of superior longitudinal fasciculus subcomponents. RESULTS: Probabilistic fiber tracking results showed that mean diffusion, radial diffusion, and axial diffusion values of the left arcuate fasciculus as well as the mean diffusion value of the right arcuate fasciculus and left superior longitudinal fasciculus III in high white matter hyperintensities rating group were significantly higher than those in low white matter hyperintensities rating group (p < 0.05). The mean diffusion value of the left superior longitudinal fasciculus III was negatively related to the Montreal Cognitive Assessment score of study participants (p < 0.05). CONCLUSIONS: The structural integrity injury of bilateral arcuate fasciculus and left superior longitudinal fasciculus III is more severe with the aggravation of white matter hyperintensities. The structural integrity injury of the left superior longitudinal fasciculus III correlates to cognitive impairment in cerebral small vessel disease.

Development and validation of a radiomics-based prediction pipeline for the response to stereotactic radiosurgery therapy in brain metastases.

OBJECTIVES: The first treatment strategy for brain metastases (BM) plays a pivotal role in the prognosis of patients. Among all strategies, stereotactic radiosurgery (SRS) is considered a promising therapy method. Therefore, we developed and validated a radiomics-based prediction pipeline to prospectively identify BM patients who are insensitive to SRS therapy, especially those who are at potential risk of progressive disease. METHODS: A total of 337 BM patients (277, 30, and 30 in the training set, internal validation set, and external validation set, respectively) were enrolled in the study. 19,377 radiomics features (3 masks × 3 MRI sequences × 2153 features) extracted from 9 ROIs were filtered through LASSO and Max-Relevance and Min-Redundancy (mRMR) algorithms. The selected radiomics features were combined with 4 clinical features to construct a two-stage cascaded model for the prediction of BM patients' response to SRS therapy using SVM and an ensemble learning classifier. The performance of the model was evaluated by its accuracy, specificity, sensitivity, and AUC curve. RESULTS: Radiomics features were integrated with the clinical features of patients in our optimal model, which showed excellent discriminative performance in the training set (AUC: 0.95, 95% CI: 0.88-0.98). The model was also verified in the internal validation set and external validation set (AUC 0.93, 95% CI: 0.76-0.95 and AUC 0.90, 95% CI: 0.73-0.93, respectively). CONCLUSIONS: The proposed prediction pipeline could non-invasively predict the response to SRS therapy in patients with brain metastases thus assisting doctors to precisely designate individualized first treatment decisions. CLINICAL RELEVANCE STATEMENT: The proposed prediction pipeline combines the radiomics features of multi-modal MRI with clinical features to construct machine learning models that noninvasively predict the response of patients with brain metastases to stereotactic radiosurgery therapy, assisting neuro-oncologists to develop personalized first treatment plans. KEY POINTS: • The proposed prediction pipeline can non-invasively predict the response to SRS therapy. • The combination of multi-modality and multi-mask contributes significantly to the prediction. • The edema index also shows a certain predictive value.

Impaired functional network properties contribute to white matter hyperintensity related cognitive decline in patients with cerebral small vessel disease.

BACKGROUND: White matter hyperintensity (WMH) is one of the typical neuroimaging manifestations of cerebral small vessel disease (CSVD), and the WMH correlates closely to cognitive impairment (CI). CSVD patients with WMH own altered topological properties of brain functional network, which is a possible mechanism that leads to CI. This study aims to identify differences in the characteristics of some brain functional network among patients with different grades of WMH and estimates the correlations between these different brain functional network characteristics and cognitive assessment scores. METHODS: 110 CSVD patients underwent 3.0 T Magnetic resonance imaging scans and neuropsychological cognitive assessments. WMH of each participant was graded on the basis of Fazekas grade scale and was divided into two groups: (A) WMH score of 1-2 points (n = 64), (B) WMH score of 3-6 points (n = 46). Topological indexes of brain functional network were analyzed using graph-theoretical method. T-test and Mann-Whitney U test was used to compare the differences in topological properties of brain functional network between groups. Partial correlation analysis was applied to explore the relationship between different topological properties of brain functional networks and overall cognitive function. RESULTS: Patients with high WMH scores exhibited decreased clustering coefficient values, global and local network efficiency along with increased shortest path length on whole brain level as well as decreased nodal efficiency in some brain regions on nodal level (p < 0.05). Nodal efficiency in the left lingual gyrus was significantly positively correlated with patients' total Montreal Cognitive Assessment (MoCA) scores (p < 0.05). No significant difference was found between two groups on the aspect of total MoCA and Mini-mental State Examination (MMSE) scores (p > 0.05). CONCLUSION: Therefore, we come to conclusions that patients with high WMH scores showed less optimized small-world networks compared to patients with low WMH scores. Global and local network efficiency on the whole-brain level, as well as nodal efficiency in certain brain regions on the nodal level, can be viewed as markers to reflect the course of WMH.

