Medical image analysis最新文献

Predicting the infiltration of Glioblastoma (GBM) from medical MRI scans is crucial for understanding tumor growth dynamics and designing personalized radiotherapy treatment plans. Mathematical models of GBM growth can complement the data in the prediction of spatial distributions of tumor cells. However, this requires estimating patient-specific parameters of the model from clinical data, which is a challenging inverse problem due to limited temporal data and the limited time between imaging and diagnosis. This work proposes a method that uses Physics-Informed Neural Networks (PINNs) to estimate patient-specific parameters of a reaction-diffusion partial differential equation (PDE) model of GBM growth from a single 3D structural MRI snapshot. PINNs embed both the data and the PDE into a loss function, thus integrating theory and data. Key innovations include the identification and estimation of characteristic non-dimensional parameters, a pre-training step that utilizes the non-dimensional parameters and a fine-tuning step to determine the patient specific parameters. Additionally, the diffuse-domain method is employed to handle the complex brain geometry within the PINN framework. The method is validated on both synthetic and patient datasets, showing promise for personalized GBM treatment through parametric inference within clinically relevant timeframes.

A deep comparative analysis of brain functional connectome across species in primates has the potential to yield valuable insights for both scientific and clinical applications. However, the interspecies commonality and differences are inherently entangled with each other and with other irrelevant factors. Here we develop a novel contrastive machine learning method, called shared-unique variation autoencoder (SU-VAE), to allow disentanglement of the species-shared and species-specific functional connectome variation between macaque and human brains on large-scale resting-state fMRI datasets. The method was validated by confirming that human-specific features are differentially related to cognitive scores, while features shared with macaque better capture sensorimotor ones. The projection of disentangled connectomes to the cortex revealed a gradient that reflected species divergence. In contrast to macaque, the introduction of human-specific connectomes to the shared ones enhanced network efficiency. We identified genes enriched on 'axon guidance' that could be related to the human-specific connectomes. The code contains the model and analysis can be found in https://github.com/BBBBrain/SU-VAE.

Continual test-time adaptation (CTTA) aims to continuously adapt a source-trained model to a target domain with minimal performance loss while assuming no access to the source data. Typically, source models are trained with empirical risk minimization (ERM) and assumed to perform reasonably on the target domain to allow for further adaptation. However, ERM-trained models often fail to perform adequately on a severely drifted target domain, resulting in unsatisfactory adaptation results. To tackle this issue, we propose a generalizable CTTA framework. First, we incorporate domain-invariant shape modeling into the model and train it using domain-generalization (DG) techniques, promoting target-domain adaptability regardless of the severity of the domain shift. Then, an uncertainty and shape-aware mean teacher network performs adaptation with uncertainty-weighted pseudo-labels and shape information. As part of this process, a novel uncertainty-ranked cross-task regularization scheme is proposed to impose consistency between segmentation maps and their corresponding shape representations, both produced by the student model, at the patch and global levels to enhance performance further. Lastly, small portions of the model's weights are stochastically reset to the initial domain-generalized state at each adaptation step, preventing the model from 'diving too deep' into any specific test samples. The proposed method demonstrates strong continual adaptability and outperforms its peers on five cross-domain segmentation tasks, showcasing its effectiveness and generalizability.

There is no doubt that advanced artificial intelligence models and high quality data are the keys to success in developing computational pathology tools. Although the overall volume of pathology data keeps increasing, a lack of quality data is a common issue when it comes to a specific task due to several reasons including privacy and ethical issues with patient data. In this work, we propose to exploit knowledge distillation, i.e., utilize the existing model to learn a new, target model, to overcome such issues in computational pathology. Specifically, we employ a student-teacher framework to learn a target model from a pre-trained, teacher model without direct access to source data and distill relevant knowledge via momentum contrastive learning with multi-head attention mechanism, which provides consistent and context-aware feature representations. This enables the target model to assimilate informative representations of the teacher model while seamlessly adapting to the unique nuances of the target data. The proposed method is rigorously evaluated across different scenarios where the teacher model was trained on the same, relevant, and irrelevant classification tasks with the target model. Experimental results demonstrate the accuracy and robustness of our approach in transferring knowledge to different domains and tasks, outperforming other related methods. Moreover, the results provide a guideline on the learning strategy for different types of tasks and scenarios in computational pathology.

