MA-SAM: Modality-agnostic SAM adaptation for 3D medical image segmentation.

The Segment Anything Model (SAM), a foundation model for general image segmentation, has demonstrated impressive zero-shot performance across numerous natural image segmentation tasks. However, SAM's performance significantly declines when applied to medical images, primarily due to the substantial disparity between natural and medical image domains. To effectively adapt SAM to medical images, it is important to incorporate critical third-dimensional information, i.e., volumetric or temporal knowledge, during fine-tuning. Simultaneously, we aim to harness SAM's pre-trained weights within its original 2D backbone to the fullest extent. In this paper, we introduce a modality-agnostic SAM adaptation framework, named as MA-SAM, that is applicable to various volumetric and video medical data. Our method roots in the parameter-efficient fine-tuning strategy to update only a small portion of weight increments while preserving the majority of SAM's pre-trained weights. By injecting a series of 3D adapters into the transformer blocks of the image encoder, our method enables the pre-trained 2D backbone to extract third-dimensional information from input data. We comprehensively evaluate our method on five medical image segmentation tasks, by using 11 public datasets across CT, MRI, and surgical video data. Remarkably, without using any prompt, our method consistently outperforms various state-of-the-art 3D approaches, surpassing nnU-Net by 0.9%, 2.6%, and 9.9% in Dice for CT multi-organ segmentation, MRI prostate segmentation, and surgical scene segmentation respectively. Our model also demonstrates strong generalization, and excels in challenging tumor segmentation when prompts are used. Our code is available at: https://github.com/cchen-cc/MA-SAM.

CauDR: A causality-inspired domain generalization framework for fundus-based diabetic retinopathy grading.

Diabetic retinopathy (DR) is the most common diabetic complication, which usually leads to retinal damage, vision loss, and even blindness. A computer-aided DR grading system has a significant impact on helping ophthalmologists with rapid screening and diagnosis. Recent advances in fundus photography have precipitated the development of novel retinal imaging cameras and their subsequent implementation in clinical practice. However, most deep learning-based algorithms for DR grading demonstrate limited generalization across domains. This inferior performance stems from variance in imaging protocols and devices inducing domain shifts. We posit that declining model performance between domains arises from learning spurious correlations in the data. Incorporating do-operations from causality analysis into model architectures may mitigate this issue and improve generalizability. Specifically, a novel universal structural causal model (SCM) was proposed to analyze spurious correlations in fundus imaging. Building on this, a causality-inspired diabetic retinopathy grading framework named CauDR was developed to eliminate spurious correlations and achieve more generalizable DR diagnostics. Furthermore, existing datasets were reorganized into 4DR benchmark for DG scenario. Results demonstrate the effectiveness and the state-of-the-art (SOTA) performance of CauDR. Diabetic retinopathy (DR) is the most common diabetic complication, which usually leads to retinal damage, vision loss, and even blindness. A computer-aided DR grading system has a significant impact on helping ophthalmologists with rapid screening and diagnosis. Recent advances in fundus photography have precipitated the development of novel retinal imaging cameras and their subsequent implementation in clinical practice. However, most deep learning-based algorithms for DR grading demonstrate limited generalization across domains. This inferior performance stems from variance in imaging protocols and devices inducing domain shifts. We posit that declining model performance between domains arises from learning spurious correlations in the data. Incorporating do-operations from causality analysis into model architectures may mitigate this issue and improve generalizability. Specifically, a novel universal structural causal model (SCM) was proposed to analyze spurious correlations in fundus imaging. Building on this, a causality-inspired diabetic retinopathy grading framework named CauDR was developed to eliminate spurious correlations and achieve more generalizable DR diagnostics. Furthermore, existing datasets were reorganized into 4DR benchmark for DG scenario. Results demonstrate the effectiveness and the state-of-the-art (SOTA) performance of CauDR.

SC-SSL: Self-Correcting Collaborative and Contrastive Co-Training Model for Semi-Supervised Medical Image Segmentation.

