Medical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention最新文献

During a fetal ultrasound scan, a sonographer will zoom in and zoom out as they attempt to get clearer images of the anatomical structures of interest. This paper explores how to use this zoom information which is an under-utilised piece of information that is extractable from fetal ultrasound images. We explore associating zooming patterns to specific structures. The presence of such patterns would indicate that each individual anatomical structure has a unique signature associated with it, thereby allowing for classification of fetal ultrasound clips without directly reading the actual fetal ultrasound images in a convolutional neural network.

Supervised methods for 3D anatomy segmentation demonstrate superior performance but are often limited by the availability of annotated data. This limitation has led to a growing interest in self-supervised approaches in tandem with the abundance of available unannotated data. Slice propagation has emerged as a self-supervised approach that leverages slice registration as a self-supervised task to achieve full anatomy segmentation with minimal supervision. This approach significantly reduces the need for domain expertise, time, and the cost associated with building fully annotated datasets required for training segmentation networks. However, this shift toward reduced supervision via deterministic networks raises concerns about the trustworthiness and reliability of predictions, especially when compared with more accurate supervised approaches. To address this concern, we propose integrating calibrated uncertainty quantification (UQ) into slice propagation methods, which would provide insights into the model's predictive reliability and confidence levels. Incorporating uncertainty measures enhances user confidence in self-supervised approaches, thereby improving their practical applicability. We conducted experiments on three datasets for 3D abdominal segmentation using five UQ methods. The results illustrate that incorporating UQ improves not only model trustworthiness but also segmentation accuracy. Furthermore, our analysis reveals various failure modes of slice propagation methods that might not be immediately apparent to end-users. This study opens up new research avenues to improve the accuracy and trustworthiness of slice propagation methods.

Tagged magnetic resonance imaging (MRI) has been successfully used to track the motion of internal tissue points within moving organs. Typically, to analyze motion using tagged MRI, cine MRI data in the same coordinate system are acquired, incurring additional time and costs. Consequently, tagged-to-cine MR synthesis holds the potential to reduce the extra acquisition time and costs associated with cine MRI, without disrupting downstream motion analysis tasks. Previous approaches have processed each frame independently, thereby overlooking the fact that complementary information from occluded regions of the tag patterns could be present in neighboring frames exhibiting motion. Furthermore, the inconsistent visual appearance, e.g., tag fading, across frames can reduce synthesis performance. To address this, we propose an efficient framework for tagged-to-cine MR sequence synthesis, leveraging both spatial and temporal information with relatively limited data. Specifically, we follow a split-and-integral protocol to balance spatialtemporal modeling efficiency and consistency. The light spatial-temporal transformer (LiST²) is designed to exploit the local and global attention in motion sequence with relatively lightweight training parameters. The directional product relative position-time bias is adapted to make the model aware of the spatial-temporal correlation, while the shifted window is used for motion alignment. Then, a recurrent sliding fine-tuning (ReST) scheme is applied to further enhance the temporal consistency. Our framework is evaluated on paired tagged and cine MRI sequences, demonstrating superior performance over comparison methods.

Asymptomatic neurocognitive impairment (ANI) is a predominant form of cognitive impairment among individuals infected with human immunodeficiency virus (HIV). The current diagnostic criteria for ANI primarily rely on subjective clinical assessments, possibly leading to different interpretations among clinicians. Some recent studies leverage structural or functional MRI containing objective biomarkers for ANI analysis, offering clinicians companion diagnostic tools. However, they mainly utilize a single imaging modality, neglecting complementary information provided by structural and functional MRI. To this end, we propose an attention-enhanced structural and functional MRI fusion (ASFF) framework for HIV-associated ANI analysis. Specifically, the ASFF first extracts data-driven and human-engineered features from structural MRI, and also captures functional MRI features via a graph isomorphism network and Transformer. A mutual cross-attention fusion module is then designed to model the underlying relationship between structural and functional MRI. Additionally, a semantic inter-modality constraint is introduced to encourage consistency of multimodal features, facilitating effective feature fusion. Experimental results on 137 subjects from an HIV-associated ANI dataset with T1-weighted MRI and resting-state functional MRI show the effectiveness of our ASFF in ANI identification. Furthermore, our method can identify both modality-shared and modality-specific brain regions, which may advance our understanding of the structural and functional pathology underlying ANI.

Graph neural networks (GNNs) are proficient machine learning models in handling irregularly structured data. Nevertheless, their generic formulation falls short when applied to the analysis of brain connectomes in Alzheimer's Disease (AD), necessitating the incorporation of domain-specific knowledge to achieve optimal model performance. The integration of AD-related expertise into GNNs presents a significant challenge. Current methodologies reliant on manual design often demand substantial expertise from external domain specialists to guide the development of novel models, thereby consuming considerable time and resources. To mitigate the need for manual curation, this paper introduces a novel self-guided knowledge-infused multimodal GNN to autonomously integrate domain knowledge into the model development process. We propose to conceptualize existing domain knowledge as natural language, and devise a specialized multimodal GNN framework tailored to leverage this uncurated knowledge to direct the learning of the GNN submodule, thereby enhancing its efficacy and improving prediction interpretability. To assess the effectiveness of our framework, we compile a comprehensive literature dataset comprising recent peer-reviewed publications on AD. By integrating this literature dataset with several real-world AD datasets, our experimental results illustrate the effectiveness of the proposed method in extracting curated knowledge and offering explanations on graphs for domain-specific applications. Furthermore, our approach successfully utilizes the extracted information to enhance the performance of the GNN.

