Search

The success of deep learning methodologies hinges upon the availability of meticulously labeled extensive datasets. However, when dealing with medical images, the annotation process for such abundant training data often necessitates the involvement of experienced radiologists, thereby consuming their limited time resources. In order to alleviate this burden, few-shot learning approaches have been developed, which manage to achieve competitive performance levels with only several labeled images. Nevertheless, a crucial yet previously overlooked problem in few-shot learning is about the selection of template images for annotation before learning, which affects the final performance. In this study, we propose a novel TEmplate Choosing Policy (TECP) that aims to identify and select "the most worthy" images for annotation, particularly within the context of multiple few-shot medical tasks, including landmark detection, anatomy detection, and anatomy segmentation. TECP is composed of four integral components: (1) Self-supervised training, which entails training a pre-existing deep model to extract salient features from radiological images; (2) Alternative proposals for localizing informative regions within the images; and (3) Representative Score Estimation, which involves the evaluation and identification of the most representative samples or templates. (4) Ranking, which rank all candidates and select one with highest representative score. The efficacy of the TECP approach is demonstrated through a series of comprehensive experiments conducted on multiple public datasets. Across all three medical tasks, the utilization of TECP yields noticeable improvements in model performance., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2024 Elsevier B.V. All rights reserved.)

Recently, there has been a trend of designing neural data structures to go beyond handcrafted data structures by leveraging patterns of data distributions for better accuracy and adaptivity. Sketches are widely used data structures in real-time web analysis, network monitoring, and self-driving to estimate item frequencies of data streams within limited space. However, existing sketches have not fully exploited the patterns of the data stream distributions, making it challenging to tightly couple them with neural networks that excel at memorizing pattern information. Starting from the premise, we envision a pure neural data structure as a base sketch, which we term the meta-sketch, to reinvent the base structure of conventional sketches. The meta-sketch learns basic sketching abilities from meta-tasks constituted with synthetic datasets following Zipf distributions in the pre-training phase and can be quickly adapted to real (skewed) distributions in the adaption phase. The meta-sketch not only surpasses its competitors in sketching conventional data streams but also holds good potential in supporting more complex streaming data, such as multimedia and graph stream scenarios. Extensive experiments demonstrate the superiority of the meta-sketch and offer insights into its working mechanism.

Radiologists possess diverse training and clinical experiences, leading to variations in the segmentation annotations of lung nodules and resulting in segmentation uncertainty. Conventional methods typically select a single annotation as the learning target or attempt to learn a latent space comprising multiple annotations. However, these approaches fail to leverage the valuable information inherent in the consensus and disagreements among the multiple annotations. In this paper, we propose an Uncertainty-Aware Attention Mechanism (UAAM) that utilizes consensus and disagreements among multiple annotations to facilitate better segmentation. To this end, we introduce the Multi-Confidence Mask (MCM), which combines a Low-Confidence (LC) Mask and a High-Confidence (HC) Mask. The LC mask indicates regions with low segmentation confidence, where radiologists may have different segmentation choices. Following UAAM, we further design an Uncertainty-Guide Multi-Confidence Segmentation Network (UGMCS-Net), which contains three modules: a Feature Extracting Module that captures a general feature of a lung nodule, an Uncertainty-Aware Module that produces three features for the annotations' union, intersection, and annotation set, and an Intersection-Union Constraining Module that uses distances between the three features to balance the predictions of final segmentation and MCM. To comprehensively demonstrate the performance of our method, we propose a Complex-Nodule Validation on LIDC-IDRI, which tests UGMCS-Net's segmentation performance on lung nodules that are difficult to segment using common methods. Experimental results demonstrate that our method can significantly improve the segmentation performance on nodules that are difficult to segment using conventional methods.

For medical image segmentation, contrastive learning is the dominant practice to improve the quality of visual representations by contrasting semantically similar and dissimilar pairs of samples. This is enabled by the observation that without accessing ground truth labels, negative examples with truly dissimilar anatomical features, if sampled, can significantly improve the performance. In reality, however, these samples may come from similar anatomical regions and the models may struggle to distinguish the minority tail-class samples, making the tail classes more prone to misclassification, both of which typically lead to model collapse. In this paper, we propose ARCO, a semi-supervised contrastive learning (CL) framework with stratified group theory for medical image segmentation. In particular, we first propose building ARCO through the concept of variance-reduced estimation and show that certain variance-reduction techniques are particularly beneficial in pixel/voxel-level segmentation tasks with extremely limited labels. Furthermore, we theoretically prove these sampling techniques are universal in variance reduction. Finally, we experimentally validate our approaches on eight benchmarks, i.e ., five 2D/3D medical and three semantic segmentation datasets, with different label settings, and our methods consistently outperform state-of-the-art semi-supervised methods. Additionally, we augment the CL frameworks with these sampling techniques and demonstrate significant gains over previous methods. We believe our work is an important step towards semi-supervised medical image segmentation by quantifying the limitation of current self-supervision objectives for accomplishing such challenging safety-critical tasks.

Low-count PET is an efficient way to reduce radiation exposure and acquisition time, but the reconstructed images often suffer from low signal-to-noise ratio (SNR), thus affecting diagnosis and other downstream tasks. Recent advances in deep learning have shown great potential in improving low-count PET image quality, but acquiring a large, centralized, and diverse dataset from multiple institutions for training a robust model is difficult due to privacy and security concerns of patient data. Moreover, low-count PET data at different institutions may have different data distribution, thus requiring personalized models. While previous federated learning (FL) algorithms enable multi-institution collaborative training without the need of aggregating local data, addressing the large domain shift in the application of multi-institutional low-count PET denoising remains a challenge and is still highly under-explored. In this work, we propose FedFTN, a personalized federated learning strategy that addresses these challenges. FedFTN uses a local deep feature transformation network (FTN) to modulate the feature outputs of a globally shared denoising network, enabling personalized low-count PET denoising for each institution. During the federated learning process, only the denoising network's weights are communicated and aggregated, while the FTN remains at the local institutions for feature transformation. We evaluated our method using a large-scale dataset of multi-institutional low-count PET imaging data from three medical centers located across three continents, and showed that FedFTN provides high-quality low-count PET images, outperforming previous baseline FL reconstruction methods across all low-count levels at all three institutions., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2023 Elsevier B.V. All rights reserved.)

Multi-modal tumor segmentation exploits complementary information from different modalities to help recognize tumor regions. Known multi-modal segmentation methods mainly have deficiencies in two aspects: First, the adopted multi-modal fusion strategies are built upon well-aligned input images, which are vulnerable to spatial misalignment between modalities (caused by respiratory motions, different scanning parameters, registration errors, etc). Second, the performance of known methods remains subject to the uncertainty of segmentation, which is particularly acute in tumor boundary regions. To tackle these issues, in this paper, we propose a novel multi-modal tumor segmentation method with deformable feature fusion and uncertain region refinement. Concretely, we introduce a deformable aggregation module, which integrates feature alignment and feature aggregation in an ensemble, to reduce inter-modality misalignment and make full use of cross-modal information. Moreover, we devise an uncertain region inpainting module to refine uncertain pixels using neighboring discriminative features. Experiments on two clinical multi-modal tumor datasets demonstrate that our method achieves promising tumor segmentation results and outperforms state-of-the-art methods.

