Automated classification of liver lesions in multi-phase CT and MR scans is of clinical significance but challenging. This study proposes a novel Siamese Dual-Resolution Transformer (SDR-Former) framework, specifically designed for liver lesion classification in 3D multi-phase CT and MR imaging with varying phase counts. The proposed SDR-Former utilizes a streamlined Siamese Neural Network (SNN) to process multi-phase imaging inputs, possessing robust feature representations while maintaining computational efficiency. The weight-sharing feature of the SNN is further enriched by a hybrid Dual-Resolution Transformer (DR-Former), comprising a 3D Convolutional Neural Network (CNN) and a tailored 3D Transformer for processing high- and low-resolution images, respectively. This hybrid sub-architecture excels in capturing detailed local features and understanding global contextual information, thereby, boosting the SNN's feature extraction capabilities. Additionally, a novel Adaptive Phase Selection Module (APSM) is introduced, promoting phase-specific intercommunication and dynamically adjusting each phase's influence on the diagnostic outcome. The proposed SDR-Former framework has been validated through comprehensive experiments on two clinical datasets: a three-phase CT dataset and an eight-phase MR dataset. The experimental results affirm the efficacy of the proposed framework. To support the scientific community, we are releasing our extensive multi-phase MR dataset for liver lesion analysis to the public. This pioneering dataset, being the first publicly available multi-phase MR dataset in this field, also underpins the MICCAI LLD-MMRI Challenge. The dataset is accessible at:https://bit.ly/3IyYlgN.
Accurate medical image segmentation demands the integration of multi-scale information, spanning from local features to global dependencies. However, it is challenging for existing methods to model long-range global information, where convolutional neural networks (CNNs) are constrained by their local receptive fields, and vision transformers (ViTs) suffer from high quadratic complexity of their attention mechanism. Recently, Mamba-based models have gained great attention for their impressive ability in long sequence modeling. Several studies have demonstrated that these models can outperform popular vision models in various tasks, offering higher accuracy, lower memory consumption, and less computational burden. However, existing Mamba-based models are mostly trained from scratch and do not explore the power of pretraining, which has been proven to be quite effective for data-efficient medical image analysis. This paper introduces a novel Mamba-based model, Swin-UMamba, designed specifically for medical image segmentation tasks, leveraging the advantages of ImageNet-based pretraining. Our experimental results reveal the vital role of ImageNet-based training in enhancing the performance of Mamba-based models. Swin-UMamba demonstrates superior performance with a large margin compared to CNNs, ViTs, and latest Mamba-based models. Notably, on AbdomenMRI, Encoscopy, and Microscopy datasets, Swin-UMamba outperforms its closest counterpart U-Mamba by an average score of 3.58%. The code and models of Swin-UMamba are publicly available at: https://github.com/JiarunLiu/Swin-UMamba
Parameter-efficient fine-tuning (PEFT) that was initially developed for exploiting pre-trained large language models has recently emerged as an effective approach to perform transfer learning on computer vision tasks. However, the effectiveness of PEFT on medical vision foundation models is still unclear and remains to be explored. As a proof of concept, we conducted a detailed empirical study on applying PEFT to chest radiography foundation models. Specifically, we delved into LoRA, a representative PEFT method, and compared it against full-parameter fine-tuning (FFT) on two self-supervised radiography foundation models across three well-established chest radiograph datasets. Our results showed that LoRA outperformed FFT in 13 out of 18 transfer learning tasks by at most 2.9% using fewer than 1% tunable parameters. Combining LoRA with foundation models, we set up new state-of-the-art on a range of data-efficient learning tasks, such as an AUROC score of 80.6% using 1% labeled data on NIH ChestX-ray14. We hope this study can evoke more attention from the community in the use of PEFT for transfer learning on medical imaging tasks. Code and models are available at https://github.com/RL4M/MED-PEFT.
Locating pathologies automatically from medical images aids the understanding of the emergence and progression of diseases, and such an ability can significantly benefit clinical diagnostics. However, existing deep learning models heavily rely on expert annotations and lack generalization capabilities in open clinical environments. In this study, we present a generalizable vision-language pre-training model for Annotation-Free pathology Localization (AFLoc). The core strength of AFLoc lies in its image annotation-free multi-level semantic structure-based contrastive learning, which comprehensively aligns multi-granularity medical concepts from reports with abundant image features, to adapt to the diverse expressions of observed and emerging unseen pathologies. We conducted extensive experimental validation across 4 distinct external datasets, encompassing 11 types of chest pathologies, to verify its generalization ability. The results demonstrate that AFLoc surpasses 6 state-of-the-art methods and even outperforms the human benchmark in locating 5 different pathologies, underscoring its suitability for complex clinical environments.
Existing contrastive language-image pre-training aims to learn a joint representation by matching abundant image-text pairs. However, the number of image-text pairs in medical datasets is usually orders of magnitude smaller than that in natural datasets. Besides, medical image-text pairs often involve numerous complex fine-grained correspondences. This paper aims to enhance the data efficiency by introducing multiple-to-multiple local relationship modeling to capture denser supervisions. More specifically, we propose a Medical Language-Image Pre-training (MLIP) framework, which exploits the limited image-text medical data more efficiently through patch-sentence matching. Furthermore, we introduce a masked contrastive learning strategy with semantic integrity estimation to reduce redundancy in images while preserving the underlying semantics. Our evaluation results show that MLIP outperforms previous work in zero/few-shot classification and few-shot segmentation tasks by a large margin.