The MRI-derived brain network serves as a pivotal instrument in elucidating both the structural and functional aspects of the brain, encompassing the ramifications of diseases and developmental processes. However, prevailing methodologies, often focusing on synchronous BOLD signals from functional MRI (fMRI), may not capture directional influences among brain regions and rarely tackle temporal functional dynamics. In this study, we first construct the brain-effective network via the dynamic causal model. Subsequently, we introduce an interpretable graph learning framework termed Spatio-Temporal Embedding ODE (STE-ODE). This framework incorporates specifically designed directed node embedding layers, aiming at capturing the dynamic inter-play between structural and effective networks via an ordinary differential equation (ODE) model, which characterizes spatial-temporal brain dynamics. Our framework is validated on several clinical phenotype prediction tasks using two independent publicly available datasets (HCP and OASIS). The experimental results clearly demonstrate the advantages of our model compared to several state-of-the-art methods.

Multiple imaging modalities are often used for disease diagnosis, prediction, or population-based analyses. However, not all modalities might be available due to cost, different study designs, or changes in imaging technology. If the differences between the types of imaging are small, data harmonization approaches can be used; for larger changes, direct image synthesis approaches have been explored. In this paper, we develop an approach based on multi-modal metric learning to synthesize images of diverse modalities. We use metric learning via multi-modal image retrieval, resulting in embeddings that can relate images of different modalities. Given a large image database, the learned image embeddings allow us to use k-nearest neighbor (k-NN) regression for image synthesis. Our driving medical problem is knee osteoarthritis (KOA), but our developed method is general after proper image alignment. We test our approach by synthesizing cartilage thickness maps obtained from 3D magnetic resonance (MR) images using 2D radiographs. Our experiments show that the proposed method outperforms direct image synthesis and that the synthesized thickness maps retain information relevant to downstream tasks such as progression prediction and Kellgren-Lawrence grading (KLG). Our results suggest that retrieval approaches can be used to obtain high-quality and meaningful image synthesis results given large image databases.

Pruning has emerged as a powerful technique for compressing deep neural networks, reducing memory usage and inference time without significantly affecting overall performance. However, the nuanced ways in which pruning impacts model behavior are not well understood, particularly for long-tailed, multi-label datasets commonly found in clinical settings. This knowledge gap could have dangerous implications when deploying a pruned model for diagnosis, where unexpected model behavior could impact patient well-being. To fill this gap, we perform the first analysis of pruning's effect on neural networks trained to diagnose thorax diseases from chest X-rays (CXRs). On two large CXR datasets, we examine which diseases are most affected by pruning and characterize class "forgettability" based on disease frequency and co-occurrence behavior. Further, we identify individual CXRs where uncompressed and heavily pruned models disagree, known as pruning-identified exemplars (PIEs), and conduct a human reader study to evaluate their unifying qualities. We find that radiologists perceive PIEs as having more label noise, lower image quality, and higher diagnosis difficulty. This work represents a first step toward understanding the impact of pruning on model behavior in deep long-tailed, multi-label medical image classification. All code, model weights, and data access instructions can be found at https://github.com/VITA-Group/PruneCXR.

One-shot federated learning (FL) has emerged as a promising solution in scenarios where multiple communication rounds are not practical. Notably, as feature distributions in medical data are less discriminative than those of natural images, robust global model training with FL is non-trivial and can lead to overfitting. To address this issue, we propose a novel one-shot FL framework leveraging Image Synthesis and Client model Adaptation (FedISCA) with knowledge distillation (KD). To prevent overfitting, we generate diverse synthetic images ranging from random noise to realistic images. This approach (i) alleviates data privacy concerns and (ii) facilitates robust global model training using KD with decentralized client models. To mitigate domain disparity in the early stages of synthesis, we design noise-adapted client models where batch normalization statistics on random noise (synthetic images) are updated to enhance KD. Lastly, the global model is trained with both the original and noise-adapted client models via KD and synthetic images. This process is repeated till global model convergence. Extensive evaluation of this design on five small- and three large-scale medical image classification datasets reveals superior accuracy over prior methods. Code is available at https://github.com/myeongkyunkang/FedISCA.

Vision Transformer (ViT) models have demonstrated a breakthrough in a wide range of computer vision tasks. However, compared to the Convolutional Neural Network (CNN) models, it has been observed that the ViT models struggle to capture high-frequency components of images, which can limit their ability to detect local textures and edge information. As abnormalities in human tissue, such as tumors and lesions, may greatly vary in structure, texture, and shape, high-frequency information such as texture is crucial for effective semantic segmentation tasks. To address this limitation in ViT models, we propose a new technique, Laplacian-Former, that enhances the self-attention map by adaptively re-calibrating the frequency information in a Laplacian pyramid. More specifically, our proposed method utilizes a dual attention mechanism via efficient attention and frequency attention while the efficient attention mechanism reduces the complexity of self-attention to linear while producing the same output, selectively intensifying the contribution of shape and texture features. Furthermore, we introduce a novel efficient enhancement multi-scale bridge that effectively transfers spatial information from the encoder to the decoder while preserving the fundamental features. We demonstrate the efficacy of Laplacian-former on multi-organ and skin lesion segmentation tasks with +1.87% and +0.76% dice scores compared to SOTA approaches, respectively. Our implementation is publically available at GitHub.