IEEE transactions on medical imaging最新文献

In the field of medical Vision-Language Pretraining (VLP), significant efforts have been devoted to deriving text and image features from both clinical reports and associated medical images. However, most existing methods may have overlooked the opportunity in leveraging the inherent hierarchical structure of clinical reports, which are generally split into 'findings' for descriptive content and 'impressions' for conclusive observation. Instead of utilizing this rich, structured format, current medical VLP approaches often simplify the report into either a unified entity or fragmented tokens. In this work, we propose a novel clinical prior guided VLP framework named IMITATE to learn the structure information from medical reports with hierarchical vision-language alignment. The framework derives multi-level visual features from the chest X-ray (CXR) images and separately aligns these features with the descriptive and the conclusive text encoded in the hierarchical medical report. Furthermore, a new clinical-informed contrastive loss is introduced for cross-modal learning, which accounts for clinical prior knowledge in formulating sample correlations in contrastive learning. The proposed model, IMITATE, outperforms baseline VLP methods across six different datasets, spanning five medical imaging downstream tasks. Comprehensive experimental results highlight the advantages of integrating the hierarchical structure of medical reports for vision-language alignment.

Reducing the dose of radiation in computed tomography (CT) is vital to decreasing secondary cancer risk. However, the use of low-dose CT (LDCT) images is accompanied by increased noise that can negatively impact diagnoses. Although numerous deep learning algorithms have been developed for LDCT denoising, several challenges persist, including the visual incongruence experienced by radiologists, unsatisfactory performances across various metrics, and insufficient exploration of the networks' robustness in other CT domains. To address such issues, this study proposes three novel accretions. First, we propose a generative adversarial network (GAN) with a robust discriminator through multi-task learning that simultaneously performs three vision tasks: restoration, image-level, and pixel-level decisions. The more multi-tasks that are performed, the better the denoising performance of the generator, which means multi-task learning enables the discriminator to provide more meaningful feedback to the generator. Second, two regulatory mechanisms, restoration consistency (RC) and non-difference suppression (NDS), are introduced to improve the discriminator's representation capabilities. These mechanisms eliminate irrelevant regions and compare the discriminator's results from the input and restoration, thus facilitating effective GAN training. Lastly, we incorporate residual fast Fourier transforms with convolution (Res-FFT-Conv) blocks into the generator to utilize both frequency and spatial representations. This approach provides mixed receptive fields by using spatial (or local), spectral (or global), and residual connections. Our model was evaluated using various pixel- and feature-space metrics in two denoising tasks. Additionally, we conducted visual scoring with radiologists. The results indicate superior performance in both quantitative and qualitative measures compared to state-of-the-art denoising techniques.

CT-based bronchial tree analysis is a key step for the diagnosis of lung and airway diseases. However, the topology of bronchial trees varies across individuals, which presents a challenge to the automatic bronchus classification. To solve this issue, we propose the Bronchus Classification Network (BCNet), a structure-guided framework that exploits the segment-level topological information using point clouds to learn the voxel-level features. BCNet has two branches, a Point-Voxel Graph Neural Network (PV-GNN) for segment classification, and a Convolutional Neural Network (CNN) for voxel labeling. The two branches are simultaneously trained to learn topology-aware features for their shared backbone while it is feasible to run only the CNN branch for the inference. Therefore, BCNet maintains the same inference efficiency as its CNN baseline. Experimental results show that BCNet significantly exceeds the state-of-the-art methods by over 8.0% both on F1-score for classifying bronchus. Furthermore, we contribute BronAtlas: an open-access benchmark of bronchus imaging analysis with high-quality voxel-wise annotations of both anatomical and abnormal bronchial segments. The benchmark is available at link¹.

Multiple instance learning (MIL) based whole slide image (WSI) classification is often carried out on the representations of patches extracted from WSI with a pre-trained patch encoder. The performance of classification relies on both patch-level representation learning and MIL classifier training. Most MIL methods utilize a frozen model pre-trained on ImageNet or a model trained with self-supervised learning on histopathology image dataset to extract patch image representations and then fix these representations in the training of the MIL classifiers for efficiency consideration. However, the invariance of representations cannot meet the diversity requirement for training a robust MIL classifier, which has significantly limited the performance of the WSI classification. In this paper, we propose a Self-Supervised Representation Distribution Learning framework (SSRDL) for patch-level representation learning with an online representation sampling strategy (ORS) for both patch feature extraction and WSI-level data augmentation. The proposed method was evaluated on three datasets under three MIL frameworks. The experimental results have demonstrated that the proposed method achieves the best performance in histopathology image representation learning and data augmentation and outperforms state-of-the-art methods under different WSI classification frameworks. The code is available at https://github.com/lazytkm/SSRDL.