Artificial intelligence using deep neural network learning for automatic location of the interscalene brachial plexus in ultrasound images.

BACKGROUND: Identifying the interscalene brachial plexus can be challenging during ultrasound-guided interscalene block. OBJECTIVE: We hypothesised that an algorithm based on deep learning could locate the interscalene brachial plexus in ultrasound images better than a nonexpert anaesthesiologist, thus possessing the potential to aid anaesthesiologists. DESIGN: Observational study. SETTING: A tertiary hospital in Shanghai, China. PATIENTS: Patients undergoing elective surgery. INTERVENTIONS: Ultrasound images at the interscalene level were collected from patients. Two independent image datasets were prepared to train and evaluate the deep learning model. Three senior anaesthesiologists who were experts in regional anaesthesia annotated the images. A deep convolutional neural network was developed, trained and optimised to locate the interscalene brachial plexus in the ultrasound images. Expert annotations on the datasets were regarded as an accurate baseline (ground truth). The test dataset was also annotated by five nonexpert anaesthesiologists. MAIN OUTCOME MEASURES: The primary outcome of the research was the distance between the lateral midpoints of the nerve sheath contours of the model predictions and ground truth. RESULTS: The data set was obtained from 1126 patients. The training dataset comprised 11 392 images from 1076 patients. The test dataset constituted 100 images from 50 patients. In the test dataset, the median [IQR] distance between the lateral midpoints of the nerve sheath contours of the model predictions and ground truth was 0.8 [0.4 to 2.9] mm: this was significantly shorter than that between nonexpert predictions and ground truth (3.4 mm [2.1 to 4.5] mm; P < 0.001). CONCLUSION: The proposed model was able to locate the interscalene brachial plexus in ultrasound images more accurately than nonexperts. TRIAL REGISTRATION: ClinicalTrials.gov (https://clinicaltrials.gov) identifier: NCT04183972.

Deep Learning for Automatic Differential Diagnosis of Primary Central Nervous System Lymphoma and Glioblastoma: Multi-Parametric Magnetic Resonance Imaging Based Convolutional Neural Network Model.

BACKGROUND: Differential diagnosis of primary central nervous system lymphoma (PCNSL) and glioblastoma (GBM) is useful to guide treatment strategies. PURPOSE: To investigate the use of a convolutional neural network (CNN) model for differentiation of PCNSL and GBM without tumor delineation. STUDY TYPE: Retrospective. POPULATION: A total of 289 patients with PCNSL (136) or GBM (153) were included, the average age of the cohort was 54 years, and there were 173 men and 116 women. FIELD STRENGTH/SEQUENCE: 3.0 T Axial contrast-enhanced T1 -weighted spin-echo inversion recovery sequence (CE-T1 WI), T2 -weighted fluid-attenuation inversion recovery sequence (FLAIR), and diffusion weighted imaging (DWI, b = 0 second/mm2 , 1000 seconds/mm2 ). ASSESSMENT: A single-parametric CNN model was built using CE-T1 WI, FLAIR, and the apparent diffusion coefficient (ADC) map derived from DWI, respectively. A decision-level fusion based multi-parametric CNN model (DF-CNN) was built by combining the predictions of single-parametric CNN models through logistic regression. An image-level fusion based multi-parametric CNN model (IF-CNN) was built using the integrated multi-parametric MR images. The radiomics models were developed. The diagnoses by three radiologists with 6 years (junior radiologist Y.Y.), 11 years (intermediate-level radiologist Y.T.), and 21 years (senior radiologist Y.L.) of experience were obtained. STATISTICAL ANALYSIS: The 5-fold cross validation was used for model evaluation. The Pearson's chi-squared test was used to compare the accuracies. U-test and Fisher's exact test were used to compare clinical characteristics. RESULTS: The CE-T1 WI, FLAIR, and ADC based single-parametric CNN model had accuracy of 0.884, 0.782, and 0.700, respectively. The DF-CNN model had an accuracy of 0.899 which was higher than the IF-CNN model (0.830, P = 0.021), but had no significant difference in accuracy compared to the radiomics model (0.865, P = 0.255), and the senior radiologist (0.906, P = 0.886). DATA CONCLUSION: A CNN model can differentiate PCNSL from GBM without tumor delineation, and comparable to the radiomics models and radiologists. LEVEL OF EVIDENCE: 4 TECHNICAL EFFICACY: Stage 2.