Background: Automated surgical workflow recognition is the foundation for computational models of medical knowledge to interpret surgical procedures. The fine-grained segmentation of the surgical process and the improvement of the accuracy of surgical workflow recognition facilitate the realization of autonomous robotic surgery. This study aimed to construct a multigranularity temporal annotation dataset of the standardized robotic left lateral sectionectomy (RLLS) and develop a deep learning-based automated model for multilevel overall and effective surgical workflow recognition., Methods: From December 2016 to May 2019, 45 cases of RLLS videos were enrolled in our dataset. All frames of RLLS videos in this study are labeled with temporal annotations. The authors defined those activities that truly contribute to the surgery as effective frames, while other activities are labeled as under-effective frames. Effective frames of all RLLS videos are annotated with three hierarchical levels of 4 steps, 12 tasks, and 26 activities. A hybrid deep learning model were used for surgical workflow recognition of steps, tasks, activities, and under-effective frames. Moreover, the authors also carried out multilevel effective surgical workflow recognition after removing under-effective frames., Results: The dataset comprises 4 383 516 annotated RLLS video frames with multilevel annotation, of which 2 418 468 frames are effective. The overall accuracies of automated recognition for Steps, Tasks, Activities, and under-effective frames are 0.82, 0.80, 0.79, and 0.85, respectively, with corresponding precision values of 0.81, 0.76, 0.60, and 0.85. In multilevel effective surgical workflow recognition, the overall accuracies were increased to 0.96, 0.88, and 0.82 for Steps, Tasks, and Activities, respectively, while the precision values were increased to 0.95, 0.80, and 0.68., Conclusion: In this study, the authors created a dataset of 45 RLLS cases with multilevel annotations and developed a hybrid deep learning model for surgical workflow recognition. The authors demonstrated a fairly higher accuracy in multilevel effective surgical workflow recognition when under-effective frames were removed. Our research could be helpful in the development of autonomous robotic surgery., (Copyright © 2023 The Author(s). Published by Wolters Kluwer Health, Inc.)

Objective. The steady-state visual evoked potential (SSVEP)-based brain-computer interface has received extensive attention in research due to its simple system, less training data, and high information transfer rate. There are currently two prominent methods dominating the classification of SSVEP signals. One is the knowledge-based task-related component analysis (TRCA) method, whose core idea is to find the spatial filters by maximizing the inter-trial covariance. The other is the deep learning-based approach, which directly learns a classification model from data. However, how to integrate the two methods to achieve better performance has not been studied before. Approach. In this study, we develop a novel algorithm named TRCA-Net (TRCA-Net) to enhance SSVEP signal classification, which enjoys the advantages of both the knowledge-based method and the deep model. Specifically, the proposed TRCA-Net first performs TRCA to obtain spatial filters, which extract task-related components of data. Then the TRCA-filtered features from different filters are rearranged as new multi-channel signals for a deep convolutional neural network (CNN) for classification. Introducing the TRCA filters to a deep learning-based approach improves the signal-to-noise ratio of input data, hence benefiting the deep learning model. Main results. We evaluate the performance of TRCA-Net using two publicly available large-scale benchmark datasets, and the results demonstrate the effectiveness of TRCA-Net. Additionally, offline and online experiments separately testing ten and five subjects further validate the robustness of TRCA-Net. Further, we conduct ablation studies on different CNN backbones and demonstrate that our approach can be transplanted into other CNN models to boost their performance. Significance. The proposed approach is believed to have a promising potential for SSVEP classification and promote its practical applications in communication and control. The code is available athttps://github.com/Sungden/TRCA-Net., (© 2023 IOP Publishing Ltd.)

Gene mutation detection is usually carried out by molecular biological methods, which is expensive and has a long-time cycle. In contrast, pathological images are ubiquitous. If clinically significant gene mutations can be predicted only through pathological images, it will greatly promote the widespread use of gene mutation detection in clinical practice. However, current gene mutation prediction methods based on pathological images are ineffective because of the inability to identify mutated regions in gigapixel Whole Slide Image (WSI). To address this challenge, hereby we propose a carefully designed framework for WSI-based gene mutation prediction, which consists of three parts. (i) The first part of cancerous area segmentation, based on supervised learning, quickly filters out a large number of non-mutated regions; (ii) the second part of cancerous patch clustering, based on the representations derived from contrastive learning, ensures the comprehensiveness of patch selection; and (iii) the third part of mutation classification, based on the proposed hierarchical deep multi-instance learning method (HDMIL), ensures that sufficient patches are considered and inaccurate selections are ignored. In addition, benefiting from a two-stage attention mechanism in HDMIL, the patches that are highly correlated with gene mutations can be identified. This interpretability can help a pathologist to analyze the correlation between gene mutation and histopathological morphology. Experimental results demonstrate that the proposed gene mutation prediction framework significantly outperforms the state-of-the-art methods. In the TCGA bladder cancer dataset, five clinically relevant gene mutations are well predicted., Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2023 Elsevier B.V. All rights reserved.)

We used deep neural networks to process the mass spectrometry imaging (MSI) data of mouse muscle (young vs aged) and human cancer (tumor vs normal adjacent) tissues, with the aim of using explainable artificial intelligence (XAI) methods to rapidly identify biomarkers that can distinguish different classes of tissues, from several thousands of metabolite features. We also modified classic neural network architectures to construct a deep convolutional neural network that is more suitable for processing high-dimensional MSI data directly, instead of using dimension reduction techniques, and compared it to seven other machine learning analysis methods' performance in classification accuracy. After ascertaining the superiority of Channel-ResNet10, we used a novel channel selection-based XAI method to identify the key metabolite features that were responsible for its learning accuracy. These key metabolite biomarkers were then processed using MetaboAnalyst for pathway enrichment mapping. We found that Channel-ResNet10 was superior to seven other machine learning methods for MSI analysis, reaching > 98% accuracy in muscle aging and colorectal cancer datasets. We also used a novel channel selection-based XAI method to find that in young and aged muscle tissues, the differentially distributed metabolite biomarkers were especially enriched in the propanoate metabolism pathway, suggesting it as a novel target pathway for anti-aging therapy., (© 2023. Springer-Verlag GmbH Germany, part of Springer Nature.)

Background: The determination of pathological grading has a guiding significance for the treatment of pancreatic ductal adenocarcinoma (PDAC) patients. However, there is a lack of an accurate and safe method to obtain pathological grading before surgery. The aim of this study is to develop a deep learning (DL) model based on 18 F-fluorodeoxyglucose-positron emission tomography/computed tomography ( 18 F-FDG-PET/CT) for a fully automatic prediction of preoperative pathological grading of pancreatic cancer., Methods: A total of 370 PDAC patients from January 2016 to September 2021 were collected retrospectively. All patients underwent 18 F-FDG-PET/CT examination before surgery and obtained pathological results after surgery. A DL model for pancreatic cancer lesion segmentation was first developed using 100 of these cases and applied to the remaining cases to obtain lesion regions. After that, all patients were divided into training set, validation set, and test set according to the ratio of 5:1:1. A predictive model of pancreatic cancer pathological grade was developed using the features computed from the lesion regions obtained by the lesion segmentation model and key clinical characteristics of the patients. Finally, the stability of the model was verified by sevenfold cross-validation., Results: The Dice score of the developed PET/CT-based tumor segmentation model for PDAC was 0.89. The area under curve (AUC) of the PET/CT-based DL model developed on the basis of the segmentation model was 0.74, with an accuracy, sensitivity, and specificity of 0.72, 0.73, and 0.72, respectively. After integrating key clinical data, the AUC of the model improved to 0.77, with its accuracy, sensitivity, and specificity boosted to 0.75, 0.77, and 0.73, respectively., Conclusion: To the best of our knowledge, this is the first deep learning model to end-to-end predict the pathological grading of PDAC in a fully automatic manner, which is expected to improve clinical decision-making., (© 2023. The Author(s).)