The development of multi-modal medical foundation models has attracted significant attention in the field of medicine and healthcare due to their promising prospects in various clinical applications. One area of focus in this research direction is the extractions of features at different scales. While previous studies have explored feature learning at individual scales, investigation on integrating the diverse scales and modalities of information is lacking, which may hinder the potential for mutual reinforcement among these features. This paper aims to bridge this gap by proposing a method that effectively exploits multi-scale and cross-modality information to enhance the performance of medical foundation models. The proposed method simultaneously exploit features at the local, instance, modality and global aspects, facilitating comprehensive representation learning within the models. We evaluate the effectiveness of the proposed method on six open-source datasets across different clinical tasks, demonstrating its ability to enhance the performance of medical foundation models.
Multimodal deep learning utilizing imaging and diagnostic reports has made impressive progress in the field of medical imaging diagnostics, demonstrating a particularly strong capability for auxiliary diagnosis in cases where sufficient annotation information is lacking. Nonetheless, localizing diseases accurately without detailed positional annotations remains a challenge. Although existing methods have attempted to utilize local information to achieve fine-grained semantic alignment, their capability in extracting the fine-grained semantics of the comprehensive contextual within reports is limited. To solve this problem, we introduce a new method that takes full sentences from textual reports as the basic units for local semantic alignment. Our approach combines chest X-ray images with their corresponding textual reports, performing contrastive learning at both global and local levels. The leading results obtained by our method on multiple datasets confirm its efficacy in the task of lesion localization.
Recent studies have integrated convolution into transformers to introduce inductive bias and improve generalization performance. However, the static nature of conventional convolution prevents it from dynamically adapting to input variations, resulting in a representation discrepancy between convolution and self-attention as self-attention calculates attention matrices dynamically. Furthermore, when stacking token mixers that consist of convolution and self-attention to form a deep network, the static nature of convolution hinders the fusion of features previously generated by self-attention into convolution kernels. These two limitations result in a sub-optimal representation capacity of the constructed networks. To find a solution, we propose a lightweight Dual Dynamic Token Mixer (D-Mixer) that aggregates global information and local details in an input-dependent way. D-Mixer works by applying an efficient global attention module and an input-dependent depthwise convolution separately on evenly split feature segments, endowing the network with strong inductive bias and an enlarged effective receptive field. We use D-Mixer as the basic building block to design TransXNet, a novel hybrid CNN-Transformer vision backbone network that delivers compelling performance. In the ImageNet-1K image classification task, TransXNet-T surpasses Swin-T by 0.3\% in top-1 accuracy while requiring less than half of the computational cost. Furthermore, TransXNet-S and TransXNet-B exhibit excellent model scalability, achieving top-1 accuracy of 83.8\% and 84.6\% respectively, with reasonable computational costs. Additionally, our proposed network architecture demonstrates strong generalization capabilities in various dense prediction tasks, outperforming other state-of-the-art networks while having lower computational costs.
A long-standing goal of AI systems is to perform complex multimodal reasoning like humans. Recently, large language models (LLMs) have made remarkable strides in such multi-step reasoning on the language modality solely by leveraging the chain of thought (CoT) to mimic human thinking. However, the transfer of these advancements to multimodal contexts introduces heightened challenges, including but not limited to the impractical need for labor-intensive annotation and the limitations in terms of flexibility, generalizability, and explainability. To evoke CoT reasoning in multimodality, this work first conducts an in-depth analysis of these challenges posed by multimodality and presents two key insights: "keeping critical thinking" and "letting everyone do their jobs" in multimodal CoT reasoning. Furthermore, this study proposes a novel DDCoT prompting that maintains a critical attitude through negative-space prompting and incorporates multimodality into reasoning by first dividing the reasoning responsibility of LLMs into reasoning and recognition and then integrating the visual recognition capability of visual models into the joint reasoning process. The rationales generated by DDCoT not only improve the reasoning abilities of both large and small language models in zero-shot prompting and fine-tuning learning, significantly outperforming state-of-the-art methods but also exhibit impressive generalizability and explainability.
Albeit the notable performance on in-domain test points, it is non-trivial for deep neural networks to attain satisfactory accuracy when deploying in the open world, where novel domains and object classes often occur. In this paper, we study a practical problem of Domain Generalization under Category Shift (DGCS), which aims to simultaneously detect unknown-class samples and classify known-class samples in the target domains. Compared to prior DG works, we face two new challenges: 1) how to learn the concept of ``unknown'' during training with only source known-class samples, and 2) how to adapt the source-trained model to unseen environments for safe model deployment. To this end, we propose a novel Activate and Reject (ART) framework to reshape the model's decision boundary to accommodate unknown classes and conduct post hoc modification to further discriminate known and unknown classes using unlabeled test data. Specifically, during training, we promote the response to the unknown by optimizing the unknown probability and then smoothing the overall output to mitigate the overconfidence issue. At test time, we introduce a step-wise online adaptation method that predicts the label by virtue of the cross-domain nearest neighbor and class prototype information without updating the network's parameters or using threshold-based mechanisms. Experiments reveal that ART consistently improves the generalization capability of deep networks on different vision tasks. For image classification, ART improves the H-score by 6.1% on average compared to the previous best method. For object detection and semantic segmentation, we establish new benchmarks and achieve competitive performance.