[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-81746-en":3,"doc-seo-81746-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":13,"seo_description":14,"update_tm":28,"read_time":29},81746,4398048949847,"Eliana","https://ap-avatar.wpscdn.com/avatar/400002536579ef2da7f?_k=1778318612642679267",8,"Research & Report","Joint Medical Image Enhancement and Segmentation with Diffusion-based Symbiotic Information Interaction","Medical image quality underpins accurate diagnosis, yet MRI, CT, and ultrasound scans are often limited by cost-related constraints that reduce resolution, sharpness, and signal quality, making anatomical structures and lesions harder to visualize. Traditional pipelines enhance images as a separate preprocessing step and miss synergy with segmentation. DiSIINet unifies both tasks within a diffusion framework, using DDIM and a symbiotic information interaction module to enable cross-attention feature exchange during reverse diffusion for iterative mutual improvement. Experiments on multi-modal datasets show substantial gains over sequential or independent baselines.","Joint Medical Image Enhancement and Segmentation with Diffusion-based  \nSymbiotic Information Interaction  \nYing Chen 1 , Jinyue Li2 , Qiankun Li3†  \n1 Shenzhen Research Institute, The Chinese University of Hong Kong  \n2University of Science and Technology of China  \n3Imperial Global Singapore, Imperial College London  \n[q.li2@imperial.ac.uk](q.li2@imperial.ac.uk)  \narXiv :2607 .00058v1 [ cs .CV] 30 Jun 2026  \nAbstract  \nImage quality is critical for accurate medical diagnosis. However, MRI, CT, and ultrasound images are often of low resolution and quality due to cost constraints, complicating the visualization of key anatomical structures and lesions. While such limitations are common in practice, traditional methods treat image enhancement as a separate preprocessing step, failing to fully leverage its potential synergy with image segmentation. To address this, we propose DiSIINet (Diffusion-based Symbiotic Information Interaction Network), which is built on the principle that enhancement and segmentation should mutually reinforce each other ina unified model. Based on Denoising Diffusion Implicit Models (DDIM), DiSIINet integrates an enhancement branch and a segmentation branch.  \nThese branches interact through a novel Symbiotic Information Interaction (SII) module, which facilitates dynamic, feature-level information exchange via cross-attention during the reverse diffusion process. This design enables both tasks to iteratively improve each other. The DDIM backbone ensures high-quality output and efficient inference through deterministic sampling. Experiments on multi-modal medical datasets (MRI, CT, ultrasound) show that DiSIINet achieves significant performance improvements compared to sequential or independent enhancement and segmen  \ntation approaches. The code is available at: [https:](https:)//[github.com/Reconsider80/DiSIINet](github.com/Reconsider80/DiSIINet).  \n1 Introduction  \nMedical images acquired under suboptimal conditions may encounter quality issues, including blurriness, poor lighting, low resolution, and noise, which can result in misdiagnosis. While advanced medical image enhancement techniques often face challenges in improving high-resolution image quality while preserving clear local anatomical features. Traditional image enhancement methods, particularly those  \n1Corresponding author†([q.li2@imperial.ac.uk](q.li2@imperial.ac.uk)) .  \nbased on transformers, typically operate under the assumption of known degradation types, which often leads to overly smoothed results. StillGAN [Ma et al., 2021] introducesan innovative and versatile bi-directional GAN for enhancing medical image quality. Although there have been significant advancements in improving visual perception, methods based on generative adversarial networks (GANs) [Liang et al., 2022; Park et al., 2023] still encounter challenges in balancing perceptual quality with fidelity, often causing artifacts.  \nRecently, the emergence of diffusion models (DMs) [Ho et al., 2020] has demonstrated remarkable capabilities in approximating intricate distributions and generating realistic images. Stable Diffusion [Rombach et al., 2022] is a conditional diffusion model, whose core concept lies in being a generation system centrally guided by conditional input. Among diffusion-based super-resolution techniques, StableSR [Wang et al., 2023] employs the latent representation of low-quality (LQ) images as the guiding condition for StableDiffusion to achieve super-resolution. In contrast, DiffBIR [Lin et al., 2024] first restores LQ images before using generative priors to maintain a balance between quality and fidelity during the diffusion process. These approaches highlight the significant potential of generative priors in SR tasks; however, relying solely on LQ image data without additional semantic control may lead to inaccurate content reconstruction. Given the inherent advantages of textual prompts in directing generation within pretrained text-to-image mo","cbCailIoO5eQZsNQ","https://ap.wps.com/l/cbCailIoO5eQZsNQ","pdf",2940839,3,1,9,"English","en",105,"# Abstract\n# Introduction","[{\"question\":\"Why is joint enhancement and segmentation important for medical imaging?\",\"answer\":\"Medical scans often suffer from low resolution and quality, which can obscure anatomical structures and lesions. Traditional approaches treat enhancement separately from segmentation, missing opportunities for the two tasks to benefit each other during restoration and interpretation.\"},{\"question\":\"How does DiSIINet combine enhancement and segmentation?\",\"answer\":\"DiSIINet uses a dual diffusion architecture with an enhancement branch and a segmentation branch. The branches interact through a Symbiotic Information Interaction (SII) module that performs cross-attention feature exchange during the reverse diffusion process.\"},{\"question\":\"What role does DDIM play in DiSIINet?\",\"answer\":\"DiSIINet is built on denoising diffusion implicit models (DDIM), which supports high-quality outputs while enabling efficient inference via deterministic sampling.\"}]",1784175797,23,{"code":4,"msg":31,"data":32},"ok",{"site_id":25,"language":24,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"joint-medical-image-enhancement-and-segmentation-with-diffusion-based-symbiotic-information-interaction","",{"@graph":36,"@context":85},[37,53,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,50],{"item":41,"name":42,"@type":43,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":20},"https://docshare.wps.com/document/research-report/",{"item":51,"name":13,"@type":43,"position":52},"https://docshare.wps.com/document/joint-medical-image-enhancement-and-segmentation-with-diffusion-based-symbiotic-information-interaction/81746/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":24,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":41,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-25","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"Why is joint enhancement and segmentation important for medical imaging?","Question",{"text":75,"@type":76},"Medical scans often suffer from low resolution and quality, which can obscure anatomical structures and lesions. Traditional approaches treat enhancement separately from segmentation, missing opportunities for the two tasks to benefit each other during restoration and interpretation.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does DiSIINet combine enhancement and segmentation?",{"text":80,"@type":76},"DiSIINet uses a dual diffusion architecture with an enhancement branch and a segmentation branch. The branches interact through a Symbiotic Information Interaction (SII) module that performs cross-attention feature exchange during the reverse diffusion process.",{"name":82,"@type":73,"acceptedAnswer":83},"What role does DDIM play in DiSIINet?",{"text":84,"@type":76},"DiSIINet is built on denoising diffusion implicit models (DDIM), which supports high-quality outputs while enabling efficient inference via deterministic sampling.","https://schema.org",{"og:url":51,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":51},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,127,130,134],{"id":21,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":22,"doc_module":4,"doc_module_name":46,"category_name":124,"show_sort_weight":125,"slug":126},"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":46,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":46,"category_name":136,"show_sort_weight":106,"slug":137},19,"General","general"]