Semi-supervised learning (SSL) methods show their powerful performance to deal with the issue of data shortage in the field of medical image segmentation. However, existing SSL methods still suffer from the problem of unreliable predictions on unannotated data due to the lack of manual annotations for them. In this paper, we propose an unreliability-diluted consistency training (UDiCT) mechanism to dilute the unreliability in SSL by assembling reliable annotated data into unreliable unannotated data. Specifically, we first propose an uncertainty-based data pairing module to pair annotated data with unannotated data based on a complementary uncertainty pairing rule, which avoids two hard samples being paired off. Secondly, we develop SwapMix, a mixed sample data augmentation method, to integrate annotated data into unannotated data for training our model in a low-unreliability manner. Finally, UDiCT is trained by minimizing a supervised loss and an unreliability-diluted consistency loss, which makes our model robust to diverse backgrounds. Extensive experiments on three chest CT datasets show the effectiveness of our method for semi-supervised CT lesion segmentation.

In clinics, a radiology report is crucial for guiding a patient's treatment. However, writing radiology reports is a heavy burden for radiologists. To this end, we present an automatic, multi-modal approach for report generation from a chest x-ray. Our approach, motivated by the observation that the descriptions in radiology reports are highly correlated with specific information of the x-ray images, features two distinct modules: (i) Learned knowledge base: To absorb the knowledge embedded in the radiology reports, we build a knowledge base that can automatically distill and restore medical knowledge from textual embedding without manual labor; (ii) Multi-modal alignment: to promote the semantic alignment among reports, disease labels, and images, we explicitly utilize textual embedding to guide the learning of the visual feature space. We evaluate the performance of the proposed model using metrics from both natural language generation and clinic efficacy on the public IU-Xray and MIMIC-CXR datasets. Our ablation study shows that each module contributes to improving the quality of generated reports. Furthermore, the assistance of both modules, our approach outperforms state-of-the-art methods over almost all the metrics. Code is available at https://github.com/LX-doctorAI1/M2KT., Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2023 The Author(s). Published by Elsevier B.V. All rights reserved.)

Transformer, one of the latest technological advances of deep learning, has gained prevalence in natural language processing or computer vision. Since medical imaging bear some resemblance to computer vision, it is natural to inquire about the status quo of Transformers in medical imaging and ask the question: can the Transformer models transform medical imaging? In this paper, we attempt to make a response to the inquiry. After a brief introduction of the fundamentals of Transformers, especially in comparison with convolutional neural networks (CNNs), and highlighting key defining properties that characterize the Transformers, we offer a comprehensive review of the state-of-the-art Transformer-based approaches for medical imaging and exhibit current research progresses made in the areas of medical image segmentation, recognition, detection, registration, reconstruction, enhancement, etc. In particular, what distinguishes our review lies in its organization based on the Transformer's key defining properties, which are mostly derived from comparing the Transformer and CNN, and its type of architecture, which specifies the manner in which the Transformer and CNN are combined, all helping the readers to best understand the rationale behind the reviewed approaches. We conclude with discussions of future perspectives., Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2023 Elsevier B.V. All rights reserved.)

While deep learning methods hitherto have achieved considerable success in medical image segmentation, they are still hampered by two limitations: (i) reliance on large-scale well-labeled datasets, which are difficult to curate due to the expert-driven and time-consuming nature of pixel-level annotations in clinical practices, and (ii) failure to generalize from one domain to another, especially when the target domain is a different modality with severe domain shifts. Recent unsupervised domain adaptation (UDA) techniques leverage abundant labeled source data together with unlabeled target data to reduce the domain gap, but these methods degrade significantly with limited source annotations. In this study, we address this underexplored UDA problem, investigating a challenging but valuable realistic scenario, where the source domain not only exhibits domain shift w.r.t. the target domain but also suffers from label scarcity. In this regard, we propose a novel and generic framework called "Label-Efficient Unsupervised Domain Adaptation" (LE-UDA). In LE-UDA, we construct self-ensembling consistency for knowledge transfer between both domains, as well as a self-ensembling adversarial learning module to achieve better feature alignment for UDA. To assess the effectiveness of our method, we conduct extensive experiments on two different tasks for cross-modality segmentation between MRI and CT images. Experimental results demonstrate that the proposed LE-UDA can efficiently leverage limited source labels to improve cross-domain segmentation performance, outperforming state-of-the-art UDA approaches in the literature.

Positron emission tomography (PET) with a reduced injection dose, i.e., low-dose PET, is an efficient way to reduce radiation dose. However, low-dose PET reconstruction suffers from a low signal-to-noise ratio (SNR), affecting diagnosis and other PET-related applications. Recently, deep learning-based PET denoising methods have demonstrated superior performance in generating high-quality reconstruction. However, these methods require a large amount of representative data for training, which can be difficult to collect and share due to medical data privacy regulations. Moreover, low-dose PET data at different institutions may use different low-dose protocols, leading to non-identical data distribution. While previous federated learning (FL) algorithms enable multi-institution collaborative training without the need of aggregating local data, it is challenging for previous methods to address the large domain shift caused by different low-dose PET settings, and the application of FL to PET is still under-explored. In this work, we propose a federated transfer learning (FTL) framework for low-dose PET denoising using heterogeneous low-dose data. Our experimental results on simulated multi-institutional data demonstrate that our method can efficiently utilize heterogeneous low-dose data without compromising data privacy for achieving superior low-dose PET denoising performance for different institutions with different low-dose settings, as compared to previous FL methods.

To reduce the potential risk of radiation to the patient, low-dose computed tomography (LDCT) has been widely adopted in clinical practice for reconstructing cross-sectional images using sinograms with reduced x-ray flux. The LDCT image quality is often degraded by different levels of noise depending on the low-dose protocols. The image quality will be further degraded when the patient has metallic implants, where the image suffers from additional streak artifacts along with further amplified noise levels, thus affecting the medical diagnosis and other CT-related applications. Previous studies mainly focused either on denoising LDCT without considering metallic implants or full-dose CT metal artifact reduction (MAR). Directly applying previous LDCT or MAR approaches to the issue of simultaneous metal artifact reduction and low-dose CT (MARLD) may yield sub-optimal reconstruction results. In this work, we develop a dual-domain under-to-fully-complete progressive restoration network, called DuDoUFNet, for MARLD. Our DuDoUFNet aims to reconstruct images with substantially reduced noise and artifact by progressive sinogram to image domain restoration with a two-stage progressive restoration network design. Our experimental results demonstrate that our method can provide high-quality reconstruction, superior to previous LDCT and MAR methods under various low-dose and metal settings.

