IEEE transactions on medical imaging最新文献

Analysis of functional connectivity networks (FCNs) derived from resting-state functional magnetic resonance imaging (rs-fMRI) has greatly advanced our understanding of brain diseases, including Alzheimer's disease (AD) and attention deficit hyperactivity disorder (ADHD). Advanced machine learning techniques, such as convolutional neural networks (CNNs), have been used to learn high-level feature representations of FCNs for automated brain disease classification. Even though convolution operations in CNNs are good at extracting local properties of FCNs, they generally cannot well capture global temporal representations of FCNs. Recently, the transformer technique has demonstrated remarkable performance in various tasks, which is attributed to its effective self-attention mechanism in capturing the global temporal feature representations. However, it cannot effectively model the local network characteristics of FCNs. To this end, in this paper, we propose a novel network structure for Local sequential feature Coupling Global representation learning (LCGNet) to take advantage of convolutional operations and self-attention mechanisms for enhanced FCN representation learning. Specifically, we first build a dynamic FCN for each subject using an overlapped sliding window approach. We then construct three sequential components (i.e., edge-to-vertex layer, vertex-to-network layer, and network-to-temporality layer) with a dual backbone branch of CNN and transformer to extract and couple from local to global topological information of brain networks. Experimental results on two real datasets (i.e., ADNI and ADHD-200) with rs-fMRI data show the superiority of our LCGNet.

Magnetic Particle Imaging (MPI) is an emerging tomographic modality that allows for precise three-dimensional (3D) mapping of magnetic nanoparticles (MNPs) concentration and distribution. Although significant progress has been made towards improving MPI since its introduction, scaling it up for human applications has proven challenging. High-quality images have been obtained in animal-scale MPI scanners with gradients up to 7 T/m/μ₀, however, for MPI systems with bore diameters around 200 mm the gradients generated by electromagnets drop significantly to below 0.5 T/m/μ₀. Given the current technological limitations in image reconstruction and the properties of available MNPs, these low gradients inherently impose limitations on improving MPI resolution for higher precision medical imaging. Utilizing superconductors stands out as a promising approach for developing a human-scale MPI system. In this study, we introduce, for the first time, a human-scale amplitude-modulated (AM) MPI system with superconductor-based selection coils. The system achieves an unprecedented magnetic field gradient of up to 2.5 T/m/μ₀ within a 200 mm bore diameter, enabling large fields of view of 100 × 130 × 98 mm³ at 2.5 T/m/μ₀ for 3D imaging. While obtained spatial resolution is in the order of previous animal-scale AM MPIs, incorporating superconductors for achieving such high gradients in a 200 mm bore diameter marks a major step toward clinical MPI.

Cervical cytology is a critical screening strategy for early detection of pre-cancerous and cancerous cervical lesions. The challenge lies in accurately classifying various cervical cytology cell types. Existing automated cervical cytology methods are primarily trained on databases covering a narrow range of coarse-grained cell types, which fail to provide a comprehensive and detailed performance analysis that accurately represents real-world cytopathology conditions. To overcome these limitations, we introduce HiCervix, the most extensive, multi-center cervical cytology dataset currently available to the public. HiCervix includes 40,229 cervical cells from 4,496 whole slide images, categorized into 29 annotated classes. These classes are organized within a three-level hierarchical tree to capture fine-grained subtype information. To exploit the semantic correlation inherent in this hierarchical tree, we propose HierSwin, a hierarchical vision transformer-based classification network. HierSwin serves as a benchmark for detailed feature learning in both coarse-level and fine-level cervical cancer classification tasks. In our comprehensive experiments, HierSwin demonstrated remarkable performance, achieving 92.08% accuracy for coarse-level classification and 82.93% accuracy averaged across all three levels. When compared to board-certified cytopathologists, HierSwin achieved high classification performance (0.8293 versus 0.7359 averaged accuracy), highlighting its potential for clinical applications. This newly released HiCervix dataset, along with our benchmark HierSwin method, is poised to make a substantial impact on the advancement of deep learning algorithms for rapid cervical cancer screening and greatly improve cancer prevention and patient outcomes in real-world clinical settings.

Pulmonary Tuberculosis (PTB) is one of the world's most infectious illnesses, and its early detection is critical for preventing PTB. Digital Radiography (DR) has been the most common and effective technique to examine PTB. However, due to the variety and weak specificity of phenotypes on DR chest X-ray (DCR), it is difficult to make reliable diagnoses for radiologists. Although artificial intelligence technology has made considerable gains in assisting the diagnosis of PTB, it lacks methods to identify the lesions of PTB with few-shot classes and small objects. To solve these problems, geometric data augmentation was used to increase the size of the DCRs. For this purpose, a diffusion probability model was implemented for six few-shot classes. Importantly, we propose a new multi-lesion detector PtbNet based on RetinaNet, which was constructed to detect small objects of PTB lesions. The results showed that by two data augmentations, the number of DCRs increased by 80% from 570 to 2,859. In the pre-evaluation experiments with the baseline, RetinaNet, the AP improved by 9.9 for six few-shot classes. Our extensive empirical evaluation showed that the AP of PtbNet achieved 28.2, outperforming the other 9 state-of-the-art methods. In the ablation study, combined with BiFPN+ and PSPD-Conv, the AP increased by 2.1, AP^s increased by 5.0, and grew by an average of 9.8 in AP^m and AP^l. In summary, PtbNet not only improves the detection of small-object lesions but also enhances the ability to detect different types of PTB uniformly, which helps physicians diagnose PTB lesions accurately. The code is available at https://github.com/Wenhui-person/PtbNet/tree/master.