Pneumoconiosis computer aided diagnosis system based on X-rays and deep learning.

PURPOSE: The objective of this study is to construct a computer aided diagnosis system for normal people and pneumoconiosis using X-raysand deep learning algorithms. MATERIALS AND METHODS: 1760 anonymous digital X-ray images of real patients between January 2017 and June 2020 were collected for this experiment. In order to concentrate the feature extraction ability of the model more on the lung region and restrain the influence of external background factors, a two-stage pipeline from coarse to fine was established. First, the U-Net model was used to extract the lung regions on each sides of the collection images. Second, the ResNet-34 model with transfer learning strategy was implemented to learn the image features extracted in the lung region to achieve accurate classification of pneumoconiosis patients and normal people. RESULTS: Among the 1760 cases collected, the accuracy and the area under curve of the classification model were 92.46% and 89% respectively. CONCLUSION: The successful application of deep learning in the diagnosis of pneumoconiosis further demonstrates the potential of medical artificial intelligence and proves the effectiveness of our proposed algorithm. However, when we further classified pneumoconiosis patients and normal subjects into four categories, we found that the overall accuracy decreased to 70.1%. We will use the CT modality in future studies to provide more details of lung regions.

Rapid identification of COVID-19 severity in CT scans through classification of deep features.

BACKGROUND: Chest CT is used for the assessment of the severity of patients infected with novel coronavirus 2019 (COVID-19). We collected chest CT scans of 202 patients diagnosed with the COVID-19, and try to develop a rapid, accurate and automatic tool for severity screening follow-up therapeutic treatment. METHODS: A total of 729 2D axial plan slices with 246 severe cases and 483 non-severe cases were employed in this study. By taking the advantages of the pre-trained deep neural network, four pre-trained off-the-shelf deep models (Inception-V3, ResNet-50, ResNet-101, DenseNet-201) were exploited to extract the features from these CT scans. These features are then fed to multiple classifiers (linear discriminant, linear SVM, cubic SVM, KNN and Adaboost decision tree) to identify the severe and non-severe COVID-19 cases. Three validation strategies (holdout validation, tenfold cross-validation and leave-one-out) are employed to validate the feasibility of proposed pipelines. RESULTS AND CONCLUSION: The experimental results demonstrate that classification of the features from pre-trained deep models shows the promising application in COVID-19 severity screening, whereas the DenseNet-201 with cubic SVM model achieved the best performance. Specifically, it achieved the highest severity classification accuracy of 95.20% and 95.34% for tenfold cross-validation and leave-one-out, respectively. The established pipeline was able to achieve a rapid and accurate identification of the severity of COVID-19. This may assist the physicians to make more efficient and reliable decisions.

Retinal image synthesis from multiple-landmarks input with generative adversarial networks.

BACKGROUND: Medical datasets, especially medical images, are often imbalanced due to the different incidences of various diseases. To address this problem, many methods have been proposed to synthesize medical images using generative adversarial networks (GANs) to enlarge training datasets for facilitating medical image analysis. For instance, conventional methods such as image-to-image translation techniques are used to synthesize fundus images with their respective vessel trees in the field of fundus image. METHODS: In order to improve the image quality and details of the synthetic images, three key aspects of the pipeline are mainly elaborated: the input mask, architecture of GANs, and the resolution of paired images. We propose a new preprocessing pipeline named multiple-channels-multiple-landmarks (MCML), aiming to synthesize color fundus images from a combination of vessel tree, optic disc, and optic cup images. We compared both single vessel mask input and MCML mask input on two public fundus image datasets (DRIVE and DRISHTI-GS) with different kinds of Pix2pix and Cycle-GAN architectures. A new Pix2pix structure with ResU-net generator is also designed, which has been compared with the other models. RESULTS AND CONCLUSION: As shown in the results, the proposed MCML method outperforms the single vessel-based methods for each architecture of GANs. Furthermore, we find that our Pix2pix model with ResU-net generator achieves superior PSNR and SSIM performance than the other GANs. High-resolution paired images are also beneficial for improving the performance of each GAN in this work. Finally, a Pix2pix network with ResU-net generator using MCML and high-resolution paired images are able to generate good and realistic fundus images in this work, indicating that our MCML method has great potential in the field of glaucoma computer-aided diagnosis based on fundus image.