Automatic chest radiology report generation is critical in clinics which can relieve experienced radiologists from the heavy workload and remind inexperienced radiologists of misdiagnosis or missed diagnose. Existing approaches mainly formulate chest radiology report generation as an image captioning task and adopt the encoder-decoder framework. However, in the medical domain, such pure data-driven approaches suffer from the following problems: 1) visual and textual bias problem; 2) lack of expert knowledge. In this paper, we propose a knowledge-enhanced radiology report generation approach introduces two types of medical knowledge: 1) General knowledge, which is input independent and provides the broad knowledge for report generation; 2) Specific knowledge, which is input dependent and provides the fine-grained knowledge for chest X-ray report generation. To fully utilize both the general and specific knowledge, we also propose a knowledge-enhanced multi-head attention mechanism. By merging the visual features of the radiology image with general knowledge and specific knowledge, the proposed model can improve the quality of generated reports. The experimental results on the publicly available IU-Xray dataset show that the proposed knowledge-enhanced approach outperforms state-of-the-art methods in almost all metrics. And the results of MIMIC-CXR dataset show that the proposed knowledge-enhanced approach is on par with state-of-the-art methods. Ablation studies also demonstrate that both general and specific knowledge can help to improve the performance of chest radiology report generation., Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2022 The Author(s). Published by Elsevier B.V. All rights reserved.)

Objective and Impact Statement . In this work, we develop a universal anatomical landmark detection model which learns once from multiple datasets corresponding to different anatomical regions. Compared with the conventional model trained on a single dataset, this universal model not only is more light weighted and easier to train but also improves the accuracy of the anatomical landmark location. Introduction . The accurate and automatic localization of anatomical landmarks plays an essential role in medical image analysis. However, recent deep learning-based methods only utilize limited data from a single dataset. It is promising and desirable to build a model learned from different regions which harnesses the power of big data. Methods . Our model consists of a local network and a global network, which capture local features and global features, respectively. The local network is a fully convolutional network built up with depth-wise separable convolutions, and the global network uses dilated convolution to enlarge the receptive field to model global dependencies. Results . We evaluate our model on four 2D X-ray image datasets totaling 1710 images and 72 landmarks in four anatomical regions. Extensive experimental results show that our model improves the detection accuracy compared to the state-of-the-art methods. Conclusion . Our model makes the first attempt to train a single network on multiple datasets for landmark detection. Experimental results qualitatively and quantitatively show that our proposed model performs better than other models trained on multiple datasets and even better than models trained on a single dataset separately., Competing Interests: The authors declare that there is no conflict of interest regarding the publication of this article., (Copyright © 2022 Heqin Zhu et al.)

Clinical evidence has shown that rib-suppressed chest X-rays (CXRs) can improve the reliability of pulmonary disease diagnosis. However, previous approaches on generating rib-suppressed CXR face challenges in preserving details and eliminating rib residues. We hereby propose a GAN-based disentanglement learning framework called Rib Suppression GAN, or RSGAN, to perform rib suppression by utilizing the anatomical knowledge embedded in unpaired computed tomography (CT) images. In this approach, we employ a residual map to characterize the intensity difference between CXR and the corresponding rib-suppressed result. To predict the residual map in CXR domain, we disentangle the image into structure- and contrast-specific features and transfer the rib structural priors from digitally reconstructed radiographs (DRRs) computed by CT. Furthermore, we employ additional adaptive loss to suppress rib residue and preserve more details. We conduct extensive experiments based on 1673 CT volumes, and four benchmarking CXR datasets, totaling over 120K images, to demonstrate that (i) our proposed RSGAN achieves superior image quality compared to the state-of-the-art rib suppression methods; (ii) combining CXR with our rib-suppressed result leads to better performance in lung disease classification and tuberculosis area detection., Competing Interests: Declaration of Competing Interest The authors have no conflict of interest., (Copyright © 2022 Elsevier B.V. All rights reserved.)

Sparse-view computed tomography (SVCT) aims to reconstruct a cross-sectional image using a reduced number of x-ray projections. While SVCT can efficiently reduce the radiation dose, the reconstruction suffers from severe streak artifacts, and the artifacts are further amplified with the presence of metallic implants, which could adversely impact the medical diagnosis and other downstream applications. Previous methods have extensively explored either SVCT reconstruction without metallic implants, or full-view CT metal artifact reduction (MAR). The issue of simultaneous sparse-view and metal artifact reduction (SVMAR) remains under-explored, and it is infeasible to directly apply previous SVCT and MAR methods to SVMAR which may yield non-ideal reconstruction quality. In this work, we propose a dual-domain data consistent recurrent network, called DuDoDR-Net, for SVMAR. Our DuDoDR-Net aims to reconstruct an artifact-free image by recurrent image domain and sinogram domain restorations. To ensure the metal-free part of acquired projection data is preserved, we also develop the image data consistent layer (iDCL) and sinogram data consistent layer (sDCL) that are interleaved in our recurrent framework. Our experimental results demonstrate that our DuDoDR-Net is able to produce superior artifact-reduced results while preserving the anatomical structures, that outperforming previous SVCT and SVMAR methods, under different sparse-view acquisition settings., Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2021 Elsevier B.V. All rights reserved.)

Background: Mitral regurgitation (MR) is the most common valve lesion worldwide. However, the quantitative assessment of MR severity based on current guidelines is challenging and time-consuming; strict adherence to applying these guidelines is therefore relatively infrequent. We aimed to develop an automatic, reliable and reproducible artificial intelligence (AI) diagnostic system to assist physicians in grading MR severity based on color video Doppler echocardiography via a self-supervised learning (SSL) algorithm., Methods: We constructed a retrospective cohort of 2,766 consecutive echocardiographic studies of patients with MR diagnosed based on clinical criteria from two hospitals in China. One hundred and forty-eight studies with reference standards were selected in the main analysis and also served as the test set for the AI segmentation model. Five hundred and ninety-two and 148 studies were selected with stratified random sampling as the training and validation datasets, respectively. The self-supervised algorithm captures features and segments the MR jet and left atrium (LA) area, and the output is used to assist physicians in MR severity grading. The diagnostic performance of physicians without and with the support from AI was estimated and compared., Results: The performance of SSL algorithm yielded 89.2% and 85.3% average segmentation dice similarity coefficient (DICE) on the validation and test datasets, which achieved 6.2% and 8.1% improvement compared to Residual U-shape Network (ResNet-UNet), respectively. When physicians were provided the output of algorithm for grading MR severity, the sensitivity increased from 77.0% (95% CI: 70.9-82.1%) to 86.7% (95% CI: 80.3-91.2%) and the specificity was largely unchanged: 91.5% (95% CI: 87.8-94.1%) vs. 90.5% (95% CI: 86.7-93.2%)., Conclusions: This study provides a new, practical, accurate, plug-and-play AI-assisted approach for assisting physicians in MR severity grading that can be easily implemented in clinical practice., Competing Interests: Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://dx.doi.org/10.21037/atm-21-3449). FY reports funding from the Beijing Natural Science and Technology Foundation (7202198). DB reports being a consultant to BioMind. KH reported funding from Ministry of Industry and Information Technology of China (2020-0103-3-1). The other authors have no conflicts of interest to declare., (2022 Annals of Translational Medicine. All rights reserved.)