Image segmentation achieves significant improvements with deep neural networks at the premise of a large scale of labeled training data, which is laborious to assure in medical image tasks. Recently, semi-supervised learning (SSL) has shown great potential in medical image segmentation. However, the influence of the learning target quality for unlabeled data is usually neglected in these SSL methods. Therefore, this study proposes a novel self-correcting co-training scheme to learn a better target that is more similar to ground-truth labels from collaborative network outputs. Our work has three-fold highlights. First, we advance the learning target generation as a learning task, improving the learning confidence for unannotated data with a self-correcting module. Second, we impose a structure constraint to encourage the shape similarity further between the improved learning target and the collaborative network outputs. Finally, we propose an innovative pixel-wise contrastive learning loss to boost the representation capacity under the guidance of an improved learning target, thus exploring unlabeled data more efficiently with the awareness of semantic context. We have extensively evaluated our method with the state-of-the-art semi-supervised approaches on four public-available datasets, including the ACDC dataset, M&Ms dataset, Pancreas-CT dataset, and Task_07 CT dataset. The experimental results with different labeled-data ratios show our proposed method's superiority over other existing methods, demonstrating its effectiveness in semi-supervised medical image segmentation.

Adaptive Frequency Learning Network With Anti-Aliasing Complex Convolutions for Colon Diseases Subtypes.

The automatic and dependable identification of colonic disease subtypes by colonoscopy is crucial. Once successful, it will facilitate clinically more in-depth disease staging analysis and the formulation of more tailored treatment plans. However, inter-class confusion and brightness imbalance are major obstacles to colon disease subtyping. Notably, the Fourier-based image spectrum, with its distinctive frequency features and brightness insensitivity, offers a potential solution. To effectively leverage its advantages to address the existing challenges, this article proposes a framework capable of thorough learning in the frequency domain based on four core designs: the position consistency module, the high-frequency self-supervised module, the complex number arithmetic model, and the feature anti-aliasing module. The position consistency module enables the generation of spectra that preserve local and positional information while compressing the spectral data range to improve training stability. Through band masking and supervision, the high-frequency autoencoder module guides the network to learn useful frequency features selectively. The proposed complex number arithmetic model allows direct spectral training while avoiding the loss of phase information caused by current general-purpose real-valued operations. The feature anti-aliasing module embeds filters in the model to prevent spectral aliasing caused by down-sampling and improve performance. Experiments are performed on the collected five-class dataset, which contains 4591 colorectal endoscopic images. The outcomes show that our proposed method produces state-of-the-art results with an accuracy rate of 89.82%.

Automatic 3-D spine curve measurement in freehand ultrasound via structure-aware reinforcement learning spinous process localization.

Freehand 3-D ultrasound systems have been advanced in scoliosis assessment to avoid radiation hazards, especially for teenagers. This novel 3-D imaging method also makes it possible to evaluate the spine curvature automatically from the corresponding 3-D projection images. However, most approaches neglect the three-dimensional spine deformity by only using the rendering images, thus limiting their usage in clinical applications. In this study, we proposed a structure-aware localization model to directly identify the spinous processes for automatic 3-D spine curve measurement using the images acquired with freehand 3-D ultrasound imaging. The pivot is to leverage a novel reinforcement learning (RL) framework to localize the landmarks, which adopts a multi-scale agent to boost structure representation with positional information. We also introduced a structure similarity prediction mechanism to perceive the targets with apparent spinous process structures. Finally, a two-fold filtering strategy was proposed to screen the detected spinous processes landmarks iteratively, followed by a three-dimensional spine curve fitting for the spine curvature assessments. We evaluated the proposed model on 3-D ultrasound images among subjects with different scoliotic angles. The results showed that the mean localization accuracy of the proposed landmark localization algorithm was 5.95 pixels. Also, the curvature angles on the coronal plane obtained by the new method had a high linear correlation with those by manual measurement (R = 0.86, p < 0.001). These results demonstrated the potential of our proposed method for facilitating the 3-D assessment of scoliosis, especially for 3-D spine deformity assessment.