Spatial resolution enhancement for pushbroom-based microscopic hyperspectral imaging.

Hyperspectral imaging (HSI) is a promising tool for microscopic histopathology studies. Pushbroom microscopic hyperspectral imaging systems are widely used because of their low cost and easy implementation. However, the spatial resolution of pushbroom HSI systems is limited by the width of the optical entrance slit. A narrower slit leads to longer exposure time and slower imaging speed. In this paper, we explored several spatial resolution enhancement algorithms, originally designed for remote-sensing hyperspectral imaging, for pushbroom microscopic HSI systems. Our results demonstrate that those algorithms could effectively achieve a higher spatial resolution without sacrificing imaging speed.

A super-resolution method-based pipeline for fundus fluorescein angiography imaging.

BACKGROUND: Fundus fluorescein angiography (FFA) imaging is a standard diagnostic tool for many retinal diseases such as age-related macular degeneration and diabetic retinopathy. High-resolution FFA images facilitate the detection of small lesions such as microaneurysms, and other landmark changes, in the early stages; this can help an ophthalmologist improve a patient's cure rate. However, only low-resolution images are available in most clinical cases. Super-resolution (SR), which is a method to improve the resolution of an image, has been successfully employed for natural and remote sensing images. To the best of our knowledge, no one has applied SR techniques to FFA imaging so far. METHODS: In this work, we propose a SR method-based pipeline for FFA imaging. The aim of this pipeline is to enhance the image quality of FFA by using SR techniques. Several SR frameworks including neighborhood embedding, sparsity-based, locally-linear regression and deep learning-based approaches are investigated. Based on a clinical FFA dataset collected from Second Affiliated Hospital to Xuzhou Medical University, each SR method is implemented and evaluated for the pipeline to improve the resolution of FFA images. RESULTS AND CONCLUSION: As shown in our results, most SR algorithms have a positive impact on the enhancement of FFA images. Super-resolution forests (SRF), a random forest-based SR method has displayed remarkable high effectiveness and outperformed other methods. Hence, SRF should be one potential way to benefit ophthalmologists by obtaining high-resolution FFA images in a clinical setting.

HSA-net with a novel CAD pipeline boosts both clinical brain tumor MR image classification and segmentation.

Brain tumors are among the most prevalent neoplasms in current medical studies. Accurately distinguishing and classifying brain tumor types accurately is crucial for patient treatment and survival in clinical practice. However, existing computer-aided diagnostic pipelines are inadequate for practical medical use due to tumor complexity. In this study, we curated a multi-centre brain tumor dataset that includes various clinical brain tumor data types, including segmentation and classification annotations, surpassing previous efforts. To enhance brain tumor segmentation accuracy, we propose a new segmentation method: HSA-Net. This method utilizes the Shared Weight Dilated Convolution module (SWDC) and Hybrid Dense Dilated Convolution module (HDense) to capture multi-scale information while minimizing parameter count. The Effective Multi-Dimensional Attention (EMA) and Important Feature Attention (IFA) modules effectively aggregate task-related information. We introduce a novel clinical brain tumor computer-aided diagnosis pipeline (CAD) that combines HSA-Net with pipeline modification. This approach not only improves segmentation accuracy but also utilizes the segmentation mask as an additional channel feature to enhance brain tumor classification results. Our experimental evaluation of 3327 real clinical data demonstrates the effectiveness of the proposed method, achieving an average Dice coefficient of 86.85 % for segmentation and a classification accuracy of 95.35 %. We also validated the effectiveness of our proposed method using the publicly available BraTS dataset.

MHD-Net: Memory-Aware Hetero-Modal Distillation Network for Thymic Epithelial Tumor Typing With Missing Pathology Modality.