Deep reinforcement learning (DRL) augments the reinforcement learning framework, which learns a sequence of actions that maximizes the expected reward, with the representative power of deep neural networks. Recent works have demonstrated the great potential of DRL in medicine and healthcare. This paper presents a literature review of DRL in medical imaging. We start with a comprehensive tutorial of DRL, including the latest model-free and model-based algorithms. We then cover existing DRL applications for medical imaging, which are roughly divided into three main categories: (i) parametric medical image analysis tasks including landmark detection, object/lesion detection, registration, and view plane localization; (ii) solving optimization tasks including hyperparameter tuning, selecting augmentation strategies, and neural architecture search; and (iii) miscellaneous applications including surgical gesture segmentation, personalized mobile health intervention, and computational model personalization. The paper concludes with discussions of future perspectives., Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2021 Elsevier B.V. All rights reserved.)

Scarcity of annotated images hampers the building of automated solution for reliable COVID-19 diagnosis and evaluation from CT. To alleviate the burden of data annotation, we herein present a label-free approach for segmenting COVID-19 lesions in CT via voxel-level anomaly modeling that mines out the relevant knowledge from normal CT lung scans. Our modeling is inspired by the observation that the parts of tracheae and vessels, which lay in the high-intensity range where lesions belong to, exhibit strong patterns. To facilitate the learning of such patterns at a voxel level, we synthesize 'lesions' using a set of simple operations and insert the synthesized 'lesions' into normal CT lung scans to form training pairs, from which we learn a normalcy-recognizing network (NormNet) that recognizes normal tissues and separate them from possible COVID-19 lesions. Our experiments on three different public datasets validate the effectiveness of NormNet, which conspicuously outperforms a variety of unsupervised anomaly detection (UAD) methods.

Immunofixation Electrophoresis (IFE) analysis is of great importance to the diagnosis of Multiple Myeloma, which is among the top-9 cancer killers in the United States, but has rarely been studied in the context of deep learning. Two possible reasons are: 1) the recognition of IFE patterns is dependent on the co-location of bands that forms a binary relation, different from the unary relation (visual features to label) that deep learning is good at modeling; 2) deep classification models may perform with high accuracy for IFE recognition but is not able to provide firm evidence (where the co-location patterns are) for its predictions, rendering difficulty for technicians to validate the results. We propose to address these issues with collocative learning, in which a collocative tensor has been constructed to transform the binary relations into unary relations that are compatible with conventional deep networks, and a location-label-free method that utilizes the Grad-CAM saliency map for evidence backtracking has been proposed for accurate localization. In addition, we have proposed Coached Attention Gates that can regulate the inference of the learning to be more consistent with human logic and thus support the evidence backtracking. The experimental results show that the proposed method has obtained a performance gain over its base model ResNet18 by 741.30% in IoU and also outperformed popular deep networks of DenseNet, CBAM, and Inception-v3.

Multimodal image registration has many applications in diagnostic medical imaging and image-guided interventions, such as Transcatheter Arterial Chemoembolization (TACE) of liver cancer guided by intraprocedural CBCT and pre-operative MR. The ability to register peri-procedurally acquired diagnostic images into the intraprocedural environment can potentially improve the intra-procedural tumor targeting, which will significantly improve therapeutic outcomes. However, the intra-procedural CBCT often suffers from suboptimal image quality due to lack of signal calibration for Hounsfield unit, limited FOV, and motion/metal artifacts. These non-ideal conditions make standard intensity-based multimodal registration methods infeasible to generate correct transformation across modalities. While registration based on anatomic structures, such as segmentation or landmarks, provides an efficient alternative, such anatomic structure information is not always available. One can train a deep learning-based anatomy extractor, but it requires large-scale manual annotations on specific modalities, which are often extremely time-consuming to obtain and require expert radiological readers. To tackle these issues, we leverage annotated datasets already existing in a source modality and propose an anatomy-preserving domain adaptation to segmentation network (APA2Seg-Net) for learning segmentation without target modality ground truth. The segmenters are then integrated into our anatomy-guided multimodal registration based on the robust point matching machine. Our experimental results on in-house TACE patient data demonstrated that our APA2Seg-Net can generate robust CBCT and MR liver segmentation, and the anatomy-guided registration framework with these segmenters can provide high-quality multimodal registrations., Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2021 Elsevier B.V. All rights reserved.)

Limited view tomographic reconstruction aims to reconstruct a tomographic image from a limited number of projection views arising from sparse view or limited angle acquisitions that reduce radiation dose or shorten scanning time. However, such a reconstruction suffers from severe artifacts due to the incompleteness of sinogram. To derive quality reconstruction, previous methods use UNet-like neural architectures to directly predict the full view reconstruction from limited view data; but these methods leave the deep network architecture issue largely intact and cannot guarantee the consistency between the sinogram of the reconstructed image and the acquired sinogram, leading to a non-ideal reconstruction. In this work, we propose a cascaded residual dense spatial-channel attention network consisting of residual dense spatial-channel attention networks and projection data fidelity layers. We evaluate our methods on two datasets. Our experimental results on AAPM Low Dose CT Grand Challenge datasets demonstrate that our algorithm achieves a consistent and substantial improvement over the existing neural network methods on both limited angle reconstruction and sparse view reconstruction. In addition, our experimental results on Deep Lesion datasets demonstrate that our method is able to generate high-quality reconstruction for 8 major lesion types.

Annotating multiple organs in medical images is both costly and time-consuming; therefore, existing multi-organ datasets with labels are often low in sample size and mostly partially labeled, that is, a dataset has a few organs labeled but not all organs. In this paper, we investigate how to learn a single multi-organ segmentation network from a union of such datasets. To this end, we propose two types of novel loss function, particularly designed for this scenario: (i) marginal loss and (ii) exclusion loss. Because the background label for a partially labeled image is, in fact, a 'merged' label of all unlabelled organs and 'true' background (in the sense of full labels), the probability of this 'merged' background label is a marginal probability, summing the relevant probabilities before merging. This marginal probability can be plugged into any existing loss function (such as cross entropy loss, Dice loss, etc.) to form a marginal loss. Leveraging the fact that the organs are non-overlapping, we propose the exclusion loss to gauge the dissimilarity between labeled organs and the estimated segmentation of unlabelled organs. Experiments on a union of five benchmark datasets in multi-organ segmentation of liver, spleen, left and right kidneys, and pancreas demonstrate that using our newly proposed loss functions brings a conspicuous performance improvement for state-of-the-art methods without introducing any extra computation., Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2021 Elsevier B.V. All rights reserved.)

Purpose: Pelvic bone segmentation in CT has always been an essential step in clinical diagnosis and surgery planning of pelvic bone diseases. Existing methods for pelvic bone segmentation are either hand-crafted or semi-automatic and achieve limited accuracy when dealing with image appearance variations due to the multi-site domain shift, the presence of contrasted vessels, coprolith and chyme, bone fractures, low dose, metal artifacts, etc. Due to the lack of a large-scale pelvic CT dataset with annotations, deep learning methods are not fully explored., Methods: In this paper, we aim to bridge the data gap by curating a large pelvic CT dataset pooled from multiple sources, including 1184 CT volumes with a variety of appearance variations. Then, we propose for the first time, to the best of our knowledge, to learn a deep multi-class network for segmenting lumbar spine, sacrum, left hip, and right hip, from multiple-domain images simultaneously to obtain more effective and robust feature representations. Finally, we introduce a post-processor based on the signed distance function (SDF)., Results: Extensive experiments on our dataset demonstrate the effectiveness of our automatic method, achieving an average Dice of 0.987 for a metal-free volume. SDF post-processor yields a decrease of 15.1% in Hausdorff distance compared with traditional post-processor., Conclusion: We believe this large-scale dataset will promote the development of the whole community and open source the images, annotations, codes, and trained baseline models at https://github.com/ICT-MIRACLE-lab/CTPelvic1K .

