[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-seo-368008-105":3,"detail-sidebar-cat-0-en-105":81,"doc-detail-368008-en":131},{"code":4,"msg":5,"data":6},0,"ok",{"site_id":7,"language":8,"slug":9,"title":10,"keywords":11,"description":12,"schema_data":13,"social_meta":74,"head_meta":76,"extra_data":78,"updated_unix":80},105,"en","3d-fusionnet-for-synthetic-ct-based-lung-cancer-segmentation","3D FusionNet for synthetic CT based lung cancer segmentation","","A 3D segmentation framework is presented for lung CT images to improve computer-aided diagnosis of lung cancer. The approach combines a deep convolutional generative adversarial network (DCGAN) with the 3D-TDUnet++ architecture and uses non-local feature aggregation (NLFa) to enhance attention to relevant volumetric regions. DCGAN generates additional synthetic CT samples, alleviating limited annotated data and enriching training diversity, including reduced class imbalance. Experiments on a publicly available Chest CT scan dataset from Kaggle, augmented with DCGAN-generated images, yield Dice 88.94%, F1-score 88.94%, and accuracy 93.37%, outperforming baseline models such as DenseNet, ResNet, and MDDNet-ASPP.",{"@graph":14,"@context":73},[15,34,56],{"@type":16,"itemListElement":17},"BreadcrumbList",[18,23,27,31],{"item":19,"name":20,"@type":21,"position":22},"https://docshare.wps.com","Home","ListItem",1,{"item":24,"name":25,"@type":21,"position":26},"https://docshare.wps.com/document/","Document",2,{"item":28,"name":29,"@type":21,"position":30},"https://docshare.wps.com/document/research-report/","Research & Report",3,{"item":32,"name":10,"@type":21,"position":33},"https://docshare.wps.com/document/3d-fusionnet-for-synthetic-ct-based-lung-cancer-segmentation/368008/",4,{"url":32,"name":10,"@type":35,"image":36,"author":41,"headline":10,"publisher":44,"fileFormat":47,"inLanguage":8,"description":12,"dateModified":48,"datePublished":49,"encodingFormat":47,"isAccessibleForFree":50,"interactionStatistic":51},"DigitalDocument",{"url":37,"@type":38,"width":39,"height":40},"https://docshare.wps.com/thumbnails/3d-fusionnet-for-synthetic-ct-based-lung-cancer-segmentation/368008.png","ImageObject",300,407,{"name":42,"@type":43},"OmBimo","Person",{"url":19,"name":45,"@type":46},"DocShare","Organization","application/pdf","2026-11-02","2026-09-23",true,{"@type":52,"interactionType":53,"userInteractionCount":55},"InteractionCounter",{"@type":54},"ViewAction",15,{"@type":57,"mainEntity":58},"FAQPage",[59,65,69],{"name":60,"@type":61,"acceptedAnswer":62},"Why are synthetic CT images used in the proposed 3D FusionNet framework?","Question",{"text":63,"@type":64},"Synthetic CT images generated by DCGAN are used to reduce insufficient annotated data, increase training diversity, and mitigate class imbalance, improving robustness and generalization.","Answer",{"name":66,"@type":61,"acceptedAnswer":67},"What roles do DCGAN and 3D-TDUnet++ play in the hybrid system?",{"text":68,"@type":64},"DCGAN generates realistic synthetic lung images, while 3D-TDUnet++ performs the volumetric lung segmentation. Non-local feature aggregation (NLFa) supports improved attention for segmentation accuracy.",{"name":70,"@type":61,"acceptedAnswer":71},"What performance metrics does 3D-FusionNet achieve in the experiments?",{"text":72,"@type":64},"The reported quantitative results include a Dice coefficient of 88.94%, an F1-score of 88.94%, and an accuracy of 93.37% on the augmented dataset.","https://schema.org",{"og:url":32,"og:type":75,"og:title":10,"og:site_name":45,"og:description":12},"article",{"robots":77,"canonical":32},"index,follow",{"doc_id":79,"site_id":7},368008,1790220191,{"code":4,"msg":82,"data":83},"success",[84,88,92,96,101,106,111,115,120,123,127],{"id":22,"doc_module":4,"doc_module_name":25,"category_name":85,"show_sort_weight":86,"slug":87},"Story & Novel",90,"story-novel",{"id":26,"doc_module":4,"doc_module_name":25,"category_name":89,"show_sort_weight":90,"slug":91},"Literature",80,"literature",{"id":33,"doc_module":4,"doc_module_name":25,"category_name":93,"show_sort_weight":94,"slug":95},"Exam",70,"exam",{"id":97,"doc_module":4,"doc_module_name":25,"category_name":98,"show_sort_weight":99,"slug":100},5,"Comic",60,"comic",{"id":102,"doc_module":4,"doc_module_name":25,"category_name":103,"show_sort_weight":104,"slug":105},6,"Technology",50,"technology",{"id":107,"doc_module":4,"doc_module_name":25,"category_name":108,"show_sort_weight":109,"slug":110},7,"Healthcare",40,"healthcare",{"id":112,"doc_module":4,"doc_module_name":25,"category_name":29,"show_sort_weight":113,"slug":114},8,30,"research-report",{"id":116,"doc_module":4,"doc_module_name":25,"category_name":117,"show_sort_weight":118,"slug":119},9,"Religion & Spirituality",20,"religion-spirituality",{"id":118,"doc_module":4,"doc_module_name":25,"category_name":121,"show_sort_weight":118,"slug":122},"World Cup","world-cup",{"id":124,"doc_module":4,"doc_module_name":25,"category_name":125,"show_sort_weight":124,"slug":126},10,"Lifestyle","lifestyle",{"id":128,"doc_module":4,"doc_module_name":25,"category_name":129,"show_sort_weight":97,"slug":130},19,"General","general",{"code":4,"msg":82,"data":132},{"doc_id":79,"user_id":133,"nickname":42,"user_avatar":134,"doc_module":4,"category_id":112,"category_name":29,"doc_title":10,"doc_description":12,"doc_content":135,"file_id":136,"file_url":137,"file_type":138,"file_size":139,"view_count":55,"is_deleted":4,"is_public":22,"is_downloadable":22,"audit_status":22,"page_count":140,"language":141,"language_code":8,"site_id":7,"html_lang":8,"table_of_contents":142,"faqs":143,"seo_title":144,"seo_description":12,"update_tm":145,"read_time":109,"preview_page_count":33,"preview_urls":146},962090893581,"https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c","[www. nature.com/scientificreports](www. nature.com/scientificreports)  \nOPEN  \n3D FusionNet for synthetic CT based lung cancer segmentation  \nChiranjib Parida1, Saurabh Bilgaiyan1, Santwana Sagnika1, Satyabrata Roy2 & Ankur Pandey2􀀍  \nInterpretation of three-dimensional medical scans continues to be a demanding task in computerassisted diagnostic systems. In this study we present a segmentation framework specifically designed for lung CT images. Since lung cancer is one of the most fatal cancers worldwide and the demand for better detection tools remains pressing. To address this issue our proposed work combined the Deep Convolutional Generative Adversarial Networks (DCGAN) with the 3D-TDUnet++ architecture. The use of DCGAN allowed us to generate additional CT samples. Which results in reducing the problem of insufficient annotated data and improving the robustness of the training set. In this experiment the Chest CT-Scan images Dataset is used which is publicly available in KAGGLE. Which is being further enriched with synthetic images created by DCGAN. While being trained on this augmented dataset the proposed 3D-FusionNet model achieved superior performance in cancer detection. Which shows higher accuracy, sensitivity and specificity than other conventional approaches. The integration of DCGAN with 3D-TDUnet++ together with Non-Local Feature Aggregation (NLFa) made the system promising for clinical environments where limited data often restricts the application of deep learning models. Quantitative evaluation reveals that the proposed 3D-FusionNet achieves a Dice coefficient  \nof 88.94%, F1-score of 88.94% and an accuracy of 93.37%. Outperforming the benchmark models such as DenseNet, ResNet and MDDNet-ASPP. This hybrid design leverages the strengths of generative augmentation and non-local attention to improve both segmentation precision and clinical robustness. Thus the proposed 3D-FusionNet model demonstrates how deep generative modeling and volumetric segmentation can be integrated into a cohesive machine vision system. Which is capable of scalable and efficient deployment in real-world diagnostic pipelines.  \nKeywords Deep Convolutional Generative Adversarial Networks (DCGAN), 3D-TDUnet++, Non-Local Feature Aggregation (NLFa), Synthetic CT images, computer-aided diagnosis (CAD), Cross-domain image synthesis  \nLung cancer is the most prevalent and fatal diseases worldwide. Which contributes heavily to cancer-related deaths globally1. Where early diagnosis plays a crucial role in improving patient survival and treatment outcomes2. Which remains a major global health concern. Yet detecting lung cancer at an early stage is still a major challenge. Because of the symptoms that are often vague, mild or inconsistent3. These kind of difficulties emphasizes the urgent need for more advanced and reliable diagnostic techniques.  \nLung Cancer Detection is mainly based on the use of imaging technologies (CT scans/PET scans)4, which have had a significant impact on both early diagnosis and clinical decision making. These technologies however come with some disadvantages: high costs, the risk of radiating patients, the need for high quality and well annotated databases. Additionally, the low volume of available data, along with the complexity of the anatomy of lung tissue reduces the performance of traditional deep learning and machine learning-based predictive models5. The potential in traditional deep learning and machine learning-based predictive models is very great. However, the performance of these models continues to be affected by a lack of sufficient high-quality data to train them. To combat these issues this research will use a combination of synthetic image generation with deep convolutional generative adversarial networks (dcgan)6 and the state-of-the-art segmentation capabilities of the 3D-TDUnet++ model7. The DCGAN component generates highly realistic synthetic lung images which aids in increasing the diversity of the","cbCaidCEQHpi48tQ","https://ap.wps.com/l/cbCaidCEQHpi48tQ","pdf",3736117,16,"English","# Introduction\n## Problem of early lung cancer detection and data scarcity\n## Imaging-based diagnosis and limitations\n# Proposed Method\n## DCGAN-based synthetic image generation\n## 3D-TDUnet++ hybrid segmentation and NLFa\n# Objectives and Novelty\n## Main objectives of the research\n## Novelty emphasis and integrated training loop","[{\"question\":\"Why are synthetic CT images used in the proposed 3D FusionNet framework?\",\"answer\":\"Synthetic CT images generated by DCGAN are used to reduce insufficient annotated data, increase training diversity, and mitigate class imbalance, improving robustness and generalization.\"},{\"question\":\"What roles do DCGAN and 3D-TDUnet++ play in the hybrid system?\",\"answer\":\"DCGAN generates realistic synthetic lung images, while 3D-TDUnet++ performs the volumetric lung segmentation. Non-local feature aggregation (NLFa) supports improved attention for segmentation accuracy.\"},{\"question\":\"What performance metrics does 3D-FusionNet achieve in the experiments?\",\"answer\":\"The reported quantitative results include a Dice coefficient of 88.94%, an F1-score of 88.94%, and an accuracy of 93.37% on the augmented dataset.\"}]","3D FusionNet for synthetic CT based lung cancer segmentation | PDF",1790174302,[147,151,155,159],{"page_no":22,"url":148,"width":149,"height":150},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/65670c76ec78f0367af91155d317f6e4.webp",983,1291,{"page_no":26,"url":152,"width":153,"height":154},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/89fa8d3bec5626f8a70c85e685d92209.webp",982,1290,{"page_no":30,"url":156,"width":157,"height":158},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/94f328b33623cf8b57178dc1cc7587ae.webp",981,1289,{"page_no":33,"url":160,"width":161,"height":162},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/3542b2e8b259bc1bbbe6fef64b965702.webp",980,1287]