Medical image analysis plays key role in computer-aided diagnosis, where segmentation and classification are essential and interconnected tasks. While multi-task learning (MTL) has been widely explored to leverage inter-task synergies, effectively guiding knowledge transfer to prevent task conflict and negative transfer remains a key challenge, particularly in anatomically complex diagnostic scenarios. This paper presents LTRMTL-Net, a novel multi-task learning framework for medical image analysis that simultaneously addresses segmentation and classification tasks guided by lesion regions and spatial relationships of tissues. The proposed architecture integrates an Enhanced Lesion Region Fusion (ELRF) module that leverages GradCAM-guided attention mechanisms to precisely locate and enhance lesion regions, providing critical prior knowledge for both tasks. Tissue Space Structure Prediction (TSSP) component captures local-global spatial dependencies through contrastive learning, establishing effective anatomical context modeling. The core encoder employs Hybrid Wavelet-State Attention blocks that combine modulated wavelet transform convolutions with structured state space models to extract multi-scale features while maintaining computational efficiency. Dual-stream inputs with symmetric architecture accommodate single-source scenarios across diverse medical imaging applications. Experimental results on mammography and breast ultrasound datasets demonstrate that the proposed method captures fine-grained lesion boundary details while providing accurate malignancy classification. Harnessing cooperative knowledge transfer between segmentation and classification, guided by anatomical priors, boosts diagnostic performance and provides comprehensive, interpretable clinical insights.

Due to the lack of appropriate priors, generating the content of dark regions remains a challenge in low-light image enhancement tasks. Currently, diffusion models employ robust image generation capabilities for enhancing low-light images. However, diffusion models require multiple iterations at the image feature level to generate details and content, which limits the speed. Moreover, the diffusion-based methods tend to generate unexpected artifacts in the degraded regions. To address these issues, we propose a Frequency Priors-guided Image Enhancement (FPIE) network, including a frequency prior generation network and an image restoration network. FPIE significantly accelerates inference by learning abstract prior with frequency domain constraints. Concretely, to learn compacted priors at the frequency domain, we introduce a joint training approach for the prior generation and restoration models to constrain the distribution of priors. Furthermore, to better utilize frequency-domain features for enhancing the network’s generation capabilities, a wavelet-based transformer block is introduced to produce intricate details and avoid the artifacts of the output. Extensive experimental results on the commonly used benchmarks demonstrate that our approach achieves state-of-the-art performances and well generalization to real-world images.

Lesion segmentation algorithms automatically outline lesion areas in medical images, facilitating more effective identification and assessment of the clinically relevant features, and improving the efficacy and diagnosis accuracy. However, most fully convolutional network based segmentation methods suffer from spatial and contextual information loss when decreasing image resolution. To overcome this shortcoming, this paper proposes a skin lesion segmentation model , namely, the Multi-Scale Contextual Attention Network (MSCA-Net), which can exploit the multi-scale contextual information in images. Inspired by the skip connection of U-Net, we design a multi-scale bridge (MSB) module which interacts with multi-scale features to effectively fuse the multi-scale contextual information of the encoder and decoder path features. We further propose a global-local channel spatial attention module (GL-CSAM), aiming at capturing global contextual information. In addition, to take full advantage of the multi-scale features of the decoder, we propose a scale-aware deep supervision (SADS) module to achieve hierarchical iterative deep supervision. Comprehensive experimental results on the public dataset of ISIC 2017, ISIC 2018, and PH 2 show that our proposed method outperforms other state-of-the-art methods, demonstrating the efficacy of our method in skin lesion segmentation. Our code is available at https://github.com/YonghengSun1997/MSCA-Net .