Since its renaissance, deep learning has been widely used in various medical imaging tasks and has achieved remarkable success in many medical imaging applications, thereby propelling us into the so-called artificial intelligence (AI) era. It is known that the success of AI is mostly attributed to the availability of big data with annotations for a single task and the advances in high performance computing. However, medical imaging presents unique challenges that confront deep learning approaches. In this survey paper, we first present traits of medical imaging, highlight both clinical needs and technical challenges in medical imaging, and describe how emerging trends in deep learning are addressing these issues. We cover the topics of network architecture, sparse and noisy labels, federating learning, interpretability, uncertainty quantification, etc. Then, we present several case studies that are commonly found in clinical practice, including digital pathology and chest, brain, cardiovascular, and abdominal imaging. Rather than presenting an exhaustive literature survey, we instead describe some prominent research highlights related to these case study applications. We conclude with a discussion and presentation of promising future directions.

There is clinical evidence that suppressing the bone structures in Chest X-rays (CXRs) improves diagnostic value, either for radiologists or computer-aided diagnosis. However, bone-free CXRs are not always accessible. We hereby propose a coarse-to-fine CXR bone suppression approach by using structural priors derived from unpaired computed tomography (CT) images. In the low-resolution stage, we use the digitally reconstructed radiograph (DRR) image that is computed from CT as a bridge to connect CT and CXR. We then perform CXR bone decomposition by leveraging the DRR bone decomposition model learned from unpaired CTs and domain adaptation between CXR and DRR. To further mitigate the domain differences between CXRs and DRRs and speed up the learning convergence, we perform all the aboved operations in Laplacian of Gaussian (LoG) domain. After obtaining the bone decomposition result in DRR, we upsample it to a high resolution, based on which the bone region in the original high-resolution CXR is cropped and processed to produce a high-resolution bone decomposition result. Finally, such a produced bone image is subtracted from the original high-resolution CXR to obtain the bone suppression result. We conduct experiments and clinical evaluations based on two benchmarking CXR databases to show that (i) the proposed method outperforms the state-of-the-art unsupervised CXR bone suppression approaches; (ii) the CXRs with bone suppression are instrumental to radiologists for reducing their false-negative rate of lung diseases from 15% to 8%; and (iii) state-of-the-art disease classification performances are achieved by learning a deep network that takes the original CXR and its bone-suppressed image as inputs.

Due to the development of deep learning, an increasing number of research works have been proposed to establish automated analysis systems for 3D volumetric medical data to improve the quality of patient care. However, it is challenging to obtain a large number of annotated 3D medical data needed to train a neural network well, as such manual annotation by physicians is time consuming and laborious. Self-supervised learning is one of the potential solutions to mitigate the strong requirement of data annotation by deeply exploiting raw data information. In this paper, we propose a novel self-supervised learning framework for volumetric medical data. Specifically, we propose a pretext task, i.e., Rubik's cube+, to pre-train 3D neural networks. The pretext task involves three operations, namely cube ordering, cube rotating and cube masking, forcing networks to learn translation and rotation invariant features from the original 3D medical data, and tolerate the noise of the data at the same time. Compared to the strategy of training from scratch, fine-tuning from the Rubik's cube+ pre-trained weights can remarkablely boost the accuracy of 3D neural networks on various tasks, such as cerebral hemorrhage classification and brain tumor segmentation, without the use of extra data., (Copyright © 2020 Elsevier B.V. All rights reserved.)

Current deep neural network based approaches to computed tomography (CT) metal artifact reduction (MAR) are supervised methods that rely on synthesized metal artifacts for training. However, as synthesized data may not accurately simulate the underlying physical mechanisms of CT imaging, the supervised methods often generalize poorly to clinical applications. To address this problem, we propose, to the best of our knowledge, the first unsupervised learning approach to MAR. Specifically, we introduce a novel artifact disentanglement network that disentangles the metal artifacts from CT images in the latent space. It supports different forms of generations (artifact reduction, artifact transfer, and self-reconstruction, etc.) with specialized loss functions to obviate the need for supervision with synthesized data. Extensive experiments show that when applied to a synthesized dataset, our method addresses metal artifacts significantly better than the existing unsupervised models designed for natural image-to-image translation problems, and achieves comparable performance to existing supervised models for MAR. When applied to clinical datasets, our method demonstrates better generalization ability over the supervised models. The source code of this paper is publicly available at https:// github.com/liaohaofu/adn.

Routine ultrasound exam in the second and third trimesters of pregnancy involves manually measuring fetal head and brain structures in 2-D scans. The procedure requires a sonographer to find the standardized visualization planes with a probe and manually place measurement calipers on the structures of interest. The process is tedious, time consuming, and introduces user variability into the measurements. This paper proposes an automatic fetal head and brain (AFHB) system for automatically measuring anatomical structures from 3-D ultrasound volumes. The system searches the 3-D volume in a hierarchy of resolutions and by focusing on regions that are likely to be the measured anatomy. The output is a standardized visualization of the plane with correct orientation and centering as well as the biometric measurement of the anatomy. The system is based on a novel framework for detecting multiple structures in 3-D volumes. Since a joint model is difficult to obtain in most practical situations, the structures are detected in a sequence, one-by-one. The detection relies on Sequential Estimation techniques, frequently applied to visual tracking. The interdependence of structure poses and strong prior information embedded in our domain yields faster and more accurate results than detecting the objects individually. The posterior distribution of the structure pose is approximated at each step by sequential Monte Carlo. The samples are propagated within the sequence across multiple structures and hierarchical levels. The probabilistic model helps solve many challenges present in the ultrasound images of the fetus such as speckle noise, signal drop-out, shadows caused by bones, and appearance variations caused by the differences in the fetus gestational age. This is possible by discriminative learning on an extensive database of scans comprising more than two thousand volumes and more than thirteen thousand annotations. The average difference between ground truth and automatic measurements is below 2 mm with a running time of 6.9 s (GPU) or 14.7 s (CPU). The accuracy of the AFHB system is within inter-user variability and the running time is fast, which meets the requirements for clinical use.

Patient-specific orthopedic knee surgery planning requires precisely segmenting from 3D CT images multiple knee bones, namely femur, tibia, fibula, and patella, around the knee joint with severe pathologies. In this work, we propose a fully automated, highly precise, and computationally efficient segmentation approach for multiple bones. First, each bone is initially segmented using a model-based marginal space learning framework for pose estimation followed by non-rigid boundary deformation. To recover shape details, we then refine the bone segmentation using graph cut that incorporates the shape priors derived from the initial segmentation. Finally we remove overlap between neighboring bones using multi-layer graph partition. In experiments, we achieve simultaneous segmentation of femur, tibia, patella, and fibula with an overall accuracy of less than 1mm surface-to-surface error in less than 90s on hundreds of 3D CT scans with pathological knee joints.

The diversity in appearance of diseased lung tissue makes automatic segmentation of lungs from CT with severe pathologies challenging. To overcome this challenge, we rely on contextual constraints from neighboring anatomies to detect and segment lung tissue across a variety of pathologies. We propose an algorithm that combines statistical learning with these anatomical constraints to seek a segmentation of the lung consistent with adjacent structures, such as the heart, liver, spleen, and ribs. We demonstrate that our algorithm reduces the number of failed detections and increases the accuracy of the segmentation on unseen test cases with severe pathologies.

