Convolutional neural networks (CNNs) have shown remarkable progress in medical image segmentation. However, the lesion segmentation remains a challenge to state-of-the-art CNN-based algorithms due to the variance in scales and shapes. On the one hand, tiny lesions are hard to delineate precisely from the medical images which are often of low resolutions. On the other hand, segmenting large-size lesions requires large receptive fields, which exacerbates the first challenge. In this article, we present a scale-aware super-resolution (SR) network to adaptively segment lesions of various sizes from low-resolution (LR) medical images. Our proposed network contains dual branches to simultaneously conduct lesion mask SR (LMSR) and lesion image SR (LISR). Meanwhile, we introduce scale-aware dilated convolution (SDC) blocks into the multitask decoders to adaptively adjust the receptive fields of the convolutional kernels according to the lesion sizes. To guide the segmentation branch to learn from richer high-resolution (HR) features, we propose a feature affinity (FA) module and a scale affinity (SA) module to enhance the multitask learning of the dual branches. On multiple challenging lesion segmentation datasets, our proposed network achieved consistent improvements compared with other state-of-the-art methods. Code will be available at: https://github.com/poiuohke/SASR_Net.

Ultrasonography plays an essential role in breast cancer diagnosis. Current deep learning based studies train the models on either images or videos in a centralized learning manner, lacking consideration of joint benefits between two different modality models or the privacy issue of data centralization. In this study, we propose the first decentralized learning solution for joint learning with breast ultrasound video and image, called FedBCD. To enable the model to learn from images and videos simultaneously and seamlessly in client-level local training, we propose a Joint Ultrasound Video and Image Learning (JUVIL) model to bridge the dimension gap between video and image data by incorporating temporal and spatial adapters. The parameter-efficient design of JUVIL with trainable adapters and frozen backbone further reduces the computational cost and communication burden of federated learning, finally improving the overall efficiency. Moreover, considering conventional model-wise aggregation may lead to unstable federated training due to different modalities, data capacities in different clients, and different functionalities across layers. We further propose a Fisher information matrix (FIM) guided Layer-wise Aggregation method named FILA. By measuring layer-wise sensitivity with FIM, FILA assigns higher contributions to the clients with lower sensitivity, improving personalized performance during federated training. Extensive experiments on three image clients and one video client demonstrate the benefits of joint learning architecture, especially for the ones with small-scale data. FedBCD significantly outperforms nine federated learning methods on both video-based and image-based diagnoses, demonstrating the superiority and potential for clinical practice. Code is released at https://github.com/tianpeng-deng/FedBCD.

Data from multimodalities bring complementary information for deep learning-based medical image classification models. However, data fusion methods simply concatenating features or images barely consider the correlations or complementarities among different modalities and easily suffer from exponential growth in dimensions and computational complexity when the modality increases. Consequently, this article proposes a subspace fusion network with tensor decomposition (TD) to heighten multimodal medical image classification. We first introduce a Tucker low-rank TD module to map the high-level dimensional tensor to the low-rank subspace, reducing the redundancy caused by multimodal data and high-dimensional features. Then, a cross-tensor attention mechanism is utilized to fuse features from the subspace into a high-dimension tensor, enhancing the representation ability of extracted features and constructing the interaction information among components in the subspace. Extensive comparison experiments with state-of-the-art (SOTA) methods are conducted on one self-established and three public multimodal medical image datasets, verifying the effectiveness and generalization ability of the proposed method. The code is available at https://github.com/1zhang-yi/TDSFNet.

Routine knee MRI protocols for 1.5 T and 3 T scanners, do not include T2*-w gradient echo (T2*W) images, which are useful in several clinical scenarios such as the assessment of cartilage, synovial blooming (deposition of hemosiderin), chondrocalcinosis and the evaluation of the physis in pediatric patients. Herein, we aimed to develop an open-source deep learning model that creates synthetic T2*W images of the knee using fat-suppressed intermediate-weighted images. A cycleGAN model was trained with 12,118 sagittal knee MR images and tested on an independent set of 2996 images. Diagnostic interchangeability of synthetic T2*W images was assessed against a series of findings. Voxel intensity of four tissues was evaluated with Bland-Altman plots. Image quality was assessed with the use of root mean squared error (NRMSE), structural similarity index (SSIM) and peak signal-to-noise ratio (PSNR). Code, model and a standalone executable file are provided on github. The model achieved a median NRMSE, PSNR and SSIM of 0.5, 17.4, and 0.5, respectively. Images were found interchangeable with an intraclass correlation coefficient >0.95 for all findings. Mean voxel intensity was equal between synthetic and conventional images. Four types of artifacts were identified: geometrical distortion (86/163 cases), object insertion/omission (11/163 cases), a wrap-around-like (26/163 cases) and an incomplete fat-suppression artifact (120/163 cases), which had a median 0 impact (no impact) on the diagnosis. In conclusion, the developed open-source GAN model creates synthetic T2*W images of the knee of high diagnostic value and quality. The identified artifacts had no or minor effect on the diagnostic value of the images.