Fusing multi-modal radiology and pathology data with complementary information can improve the accuracy of tumor typing. However, collecting pathology data is difficult since it is high-cost and sometimes only obtainable after the surgery, which limits the application of multi-modal methods in diagnosis. To address this problem, we propose comprehensively learning multi-modal radiology-pathology data in training, and only using uni-modal radiology data in testing. Concretely, a Memory-aware Hetero-modal Distillation Network (MHD-Net) is proposed, which can distill well-learned multi-modal knowledge with the assistance of memory from the teacher to the student. In the teacher, to tackle the challenge in hetero-modal feature fusion, we propose a novel spatial-differentiated hetero-modal fusion module (SHFM) that models spatial-specific tumor information correlations across modalities. As only radiology data is accessible to the student, we store pathology features in the proposed contrast-boosted typing memory module (CTMM) that achieves type-wise memory updating and stage-wise contrastive memory boosting to ensure the effectiveness and generalization of memory items. In the student, to improve the cross-modal distillation, we propose a multi-stage memory-aware distillation (MMD) scheme that reads memory-aware pathology features from CTMM to remedy missing modal-specific information. Furthermore, we construct a Radiology-Pathology Thymic Epithelial Tumor (RPTET) dataset containing paired CT and WSI images with annotations. Experiments on the RPTET and CPTAC-LUAD datasets demonstrate that MHD-Net significantly improves tumor typing and outperforms existing multi-modal methods on missing modality situations.

SegCoFusion: An Integrative Multimodal Volumetric Segmentation Cooperating With Fusion Pipeline to Enhance Lesion Awareness.

Multimodal volumetric segmentation and fusion are two valuable techniques for surgical treatment planning, image-guided interventions, tumor growth detection, radiotherapy map generation, etc. In recent years, deep learning has demonstrated its excellent capability in both of the above tasks, while these methods inevitably face bottlenecks. On the one hand, recent segmentation studies, especially the U-Net-style series, have reached the performance ceiling in segmentation tasks. On the other hand, it is almost impossible to capture the ground truth of the fusion in multimodal imaging, due to differences in physical principles among imaging modalities. Hence, most of the existing studies in the field of multimodal medical image fusion, which fuse only two modalities at a time with hand-crafted proportions, are subjective and task-specific. To address the above concerns, this work proposes an integration of multimodal segmentation and fusion, namely SegCoFusion, which consists of a novel feature frequency dividing network named FDNet and a segmentation part using a dual-single path feature supplementing strategy to optimize the segmentation inputs and suture with the fusion part. Furthermore, focusing on multimodal brain tumor volumetric fusion and segmentation, the qualitative and quantitative results demonstrate that SegCoFusion can break the ceiling both of segmentation and fusion methods. Moreover, the effectiveness of the proposed framework is also revealed by comparing it with state-of-the-art fusion methods on 2D two-modality fusion tasks, our method achieves better fusion performance than others. Therefore, the proposed SegCoFusion develops a novel perspective that improves the performance in volumetric fusion by cooperating with segmentation and enhances lesion awareness.

Privacy-Preserving Federated Learning with Domain Adaptation for Multi-Disease Ocular Disease Recognition.

As one of the effective ways of ocular disease recognition, early fundus screening can help patients avoid unrecoverable blindness. Although deep learning is powerful for image-based ocular disease recognition, the performance mainly benefits from a large number of labeled data. For ocular disease, data collection and annotation in a single site usually take a lot of time. If multi-site data are obtained, there are two main issues: 1) the data privacy is easy to be leaked; 2) the domain gap among sites will influence the recognition performance. Inspired by the above, first, a Gaussian randomized mechanism is adopted in local sites, which are then engaged in a global model to preserve the data privacy of local sites and models. Second, to bridge the domain gap among different sites, a two-step domain adaptation method is introduced, which consists of a domain confusion module and a multi-expert learning strategy. Based on the above, a privacy-preserving federated learning framework with domain adaptation is constructed. In the experimental part, a multi-disease early fundus screening dataset, including a detailed ablation study and four experimental settings, is used to show the stepwise performance, which verifies the efficiency of our proposed framework.

TransU²-Net: An Effective Medical Image Segmentation Framework Based on Transformer and U²-Net.