Medical report generation, a cross-modal task of generating medical text information, aiming to provide professional descriptions of medical images in clinical language. Despite some methods have made progress, there are still some limitations, including insufficient focus on lesion areas, omission of internal edge features, and difficulty in aligning cross-modal data. To address these issues, we propose Dual-Modality Visual Feature Flow (DMVF) for medical report generation. Firstly, we introduce region-level features based on grid-level features to enhance the method's ability to identify lesions and key areas. Then, we enhance two types of feature flows based on their attributes to prevent the loss of key information, respectively. Finally, we align visual mappings from different visual feature with report textual embeddings through a feature fusion module to perform cross-modal learning. Extensive experiments conducted on four benchmark datasets demonstrate that our approach outperforms the state-of-the-art methods in both natural language generation and clinical efficacy metrics.

Semantic segmentation is an essential component of medical image analysis research, with recent deep learning algorithms offering out-of-the-box applicability across diverse datasets. Despite these advancements, segmentation failures remain a significant concern for real-world clinical applications, necessitating reliable detection mechanisms. This paper introduces a comprehensive benchmarking framework aimed at evaluating failure detection methodologies within medical image segmentation. Through our analysis, we identify the strengths and limitations of current failure detection metrics, advocating for the risk-coverage analysis as a holistic evaluation approach. Utilizing a collective dataset comprising five public 3D medical image collections, we assess the efficacy of various failure detection strategies under realistic test-time distribution shifts. Our findings highlight the importance of pixel confidence aggregation and we observe superior performance of the pairwise Dice score (Roy et al., 2019) between ensemble predictions, positioning it as a simple and robust baseline for failure detection in medical image segmentation. To promote ongoing research, we make the benchmarking framework available to the community.

Cardiac diffusion tensor imaging (cDTI) is highly prone to image corruption, yet robust-fitting methods are rarely used. Single voxel outlier detection (SVOD) can overlook corruptions that are visually obvious, perhaps causing reluctance to replace whole-image shot-rejection (SR) despite its own deficiencies. SVOD's deficiencies may be relatively unimportant: corrupted signals that are not statistical outliers may not be detrimental. Multiple voxel outlier detection (MVOD), using a local myocardial neighbourhood, may overcome the shared deficiencies of SR and SVOD for cDTI while keeping the benefits of both. Here, robust fitting methods using M-estimators are derived for both non-linear least squares and weighted least squares fitting, and outlier detection is applied using (i) SVOD; and (ii) SVOD and MVOD. These methods, along with non-robust fitting with/without SR, are applied to cDTI datasets from healthy volunteers and hypertrophic cardiomyopathy patients. Robust fitting methods produce larger group differences with more statistical significance for MD, FA, and E2A, versus non-robust methods, with MVOD giving the largest group differences for MD and FA. Visual analysis demonstrates the superiority of robust-fitting methods over SR, especially when it is difficult to partition the images into good and bad sets. Synthetic experiments confirm that MVOD gives lower root-mean-square-error than SVOD.

Resting-state functional magnetic resonance imaging (rs-fMRI) provides a non-invasive imaging technique to study patterns of brain activity, and is increasingly used to facilitate automated brain disorder analysis. Existing fMRI-based learning methods often rely on labeled data to construct learning models, while the data annotation process typically requires significant time and resource investment. Graph contrastive learning offers a promising solution to address the small labeled data issue, by augmenting fMRI time series for self-supervised learning. However, data augmentation strategies employed in these approaches may damage the original blood-oxygen-level-dependent (BOLD) signals, thus hindering subsequent fMRI feature extraction. In this paper, we propose a self-supervised graph contrastive learning framework with diffusion augmentation (GCDA) for functional MRI analysis. The GCDA consists of a pretext model and a task-specific model. In the pretext model, we first augment each brain functional connectivity network derived from fMRI through a graph diffusion augmentation (GDA) module, and then use two graph isomorphism networks with shared parameters to extract features in a self-supervised contrastive learning manner. The pretext model can be optimized without the need for labeled training data, while the GDA focuses on perturbing graph edges and nodes, thus preserving the integrity of original BOLD signals. The task-specific model involves fine-tuning the trained pretext model to adapt to downstream tasks. Experimental results on two rs-fMRI cohorts with a total of 1230 subjects demonstrate the effectiveness of our method compared with several state-of-the-arts.