Deep complex-valued neural networks (CVNNs) provide a powerful way to leverage complex number operations and representations and have succeeded in several phase-based applications. However, previous networks have not fully explored the impact of complex-valued networks in the frequency domain. Here, we introduce a unified complex-valued deep learning framework - Artificial Fourier Transform Network (AFTNet) - which combines domain-manifold learning and CVNNs. AFTNet can be readily used to solve image inverse problems in domain transformation, especially for accelerated magnetic resonance imaging (MRI) reconstruction and other applications. While conventional methods typically utilize magnitude images or treat the real and imaginary components of k-space data as separate channels, our approach directly processes raw k-space data in the frequency domain, utilizing complex-valued operations. This allows for a mapping between the frequency (k-space) and image domain to be determined through cross-domain learning. We show that AFTNet achieves superior accelerated MRI reconstruction compared to existing approaches. Furthermore, our approach can be applied to various tasks, such as denoised magnetic resonance spectroscopy (MRS) reconstruction and datasets with various contrasts. The AFTNet presented here is a valuable preprocessing component for different preclinical studies and provides an innovative alternative for solving inverse problems in imaging and spectroscopy. The code is available at: https://github.com/yanting-yang/AFT-Net.

Kidney irregularities pose a significant public health challenge, often leading to severe complications, yet the limited availability of nephrologists makes early detection costly and time-consuming. To address this issue, we propose a deep learning framework for automated kidney disease detection, leveraging feature fusion and sequential modeling techniques to enhance diagnostic accuracy. Our study thoroughly evaluates six pretrained models under identical experimental conditions, identifying ResNet50 and VGG19 as the highly efficient models for feature extraction due to their deep residual learning and hierarchical representations. Our proposed methodology integrates feature fusion with an inception block to extract diverse feature representations while maintaining imbalance dataset overhead. To enhance sequential learning and capture long-term dependencies in disease progression, ConvLSTM is incorporated after feature fusion. Additionally, Inception block is employed after ConvLSTM to refine hierarchical feature extraction, further strengthening the proposed model ability to leverage both spatial and temporal patterns. To validate our approach, we introduce a new named Multiple Hospital Collected (MHC-CT) dataset, consisting of 1860 tumor and 1024 normal kidney CT scans, meticulously annotated by medical experts. Our model achieves 99.60 % accuracy on this dataset, demonstrating its robustness in binary classification. Furthermore, to assess its generalization capability, we evaluate the model on a publicly available benchmark multiclass CT scan dataset, achieving 91.31 % accuracy. The superior performance is attributed to the effective feature fusion using inception blocks and the sequential learning capabilities of ConvLSTM, which together enhance spatial and temporal feature representations. These results highlight the efficacy of the proposed framework in automating kidney disease detection, providing a reliable, and efficient solution for clinical decision-making. https://github.com/VS-EYE/KidneyDiseaseDetection.git.

This work aims to perform a cross-site validation of automated segmentation for breast cancers in MRI and to compare the performance to radiologists. A three-dimensional (3D) U-Net was trained to segment cancers in dynamic contrast-enhanced axial MRIs using a large dataset from Site 1 (n = 15,266; 449 malignant and 14,817 benign). Performance was validated on site-specific test data from this and two additional sites, and common publicly available testing data. Four radiologists from each of the three clinical sites provided two-dimensional (2D) segmentations as ground truth. Segmentation performance did not differ between the network and radiologists on the test data from Sites 1 and 2 or the common public data (median Dice score Site 1, network 0.86 vs. radiologist 0.85, n = 114; Site 2, 0.91 vs. 0.91, n = 50; common: 0.93 vs. 0.90). For Site 3, an affine input layer was fine-tuned using segmentation labels, resulting in comparable performance between the network and radiologist (0.88 vs. 0.89, n = 42). Radiologist performance differed on the common test data, and the network numerically outperformed 11 of the 12 radiologists (median Dice: 0.85-0.94, n = 20). In conclusion, a deep network with a novel supervised harmonization technique matches radiologists' performance in MRI tumor segmentation across clinical sites. We make code and weights publicly available to promote reproducible AI in radiology.