Many neuropsychiatric disorders are considered to be associated with abnormalities in the functional connectivity networks of the brain. The research on the classification of functional connectivity can therefore provide new perspectives for understanding the pathology of disorders and contribute to early diagnosis and treatment. Functional connectivity exhibits a nature of dynamically changing over time, however, the majority of existing methods are unable to collectively reveal the spatial topology and time-varying characteristics. Furthermore, despite the efforts of limited spatial-temporal studies to capture rich information across different spatial scales, they have not delved into the temporal characteristics among different scales. To address above issues, we propose a novel Multi-Scale Spatial-Temporal Attention Networks (MSSTAN) to exploit the multi-scale spatial-temporal information provided by functional connectome for classification. To fully extract spatial features of brain regions, we propose a Topology Enhanced Graph Transformer module to guide the attention calculations in the learning of spatial features by incorporating topology priors. A Multi-Scale Pooling Strategy is introduced to obtain representations of brain connectome at various scales. Considering the temporal dynamic characteristics between dynamic functional connectome, we employ Locality Sensitive Hashing attention to further capture long-term dependencies in time dynamics across multiple scales and reduce the computational complexity of the original attention mechanism. Experiments on three brain fMRI datasets of MDD and ASD demonstrate the superiority of our proposed approach. In addition, benefiting from the attention mechanism in Transformer, our results are interpretable, which can contribute to the discovery of biomarkers. The code is available at https://github.com/LIST-KONG/MSSTAN.

CT and MR are currently the most common imaging techniques for pancreatic cancer diagnosis. Accurate segmentation of the pancreas in CT and MR images can provide significant help in the diagnosis and treatment of pancreatic cancer. Traditional supervised segmentation methods require a large number of labeled CT and MR training data, which is usually time-consuming and laborious. Meanwhile, due to domain shift, traditional segmentation networks are difficult to be deployed on different imaging modality datasets. Cross-domain segmentation can utilize labeled source domain data to assist unlabeled target domains in solving the above problems. In this paper, a cross-domain pancreas segmentation algorithm is proposed based on Moment-Consistent Contrastive Cycle Generative Adversarial Networks (MC-CCycleGAN). MC-CCycleGAN is a style transfer network, in which the encoder of its generator is used to extract features from real images and style transfer images, constrain feature extraction through a contrastive loss, and fully extract structural features of input images during style transfer while eliminate redundant style features. The multi-order central moments of the pancreas are proposed to describe its anatomy in high dimensions and a contrastive loss is also proposed to constrain the moment consistency, so as to maintain consistency of the pancreatic structure and shape before and after style transfer. Multi-teacher knowledge distillation framework is proposed to transfer the knowledge from multiple teachers to a single student, so as to improve the robustness and performance of the student network. The experimental results have demonstrated the superiority of our framework over state-of-the-art domain adaptation methods.

Histological image registration is a fundamental task in histological image analysis. It is challenging because of substantial appearance differences due to multiple staining. Keypoint correspondences, i.e., matched keypoint pairs, have been introduced to guide unsupervised deep learning (DL) based registration methods to handle such a registration task. This paper proposes an iterative keypoint correspondence-guided (IKCG) unsupervised network for non-rigid histological image registration. Fixed deep features and learnable deep features are introduced as keypoint descriptors to automatically establish keypoint correspondences, the distance between which is used as a loss function to train the registration network. Fixed deep features extracted from DL networks that are pre-trained on natural image datasets are more discriminative than handcrafted ones, benefiting from the deep and hierarchical nature of DL networks. The intermediate layer outputs of the registration networks trained on histological image datasets are extracted as learnable deep features, which reveal unique information for histological images. An iterative training strategy is adopted to train the registration network and optimize learnable deep features jointly. Benefiting from the excellent matching ability of learnable deep features optimized with the iterative training strategy, the proposed method can solve the local non-rigid large displacement problem, an inevitable problem usually caused by misoperation, such as tears in producing tissue slices. The proposed method is evaluated on the Automatic Non-rigid Histology Image Registration (ANHIR) website and AutomatiC Registration Of Breast cAncer Tissue (ACROBAT) website. It ranked 1st on both websites as of August 6th, 2024.