Intrathoracic airway segmentation in computed tomography is a prerequisite for various respiratory disease analyses such as chronic obstructive pulmonary disease, asthma and lung cancer. Due to the low imaging contrast and noises execrated at peripheral branches, the topological-complexity and the intra-class imbalance of airway tree, it remains challenging for deep learning-based methods to segment the complete airway tree (on extracting deeper branches). Unlike other organs with simpler shapes or topology, the airway's complex tree structure imposes an unbearable burden to generate the "ground truth" label (up to 7 or 3 hours of manual or semi-automatic annotation per case). Most of the existing airway datasets are incompletely labeled/annotated, thus limiting the completeness of computer-segmented airway. In this paper, we propose a new anatomy-aware multi-class airway segmentation method enhanced by topology-guided iterative self-learning. Based on the natural airway anatomy, we formulate a simple yet highly effective anatomy-aware multi-class segmentation task to intuitively handle the severe intra-class imbalance of the airway. To solve the incomplete labeling issue, we propose a tailored iterative self-learning scheme to segment toward the complete airway tree. For generating pseudo-labels to achieve higher sensitivity (while retaining similar specificity), we introduce a novel breakage attention map and design a topology-guided pseudo-label refinement method by iteratively connecting breaking branches commonly existed from initial pseudo-labels. Extensive experiments have been conducted on four datasets including two public challenges. The proposed method achieves the top performance in both EXACT'09 challenge using average score and ATM'22 challenge on weighted average score. In a public BAS dataset and a private lung cancer dataset, our method significantly improves previous leading approaches by extracting at least (absolute) 6.1% more detected tree length and 5.2% more tree branches, while maintaining comparable precision.

Available evidence suggests that dynamic functional connectivity can capture time-varying abnormalities in brain activity in resting-state cerebral functional magnetic resonance imaging (rs-fMRI) data and has a natural advantage in uncovering mechanisms of abnormal brain activity in schizophrenia (SZ) patients. Hence, an advanced dynamic brain network analysis model called the temporal brain category graph convolutional network (Temporal-BCGCN) was employed. Firstly, a unique dynamic brain network analysis module, DSF-BrainNet, was designed to construct dynamic synchronization features. Subsequently, a revolutionary graph convolution method, TemporalConv, was proposed based on the synchronous temporal properties of features. Finally, the first modular test tool for abnormal hemispherical lateralization in deep learning based on rs-fMRI data, named CategoryPool, was proposed. This study was validated on COBRE and UCLA datasets and achieved 83.62% and 89.71% average accuracies, respectively, outperforming the baseline model and other state-of-theart methods. The ablation results also demonstrate the advantages of TemporalConv over the traditional edge feature graph convolution approach and the improvement of CategoryPool over the classical graph pooling approach. Interestingly, this study showed that the lower-order perceptual system and higher-order network regions in the left hemisphere are more severely dysfunctional than in the right hemisphere in SZ, reaffirmings the importance of the left medial superior frontal gyrus in SZ. Our code was available at: https://github.com/swfen/Temporal-BCGCN.

Multi-modal prompt learning is a high-performance and cost-effective learning paradigm, which learns text as well as image prompts to tune pre-trained vision-language (V-L) models like CLIP for adapting multiple downstream tasks. However, recent methods typically treat text and image prompts as independent components without considering the dependency between prompts. Moreover, extending multi-modal prompt learning into the medical field poses challenges due to a significant gap between general- and medical-domain data. To this end, we propose a Multi-modal Collaborative Prompt Learning (MCPL) pipeline to tune a frozen V-L model for aligning medical text-image representations, thereby achieving medical downstream tasks. We first construct the anatomy-pathology (AP) prompt for multi-modal prompting jointly with text and image prompts. The AP prompt introduces instance-level anatomy and pathology information, thereby making a V-L model better comprehend medical reports and images. Next, we propose graph-guided prompt collaboration module (GPCM), which explicitly establishes multi-way couplings between the AP, text, and image prompts, enabling collaborative multi-modal prompt producing and updating for more effective prompting. Finally, we develop a novel prompt configuration scheme, which attaches the AP prompt to the query and key, and the text/image prompt to the value in self-attention layers for improving the interpretability of multi-modal prompts. Extensive experiments on numerous medical classification and object detection datasets show that the proposed pipeline achieves excellent effectiveness and generalization. Compared with state-of-the-art prompt learning methods, MCPL provides a more reliable multi-modal prompt paradigm for reducing tuning costs of V-L models on medical downstream tasks. Our code: https://github.com/CUHK-AIM-Group/MCPL.