Whole-body PET imaging plays an essential role in cancer diagnosis and treatment but suffers from low image quality. Traditional deep learning-based denoising methods work well for a specific acquisition but are less effective in handling diverse PET protocols. In this study, we proposed and validated a 3D Denoising Diffusion Probabilistic Model (3D DDPM) as a robust and universal solution for whole-body PET image denoising. The proposed 3D DDPM gradually injected noise into the images during the forward diffusion phase, allowing the model to learn to reconstruct the clean data during the reverse diffusion process. A 3D convolutional network was trained using high-quality data from the Biograph Vision Quadra PET/CT scanner to generate the score function, enabling the model to capture accurate PET distribution information extracted from the total-body datasets. The trained 3D DDPM was evaluated on datasets from four scanners, four tracer types, and six dose levels representing a broad spectrum of clinical scenarios. The proposed 3D DDPM consistently outperformed 2D DDPM, 3D UNet, and 3D GAN, demonstrating its superior denoising performance across all tested conditions. Additionally, the model's uncertainty maps exhibited lower variance, reflecting its higher confidence in its outputs. The proposed 3D DDPM can effectively handle various clinical settings, including variations in dose levels, scanners, and tracers, establishing it as a promising foundational model for PET image denoising. The trained 3D DDPM model of this work can be utilized off the shelf by researchers as a whole-body PET image denoising solution. The code and model are available at https://github.com/Miche11eU/PET-Image-Denoising-Using-3D-Diffusion-Model .

In computer-aided diagnosis systems, precise feature extraction from CT scans of colorectal cancer using deep learning is essential for effective prognosis. However, existing convolutional neural networks struggle to capture long-range dependencies and contextual information, resulting in incomplete CT feature extraction. To address this, the PEDRA-EFB0 architecture integrates patch embeddings and a dual residual attention mechanism for enhanced feature extraction and survival prediction in colorectal cancer CT scans. A patch embedding method processes CT scans into patches, creating positional features for global representation and guiding spatial attention computation. Additionally, a dual residual attention mechanism during the upsampling stage selectively combines local and global features, enhancing CT data utilization. Furthermore, this paper proposes a feature selection algorithm that combines autoencoders and entropy technology, encoding and compressing high-dimensional data to reduce redundant information and using entropy to assess the importance of features, thereby achieving precise feature selection. Experimental results indicate the PEDRA-EFB0 model outperforms traditional methods on colorectal cancer CT metrics, notably in C-index, BS, MCC, and AUC, enhancing survival prediction accuracy. Our code is freely available at https://github.com/smile0208z/PEDRA .

Magnetic Resonance Imaging (MRI) has become a pivotal tool in diagnosing brain diseases, with a wide array of computer-aided artificial intelligence methods being proposed to enhance diagnostic accuracy. However, early studies were often limited by small-scale datasets and a narrow range of disease types, which posed challenges in model generalization. This study presents UniBrain, a hierarchical knowledge-enhanced pre-training framework designed for universal brain MRI diagnosis. UniBrain leverages a large-scale dataset comprising 24,770 imaging-report pairs from routine diagnostics for pre-training. Unlike previous approaches that either focused solely on visual representation learning or used brute-force alignment between vision and language, the framework introduces a hierarchical alignment mechanism. This mechanism extracts structured knowledge from free-text clinical reports at multiple granularities, enabling vision-language alignment at both the sequence and case levels, thereby significantly improving feature learning efficiency. A coupled vision-language perception module is further employed for text-guided multi-label classification, which facilitates zero-shot evaluation and fine-tuning of downstream tasks without modifying the model architecture. UniBrain is validated on both in-domain and out-of-domain datasets, consistently surpassing existing state-of-the-art diagnostic models and demonstrating performance on par with radiologists in specific disease categories. It shows strong generalization capabilities across diverse tasks, highlighting its potential for broad clinical application. The code is available at https://github.com/ljy19970415/UniBrain.