Microwave-induced thermoacoustic imaging (M-TAI) allows the visualization of macroscopic and microscopic structures of bio-tissues. However, it suffers from severe inherent artifacts that might misguide the subsequent diagnostics and treatments of diseases. To overcome this limitation, we propose an optimized excitation strategy. In detail, the strategy integrates dynamically compound specific absorption rate (SAR) and co-planar configuration of polarization state, incident wave vector and imaging plane. Starting from the theoretical analysis, we interpret the underlying mechanism supporting the superiority of the optimized excitation strategy to achieve an effect equivalent to homogenizing the deposited electromagnetic energy in bio-tissues. The following numerical simulations demonstrate that the strategy enables better preservation of the conductivity weighting of samples while increasing Pearson correlation coefficient. Furthermore, the in vitro and in vivo M-TAI experiments validate the effectiveness and robustness of this optimized excitation strategy in artifact suppression, allowing the simultaneous identification of both boundary and inside fine structures within bio-tissues. All the results suggest that the optimized excitation strategy can be expanded to diverse scenarios, inspiring more suitable strategies that remarkably suppress the inherent artifacts in M-TAI.

In digital pathology, whole slide images (WSI) are crucial for cancer prognostication and treatment planning. WSI classification is generally addressed using multiple instance learning (MIL), alleviating the challenge of processing billions of pixels and curating rich annotations. Though recent MIL approaches leverage variants of the attention mechanism to learn better representations, they scarcely study the properties of the data distribution itself i.e., different staining and acquisition protocols resulting in intra-patch and inter-slide variations. In this work, we first introduce a distribution re-calibration strategy to shift the feature distribution of a WSI bag (instances) using the statistics of the max-instance (critical) feature. Second, we enforce class (bag) separation via a metric loss assuming that positive bags exhibit larger magnitudes than negatives. We also introduce a generative process leveraging Vector Quantization (VQ) for improved instance discrimination i.e., VQ helps model bag latent factors for improved classification. To model spatial and context information, a position encoding module (PEM) is employed with transformer-based pooling by multi-head self-attention (PMSA). Evaluation of popular WSI benchmark datasets reveals our approach improves over state-of-the-art MIL methods. Further, we validate the general applicability of our method on classic MIL benchmark tasks and for point cloud classification with limited points https://github.com/PhilipChicco/FRMIL.

In multi-modal magnetic resonance imaging (MRI), the tasks of imputing or reconstructing the target modality share a common obstacle: the accurate modeling of fine-grained inter-modal differences, which has been sparingly addressed in current literature. These differences stem from two sources: 1) spatial misalignment remaining after coarse registration and 2) structural distinction arising from modality-specific signal manifestations. This paper integrates the previously separate research trajectories of cross-modality synthesis (CMS) and multi-contrast super-resolution (MCSR) to address this pervasive challenge within a unified framework. Connected through generalized down-sampling ratios, this unification not only emphasizes their common goal in reducing structural differences, but also identifies the key task distinguishing MCSR from CMS: modeling the structural distinctions using the limited information from the misaligned target input. Specifically, we propose a composite network architecture with several key components: a label correction module to align the coordinates of multi-modal training pairs, a CMS module serving as the base model, an SR branch to handle target inputs, and a difference projection discriminator for structural distinction-centered adversarial training. When training the SR branch as the generator, the adversarial learning is enhanced with distinction-aware incremental modulation to ensure better-controlled generation. Moreover, the SR branch integrates deformable convolutions to address cross-modal spatial misalignment at the feature level. Experiments conducted on three public datasets demonstrate that our approach effectively balances structural accuracy and realism, exhibiting overall superiority in comprehensive evaluations for both tasks over current state-of-the-art approaches. The code is available at https://github.com/papshare/FGDL.