Segmenting prostate organ areas and lesion areas based on multi-modal magnetic resonance images is an important task in the medical field, and has important reference value for doctors’ subsequent diagnosis and treatment plan formulation. However, existing segmentation methods still have some shortcomings. The most important thing is that many existing methods are mostly limited to the segmentation stage of a single task, which not only wastes computing resources, but also fails to make full use of shared features between different tasks and effectively utilize the spatial and texture structures of organs and lesions. Correlation can enhance the model’s understanding of anatomical structures and improve the accuracy and consistency of segmentation results. In this paper, we propose a multi-task learning model for multi-scale feature interaction and fusion (MFAS-Net), drawing on the insights of Progressive layered extraction model(PLE). The fusion module (MMFF) realizes cross-task feature interaction and fusion, thereby making better use of shared features between organs and lesions. The multi-scale attention supervision (MAS) module is used to supervise the lesions using feature maps of organ segmentation, and finally achieves joint segmentation of prostate organs and lesions. Through comparative experiments and ablation experiments on the PI-CAI data set, our model showed significant superiority in the joint segmentation task of prostate organs and lesions, verifying its effectiveness and practicability in medical image segmentation.

Segmentation of brain tumors from multi-modal magnetic resonance (MR) images is a crucial task in medical imaging, with significant clinical implications for the diagnosis and treatment planning of relevant brain disorders. In this paper, we present a 3D framework, that utilizes multi-modal MR images, including T1-weighted, T1ce-weighted, T2-weighted, and FLAIR images, to capture comprehensive information about the tumor for precise and automated brain tumor segmentation. The proposed framework leverages a multi-modal MRI fusion task to facilitate the optimization and training of brain tumor segmentation. Comparative experiments have indicated the effectiveness and superiority of the proposed framework for brain tumor segmentation tasks.

High-performance deep learning models require large amounts of data with high quality annotations for model training, while the labeling work usually takes a lot of time for the experts. Meanwhile, the inter-observer variability al-ways exist between annotations from different experts and the distribution shift between the data acquired from different medical institutions. To address these challenges, we propose an end-to-end domain adaptive collaborative learning network for multi-institutional prostate MRI segmentation. Specifically, we introduce an unpaired image translation module to match the image domains between different institutions, which can alleviate the heterogeneity between 1.5T and 3T prostate MR images during model training. Moreover, we design a self-taught strategy to transfer domain-aware knowledge to jointly learn generic and unique representations. Furthermore, we evaluate our approach in scenarios with limited or without annotations, experimental results show that our approach has better adaptation performance than traditional supervised learning approaches, and has the potential to extend to unsupervised domain adaptation scenario. We also evaluate our approach with prostate MRI segmentation benchmark datasets, experimental results show that our approach outperforms several state-of-the-art methods.

Contouring prostate tumor in magnetic resonance images is a prerequisite for diagnosis. Automatically segmenting blurred lesion regions is challenging and requires fully leveraging multi-parameter MR images. This paper proposes MFSL-Net, an end-to-end network that cascades two novel sub-networks: 1) a modality fusion network that selectively fuses information of two MRI modalities by expanding a dual-stream CNN with spatial and channel attention modules; 2) a shape learning network that integrates shape learning and context learning to recognize the shape and edge information while preserving high-resolution semantic information. We justify MFSL-Net’s design by ablation experiments and compare its performance with the state-of-the-art approaches. Experimental results show a 3.6% improvement in Dice Similarity Coefficient, which confirms the effectiveness of MFSL-Net.