Lymph nodes have high clinical relevance and routinely need to be considered in clinical practice. Automatic detection is, however, challenging due to clutter and low contrast. In this paper, a method is presented that fully automatically detects and segments lymph nodes in 3-D computed tomography images of the chest. Lymph nodes can easily be confused with other structures, it is therefore vital to incorporate as much anatomical prior knowledge as possible in order to achieve a good detection performance. Here, a learned prior of the spatial distribution is used to model this knowledge. Different prior types with increasing complexity are proposed and compared to each other. This is combined with a powerful discriminative model that detects lymph nodes from their appearance. It first generates a number of candidates of possible lymph node center positions. Then, a segmentation method is initialized with a detected candidate. The graph cuts method is adapted to the problem of lymph nodes segmentation. We propose a setting that requires only a single positive seed and at the same time solves the small cut problem of graph cuts. Furthermore, we propose a feature set that is extracted from the segmentation. A classifier is trained on this feature set and used to reject false alarms. Cross-validation on 54 CT datasets showed that for a fixed number of four false alarms per volume image, the detection rate is well more than doubled when using the spatial prior. In total, our proposed method detects mediastinal lymph nodes with a true positive rate of 52.0% at the cost of only 3.1 false alarms per volume image and a true positive rate of 60.9% with 6.1 false alarms per volume image, which compares favorably to prior work on mediastinal lymph node detection., (Copyright © 2012 Elsevier B.V. All rights reserved.)

Ultrasound acquisition is a challenging task that requires simultaneous adjustment of several acquisition parameters (the depth, the focus, the frequency and its operation mode). If the acquisition parameters are not properly chosen, the resulting image will have a poor quality and will degrade the patient diagnosis and treatment workflow. Several hardware-based systems for autotuning the acquisition parameters have been previously proposed, but these solutions were largely abandoned because they failed to properly account for tissue inhomogeneity and other patient-specific characteristics. Consequently, in routine practice the clinician either uses population-based parameter presets or manually adjusts the acquisition parameters for each patient during the scan. In this paper, we revisit the problem of autotuning the acquisition parameters by taking a completely novel approach and producing a solution based on image analytics. Our solution is inspired by the autofocus capability of conventional digital cameras, but is significantly more challenging because the number of acquisition parameters is large and the determination of "good quality" images is more difficult to assess. Surprisingly, we show that the set of acquisition parameters which produce images that are favored by clinicians comprise a 1D manifold, allowing for a real-time optimization to maximize image quality. We demonstrate our method for acquisition parameter autotuning on several live patients, showing that our system can start with a poor initial set of parameters and automatically optimize the parameters to produce high quality images.

We propose a novel framework for rapid and accurate segmentation of a cohort of organs. First, it integrates local and global image context through a product rule to simultaneously detect multiple landmarks on the target organs. The global posterior integrates evidence over all volume patches, while the local image context is modeled with a local discriminative classifier. Through non-parametric modeling of the global posterior, it exploits sparsity in the global context for efficient detection. The complete surface of the target organs is then inferred by robust alignment of a shape model to the resulting landmarks and finally deformed using discriminative boundary detectors. Using our approach, we demonstrate efficient detection and accurate segmentation of liver, kidneys, heart, and lungs in challenging low-resolution MR data in less than one second, and of prostate, bladder, rectum, and femoral heads in CT scans, in roughly one to three seconds and in both cases with accuracy fairly close to inter-user variability.

This paper proposes a fully automatic approach for computing Nuchal Translucency (NT) measurement in an ultrasound scans of the mid-sagittal plane of a fetal head. This is an improvement upon current NT measurement methods which require manual placement of NT measurement points or user-guidance in semi-automatic segmentation of the NT region. The algorithm starts by finding the pose of the fetal head using discriminative learning-based detectors. The fetal head serves as a robust anchoring structure and the NT region is estimated from the statistical relationship between the fetal head and the NT region. Next, the pose of the NT region is locally refined and its inner and outer edge approximately determined via Dijkstra's shortest path applied on the edge-enhanced image. Finally, these two region edges are used to define foreground and background seeds for accurate graph cut segmentation. The NT measurement is computed from the segmented region. Experiments show that the algorithm efficiently and effectively detects the NT region and provides accurate NT measurement which suggests suitability for clinical use.

Lymph nodes are assessed routinely in clinical practice and their size is followed throughout radiation or chemotherapy to monitor the effectiveness of cancer treatment. This paper presents a robust learning-based method for automatic detection and segmentation of solid lymph nodes from CT data, with the following contributions. First, it presents a learning based approach to solid lymph node detection that relies on marginal space learning to achieve great speedup with virtually no loss in accuracy. Second, it presents a computationally efficient segmentation method for solid lymph nodes (LN). Third, it introduces two new sets of features that are effective for LN detection, one that self-aligns to high gradients and another set obtained from the segmentation result. The method is evaluated for axillary LN detection on 131 volumes containing 371 LN, yielding a 83.0% detection rate with 1.0 false positive per volume. It is further evaluated for pelvic and abdominal LN detection on 54 volumes containing 569 LN, yielding a 80.0% detection rate with 3.2 false positives per volume. The running time is 5-20 s per volume for axillary areas and 15-40 s for pelvic. An added benefit of the method is the capability to detect and segment conglomerated lymph nodes.

We present a two-stage method for effective and efficient detection of one or multiple anatomical landmarks in an arbitrary 3D volume. The first stage of nearest neighbor matching is to roughly estimate the landmark locations. It searches out of 100,000 volumes for the closest to an input volume and then transfers landmark annotations to the input. The second stage of submodular optimization is to refine the landmark locations by running discriminative landmark detectors within the search ranges constrained by the first stage results. Further it coordinates multiple detectors with a search strategy optimized on the fly to reduce the overall computation cost arising in a submodular formulation. We validate the accuracy, speed and robustness of our approach by detecting body regions and landmarks in a dataset of 2500 CT scans.

In this paper, we present a novel method by incorporating information theory into the learning-based approach for automatic and accurate pelvic organ segmentation (including the prostate, bladder and rectum). We target 3D CT volumes that are generated using different scanning protocols (e.g., contrast and non-contrast, with and without implant in the prostate, various resolution and position), and the volumes come from largely diverse sources (e.g., diseased in different organs). Three key ingredients are combined to solve this challenging segmentation problem. First, marginal space learning (MSL) is applied to efficiently and effectively localize the multiple organs in the largely diverse CT volumes. Second, learning techniques, steerable features, are applied for robust boundary detection. This enables handling of highly heterogeneous texture pattern. Third, a novel information theoretic scheme is incorporated into the boundary inference process. The incorporation of the Jensen-Shannon divergence further drives the mesh to the best fit of the image, thus improves the segmentation performance. The proposed approach is tested on a challenging dataset containing 188 volumes from diverse sources. Our approach not only produces excellent segmentation accuracy, but also runs about eighty times faster than previous state-of-the-art solutions. The proposed method can be applied to CT images to provide visual guidance to physicians during the computer-aided diagnosis, treatment planning and image-guided radiotherapy to treat cancers in pelvic region.