BACKGROUND: In the past few years, U-Net based U-shaped architecture and skip-connections have made incredible progress in the field of medical image segmentation. U2-Net achieves good performance in computer vision. However, in the medical image segmentation task, U2-Net with over nesting is easy to overfit. PURPOSE: A 2D network structure TransU2-Net combining transformer and a lighter weight U2-Net is proposed for automatic segmentation of brain tumor magnetic resonance image (MRI). METHODS: The light-weight U2-Net architecture not only obtains multi-scale information but also reduces redundant feature extraction. Meanwhile, the transformer block embedded in the stacked convolutional layer obtains more global information; the transformer with skip-connection enhances spatial domain information representation. A new multi-scale feature map fusion strategy as a postprocessing method was proposed for better fusing high and low-dimensional spatial information. RESULTS: Our proposed model TransU2-Net achieves better segmentation results, on the BraTS2021 dataset, our method achieves an average dice coefficient of 88.17%; Evaluation on the publicly available MSD dataset, we perform tumor evaluation, we achieve a dice coefficient of 74.69%; in addition to comparing the TransU2-Net results are compared with previously proposed 2D segmentation methods. CONCLUSIONS: We propose an automatic medical image segmentation method combining transformers and U2-Net, which has good performance and is of clinical importance. The experimental results show that the proposed method outperforms other 2D medical image segmentation methods. Clinical Translation Statement: We use the BarTS2021 dataset and the MSD dataset which are publicly available databases. All experiments in this paper are in accordance with medical ethics.

A brain tumor computer-aided diagnosis method with automatic lesion segmentation and ensemble decision strategy.

Objectives: Gliomas and brain metastases (Mets) are the most common brain malignancies. The treatment strategy and clinical prognosis of patients are different, requiring accurate diagnosis of tumor types. However, the traditional radiomics diagnostic pipeline requires manual annotation and lacks integrated methods for segmentation and classification. To improve the diagnosis process, a gliomas and Mets computer-aided diagnosis method with automatic lesion segmentation and ensemble decision strategy on multi-center datasets was proposed. Methods: Overall, 1,022 high-grade gliomas and 775 Mets patients' preoperative MR images were adopted in the study, including contrast-enhanced T1-weighted (T1-CE) and T2-fluid attenuated inversion recovery (T2-flair) sequences from three hospitals. Two segmentation models trained on the gliomas and Mets datasets, respectively, were used to automatically segment tumors. Multiple radiomics features were extracted after automatic segmentation. Several machine learning classifiers were used to measure the impact of feature selection methods. A weight soft voting (RSV) model and ensemble decision strategy based on prior knowledge (EDPK) were introduced in the radiomics pipeline. Accuracy, sensitivity, specificity, and the area under the receiver operating characteristic curve (AUC) were used to evaluate the classification performance. Results: The proposed pipeline improved the diagnosis of gliomas and Mets with ACC reaching 0.8950 and AUC reaching 0.9585 after automatic lesion segmentation, which was higher than those of the traditional radiomics pipeline (ACC:0.8850, AUC:0.9450). Conclusion: The proposed model accurately classified gliomas and Mets patients using MRI radiomics. The novel pipeline showed great potential in diagnosing gliomas and Mets with high generalizability and interpretability.

TUCNet: A channel and spatial attention-based graph convolutional network for teeth upsampling and completion.

With the increasing popularity of the use of 3D scanning equipment in capturing oral cavity in dental health applications, the quality of 3D dental models has become vital in oral prosthodontics and orthodontics. However, the point cloud data obtained can often be sparse and thus missing information. To address this issue, we construct a high-resolution teeth point cloud completion method named TUCNet to fill up the sparse and incomplete oral point cloud collected and output a dense and complete teeth point cloud. First, we propose a Channel and Spatial Attentive EdgeConv (CSAE) module to fuse local and global contexts in the point feature extraction. Second, we propose a CSAE-based point cloud upsample (CPCU) module to gradually increase the number of points in the point clouds. TUCNet employs a tree-based approach to generate complete point clouds, where child points are derived through a splitting process from parent points following each CPCU. The CPCU learns the up-sampling pattern of each parent point by combining the attention mechanism and the point deconvolution operation. Skip connections are introduced between CPCUs to summarize the split mode of the previous layer of CPCUs, which is used to generate the split mode of the current CPCUs. We conduct numerous experiments on the teeth point cloud completion dataset and the PCN dataset. The experimental results show that our TUCNet not only achieves the state-of-the-art performance on the teeth dataset, but also achieves excellent performance on the PCN dataset.

DCAM-NET:A novel domain generalization optic cup and optic disc segmentation pipeline with multi-region and multi-scale convolution attention mechanism.