AWSnet: An auto-weighted supervision attention network for myocardial scar and edema segmentation in multi-sequence cardiac magnetic resonance images.

Multi-sequence cardiac magnetic resonance (CMR) provides essential pathology information (scar and edema) to diagnose myocardial infarction. However, automatic pathology segmentation can be challenging due to the difficulty of effectively exploring the underlying information from the multi-sequence CMR data. This paper aims to tackle the scar and edema segmentation from multi-sequence CMR with a novel auto-weighted supervision framework, where the interactions among different supervised layers are explored under a task-specific objective using reinforcement learning. Furthermore, we design a coarse-to-fine framework to boost the small myocardial pathology region segmentation with shape prior knowledge. The coarse segmentation model identifies the left ventricle myocardial structure as a shape prior, while the fine segmentation model integrates a pixel-wise attention strategy with an auto-weighted supervision model to learn and extract salient pathological structures from the multi-sequence CMR data. Extensive experimental results on a publicly available dataset from Myocardial pathology segmentation combining multi-sequence CMR (MyoPS 2020) demonstrate our method can achieve promising performance compared with other state-of-the-art methods. Our method is promising in advancing the myocardial pathology assessment on multi-sequence CMR data. To motivate the community, we have made our code publicly available via https://github.com/soleilssss/AWSnet/tree/master.

AW3M: An auto-weighting and recovery framework for breast cancer diagnosis using multi-modal ultrasound.

Recently, more clinicians have realized the diagnostic value of multi-modal ultrasound in breast cancer identification and began to incorporate Doppler imaging and Elastography in the routine examination. However, accurately recognizing patterns of malignancy in different types of sonography requires expertise. Furthermore, an accurate and robust diagnosis requires proper weights of multi-modal information as well as the ability to process missing data in practice. These two aspects are often overlooked by existing computer-aided diagnosis (CAD) approaches. To overcome these challenges, we propose a novel framework (called AW3M) that utilizes four types of sonography (i.e. B-mode, Doppler, Shear-wave Elastography, and Strain Elastography) jointly to assist breast cancer diagnosis. It can extract both modality-specific and modality-invariant features using a multi-stream CNN model equipped with self-supervised consistency loss. Instead of assigning the weights of different streams empirically, AW3M automatically learns the optimal weights using reinforcement learning techniques. Furthermore, we design a light-weight recovery block that can be inserted to a trained model to handle different modality-missing scenarios. Experimental results on a large multi-modal dataset demonstrate that our method can achieve promising performance compared with state-of-the-art methods. The AW3M framework is also tested on another independent B-mode dataset to prove its efficacy in general settings. Results show that the proposed recovery block can learn from the joint distribution of multi-modal features to further boost the classification accuracy given single modality input during the test.

Learn Fine-Grained Adaptive Loss for Multiple Anatomical Landmark Detection in Medical Images.

Automatic and accurate detection of anatomical landmarks is an essential operation in medical image analysis with a multitude of applications. Recent deep learning methods have improved results by directly encoding the appearance of the captured anatomy with the likelihood maps (i.e., heatmaps). However, most current solutions overlook another essence of heatmap regression, the objective metric for regressing target heatmaps and rely on hand-crafted heuristics to set the target precision, thus being usually cumbersome and task-specific. In this paper, we propose a novel learning-to-learn framework for landmark detection to optimize the neural network and the target precision simultaneously. The pivot of this work is to leverage the reinforcement learning (RL) framework to search objective metrics for regressing multiple heatmaps dynamically during the training process, thus avoiding setting problem-specific target precision. We also introduce an early-stop strategy for active termination of the RL agent's interaction that adapts the optimal precision for separate targets considering exploration-exploitation tradeoffs. This approach shows better stability in training and improved localization accuracy in inference. Extensive experimental results on two different applications of landmark localization: 1) our in-house prenatal ultrasound (US) dataset and 2) the publicly available dataset of cephalometric X-Ray landmark detection, demonstrate the effectiveness of our proposed method. Our proposed framework is general and shows the potential to improve the efficiency of anatomical landmark detection.