Being able to segment the esophagus without user interaction from 3-D CT data is of high value to radiologists during oncological examinations of the mediastinum. The segmentation can serve as a guideline and prevent confusion with pathological tissue. However, limited contrast to surrounding structures and versatile shape and appearance make segmentation a challenging problem. This paper presents a multistep method. First, a detector that is trained to learn a discriminative model of the appearance is combined with an explicit model of the distribution of respiratory and esophageal air. In the next step, prior shape knowledge is incorporated using a Markov chain model. We follow a "detect and connect" approach to obtain the maximum a posteriori estimate of the approximate esophagus shape from hypothesis about the esophagus contour in axial image slices. Finally, the surface of this approximation is nonrigidly deformed to better fit the boundary of the organ. The method is compared to an alternative approach that uses a particle filter instead of a Markov chain to infer the approximate esophagus shape, to the performance of a human observer and also to state of the art methods, which are all semiautomatic. Cross-validation on 144 CT scans showed that the Markov chain based approach clearly outperforms the particle filter. It segments the esophagus with a mean error of 1.80 mm in less than 16 s on a standard PC. This is only 1 mm above the interobserver variability and can compete with the results of previously published semiautomatic methods.

In this paper, a method is described to automatically estimate the visible body region of a computed tomography (CT) volume image. In order to quantify the body region, a body coordinate (BC) axis is used that runs in longitudinal direction. Its origin and unit length are patient-specific and depend on anatomical landmarks. The body region of a test volume is estimated by registering it only along the longitudinal axis to a set of reference CT volume images with known body coordinates. During these 1D registrations, an axial image slice of the test volume is compared to an axial slice of a reference volume by extracting a descriptor from both slices and measuring the similarity of the descriptors. A slice descriptor consists of histograms of visual words. Visual words are code words of a quantized feature space and can be thought of as classes of image patches with similar appearance. A slice descriptor is formed by sampling a slice on a regular 2D grid and extracting a Speeded Up Robust Features (SURF) descriptor at each sample point. The codebook, or visual vocabulary, is generated in a training step by clustering SURF descriptors. Each SURF descriptor extracted from a slice is classified into the closest visual word (or cluster center) and counted in a histogram. A slice is finally described by a spatial pyramid of such histograms. We introduce an extension of the SURF descriptors to an arbitrary number of dimensions (N-SURF). Here, we make use of 2-SURF and 3-SURF descriptors. Cross-validation on 84 datasets shows the robustness of the results. The body portion can be estimated with an average error of 15.5mm within 9s. Possible applications of this method are automatic labeling of medical image databases and initialization of subsequent image analysis algorithms., (Copyright © 2010 Elsevier Ltd. All rights reserved.)

Cardiac computed tomography (CT) is the primary noninvasive imaging modality to diagnose coronary artery disease. Though various methods have been proposed for coronary artery segmentation, most rely on at least one user click to provide a seed point for initialization. Automatic detection of the coronary ostia (where coronaries originate from the aorta), including both the native coronary ostia and graft ostia of the bypass coronaries, can make the whole coronary exam workflow fully automatic, therefore increasing a physician's throughput. Anatomical structures (native coronary ostia) and pathological structures (graft ostia) often require significantly different detection methods. The native coronary ostia are well constrained by the surrounding structures, therefore are detected as a global object. Detecting the graft ostia is far more difficult due to the large variation in graft position. A new searching strategy is proposed to efficiently guide the focus of analysis and, at the same time, reduce the false positive detections. Since the bypass coronaries are grafted on the ascending aorta surface, the ascending aorta is first segmented to constrain the search. The quantitative prior distribution of the graft ostia on the aorta surface is learned from a training set to significantly reduce the searching space further. Efficient local image features are extracted around each candidate point on the aorta surface to train a detector. The proposed method is computationally efficient, taking about 0.40 seconds to detect both native and graft ostia in a volume with around 512 x 512 x 200 voxels.

We propose an automatic algorithm for phase labeling that relies on the intensity changes in anatomical regions due to the contrast agent propagation. The regions (specified by aorta, vena cava, liver, and kidneys) are first detected by a robust learning-based discriminative algorithm. The intensities inside each region are then used in multi-class LogitBoost classifiers to independently estimate the contrast phase. Each classifier forms a node in a decision tree which is used to obtain the final phase label. Combining independent classification from multiple regions in a tree has the advantage when one of the region detectors fail or when the phase training example database is imbalanced. We show on a dataset of 1016 volumes that the system correctly classifies native phase in 96.2% of the cases, hepatic dominant phase (92.2%), hepatic venous phase (96.7%), and equilibrium phase (86.4%) in 7 seconds on average.

Recently conducted clinical studies prove the utility of Coronary Computed Tomography Angiography (CCTA) as a viable alternative to invasive angiography for the detection of Coronary Artery Disease (CAD). This has lead to the development of several algorithms for automatic detection and grading of coronary stenoses. However, most of these methods focus on detecting calcified plaques only. A few methods that can also detect and grade non-calcified plaques require substantial user involvement. In this paper, we propose a fast and fully automatic system that is capable of detecting, grading and classifying coronary stenoses in CCTA caused by all types of plaques. We propose a four-step approach including a learning-based centerline verification step and a lumen cross-section estimation step using random regression forests. We show state-of-the-art performance of our method in experiments conducted on a set of 229 CCTA volumes. With an average processing time of 1.8 seconds per case after centerline extraction, our method is significantly faster than competing approaches.

As a minimally invasive surgery to treat left atrial (LA) fibrillation, catheter based ablation uses high radio-frequency energy to eliminate potential sources of the abnormal electrical events, especially around the ostia of pulmonary veins (PV). Due to large structural variations of the PV drainage pattern, a personalized LA model is helpful to translate a generic ablation strategy to a specific patient's anatomy. Overlaying the LA model onto 2D fluoroscopic images provides valuable visual guidance during surgery. A holistic shape model is not accurate enough to represent the whole shape population of the LA. In this paper, we propose a part based LA model (including the chamber, appendage, and four major PVs) and each part is a much simpler anatomical structure compared to the holistic one. Our approach works on un-gated C-arm CT, where thin boundaries between the LA blood pool and surrounding tissues are often blurred due to the cardiac motion artifacts (which presents a big challenge compared to the highly contrasted gated CT/MRI). To avoid segmentation leakage, the shape prior is exploited in a model based approach to segment the LA parts. However, independent detection of each part is not optimal and its robustness needs further improvement (especially for the appendage and PVs). We propose to enforce a statistical shape constraint during the estimation of pose parameters (position, orientation, and size) of different parts. Our approach is computationally efficient, taking about 1.5 s to process a volume with 256 x 256 x 250 voxels. Experiments on 469 C-arm CT datasets demonstrate its robustness.

Simple algorithms for segmenting healthy lung parenchyma in CT are unable to deal with high density tissue common in pulmonary diseases. To overcome this problem, we propose a multi-stage learning-based approach that combines anatomical information to predict an initialization of a statistical shape model of the lungs. The initialization first detects the carina of the trachea, and uses this to detect a set of automatically selected stable landmarks on regions near the lung (e.g., ribs, spine). These landmarks are used to align the shape model, which is then refined through boundary detection to obtain fine-grained segmentation. Robustness is obtained through hierarchical use of discriminative classifiers that are trained on a range of manually annotated data of diseased and healthy lungs. We demonstrate fast detection (35s per volume on average) and segmentation of 2 mm accuracy on challenging data.