Fundus images are an essential basis for diagnosing ocular diseases, and using convolutional neural networks has shown promising results in achieving accurate fundus image segmentation. However, the difference between the training data (source domain) and the testing data (target domain) will significantly affect the final segmentation performance. This paper proposes a novel framework named DCAM-NET for fundus domain generalization segmentation, which substantially improves the generalization ability of the segmentation model to the target domain data and enhances the extraction of detailed information on the source domain data. This model can effectively overcome the problem of poor model performance due to cross-domain segmentation. To enhance the adaptability of the segmentation model to target domain data, this paper proposes a multi-scale attention mechanism module (MSA) that functions at the feature extraction level. Extracting different attribute features to enter the corresponding scale attention module further captures the critical features in channel, position, and spatial regions. The MSA attention mechanism module also integrates the characteristics of the self-attention mechanism, it can capture dense context information, and the aggregation of multi-feature information effectively enhances the generalization of the model when dealing with unknown domain data. In addition, this paper proposes the multi-region weight fusion convolution module (MWFC), which is essential for the segmentation model to extract feature information from the source domain data accurately. Fusing multiple region weights and convolutional kernel weights on the image to enhance the model adaptability to information at different locations on the image, the fusion of weights deepens the capacity and depth of the model. It enhances the learning ability of the model for multiple regions on the source domain. Our experiments on fundus data for cup/disc segmentation show that the introduction of MSA and MWFC modules in this paper effectively improves the segmentation ability of the segmentation model on the unknown domain. And the performance of the proposed method is significantly better than other methods in the current domain generalization segmentation of the optic cup/disc.

BTMF-GAN: A multi-modal MRI fusion generative adversarial network for brain tumors.

Image fusion techniques have been widely used for multi-modal medical image fusion tasks. Most existing methods aim to improve the overall quality of the fused image and do not focus on the more important textural details and contrast between the tissues of the lesion in the regions of interest (ROIs). This can lead to the distortion of important tumor ROIs information and thus limits the applicability of the fused images in clinical practice. To improve the fusion quality of ROIs relevant to medical implications, we propose a multi-modal MRI fusion generative adversarial network (BTMF-GAN) for the task of multi-modal MRI fusion of brain tumors. Unlike existing deep learning approaches which focus on improving the global quality of the fused image, the proposed BTMF-GAN aims to achieve a balance between tissue details and structural contrasts in brain tumor, which is the region of interest crucial to many medical applications. Specifically, we employ a generator with a U-shaped nested structure and residual U-blocks (RSU) to enhance multi-scale feature extraction. To enhance and recalibrate features of the encoder, the multi-perceptual field adaptive transformer feature enhancement module (MRF-ATFE) is used between the encoder and the decoder instead of a skip connection. To increase contrast between tumor tissues of the fused image, a mask-part block is introduced to fragment the source image and the fused image, based on which, we propose a novel salient loss function. Qualitative and quantitative analysis of the results on the public and clinical datasets demonstrate the superiority of the proposed approach to many other commonly used fusion methods.

BTSC-TNAS: A neural architecture search-based transformer for brain tumor segmentation and classification.

Glioblastoma (GBM), isolated brain metastasis (SBM), and primary central nervous system lymphoma (PCNSL) possess a high level of similarity in histomorphology and clinical manifestations on multimodal MRI. Such similarities have led to challenges in the clinical diagnosis of these three malignant tumors. However, many existing models solely focus on either the task of segmentation or classification, which limits the application of computer-aided diagnosis in clinical diagnosis and treatment. To solve this problem, we propose a multi-task learning transformer with neural architecture search (NAS) for brain tumor segmentation and classification (BTSC-TNAS). In the segmentation stage, we use a nested transformer U-shape network (NTU-NAS) with NAS to directly predict brain tumor masks from multi-modal MRI images. In the tumor classification stage, we use the multiscale features obtained from the encoder of NTU-NAS as the input features of the classification network (MSC-NET), which are integrated and corrected by the classification feature correction enhancement (CFCE) block to improve the accuracy of classification. The proposed BTSC-TNAS achieves an average Dice coefficient of 80.86% and 87.12% for the segmentation of tumor region and the maximum abnormal region in clinical data respectively. The model achieves a classification accuracy of 0.941. The experiments performed on the BraTS 2019 dataset show that the proposed BTSC-TNAS has excellent generalizability and can provide support for some challenging tasks in the diagnosis and treatment of brain tumors.