[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-seo-457196-105":3,"detail-sidebar-cat-0-en-105":81,"doc-detail-457196-en":130},{"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","artificial-intelligence-model-outperformed-experienced-clinicians-in-differentiating-the-aetiology-of-pneumonia-on-chest-computed-tomography-a-retrospective-study","Artificial intelligence model outperformed experienced clinicians in differentiating the aetiology of pneumonia on chest computed tomography - a retrospective study","","Rapid and precise aetiological diagnosis is crucial for managing pneumonia. This retrospective study developed and validated deep learning models to differentiate ten pneumonia aetiologies using chest computed tomography images. Using 1,091 patients with confirmed diagnoses (2015-10-01 to 2022-06-30) and an external test set of 183 non-overlapping patients, performance was assessed via AUC and Top-1/Top-2/Top-3 accuracy versus expert radiologists and pulmonologists. Results showed the LVM+ approach outperformed multiple baselines and demonstrated advantages in key disease identification, including PNTM, PTB, and PJP.",{"@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/healthcare/","Healthcare",3,{"item":32,"name":10,"@type":21,"position":33},"https://docshare.wps.com/document/artificial-intelligence-model-outperformed-experienced-clinicians-in-differentiating-the-aetiology-of-pneumonia-on-chest-computed-tomography-a-retrospective-study/457196/",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/artificial-intelligence-model-outperformed-experienced-clinicians-in-differentiating-the-aetiology-of-pneumonia-on-chest-computed-tomography-a-retrospective-study/457196.png","ImageObject",300,407,{"name":42,"@type":43},"วิน","Person",{"url":19,"name":45,"@type":46},"DocShare","Organization","application/pdf","2026-11-07","2026-09-30",true,{"@type":52,"interactionType":53,"userInteractionCount":55},"InteractionCounter",{"@type":54},"ViewAction",6,{"@type":57,"mainEntity":58},"FAQPage",[59,65,69],{"name":60,"@type":61,"acceptedAnswer":62},"What was the main goal of the study?","Question",{"text":63,"@type":64},"To develop and validate deep learning models that differentiate ten pneumonia aetiologies using chest computed tomography images, enabling rapid and accurate diagnosis.","Answer",{"name":66,"@type":61,"acceptedAnswer":67},"How were the deep learning models trained and evaluated?",{"text":68,"@type":64},"The study retrospectively enrolled 1,091 pneumonia patients with 1 of 10 definite aetiological diagnoses and trained two models, then validated internally and tested them on an external dataset of 183 non-overlapping patients. Performance used AUC and Top-1/Top-2/Top-3 accuracy.",{"name":70,"@type":61,"acceptedAnswer":71},"How did the AI models compare with experienced clinicians?",{"text":72,"@type":64},"The LVM combined with a non-imaging model (LVM+) showed greater average prediction performance than the DenseNet+ approach and outperformed radiologists’ and pulmonologists’ results using non-imaging data, with reported Top-1 AUC and accuracy metrics.","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},457196,1792180439,{"code":4,"msg":82,"data":83},"success",[84,88,92,96,101,105,109,114,119,122,126],{"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":55,"doc_module":4,"doc_module_name":25,"category_name":102,"show_sort_weight":103,"slug":104},"Technology",50,"technology",{"id":106,"doc_module":4,"doc_module_name":25,"category_name":29,"show_sort_weight":107,"slug":108},7,40,"healthcare",{"id":110,"doc_module":4,"doc_module_name":25,"category_name":111,"show_sort_weight":112,"slug":113},8,"Research & Report",30,"research-report",{"id":115,"doc_module":4,"doc_module_name":25,"category_name":116,"show_sort_weight":117,"slug":118},9,"Religion & Spirituality",20,"religion-spirituality",{"id":117,"doc_module":4,"doc_module_name":25,"category_name":120,"show_sort_weight":117,"slug":121},"World Cup","world-cup",{"id":123,"doc_module":4,"doc_module_name":25,"category_name":124,"show_sort_weight":123,"slug":125},10,"Lifestyle","lifestyle",{"id":127,"doc_module":4,"doc_module_name":25,"category_name":128,"show_sort_weight":97,"slug":129},19,"General","general",{"code":4,"msg":82,"data":131},{"doc_id":79,"user_id":132,"nickname":42,"user_avatar":133,"doc_module":4,"category_id":106,"category_name":29,"doc_title":10,"doc_description":12,"doc_content":134,"file_id":135,"file_url":136,"file_type":137,"file_size":138,"view_count":55,"is_deleted":4,"is_public":22,"is_downloadable":22,"audit_status":22,"page_count":127,"language":139,"language_code":8,"site_id":7,"html_lang":8,"table_of_contents":140,"faqs":141,"seo_title":142,"seo_description":12,"update_tm":143,"read_time":144,"preview_page_count":4},2336475104736,"https://ap-avatar.wpscdn.com/avatar/22000c4c5e0e5b17e70?x-image-process=image/resize,m_fixed,w_180,h_180&k=1786591360781797222","Original Article  \nArtificial intelligence model outperformed experienced clinicians in differentiating the aetiology of pneumonia on chest computed tomography: a retrospective study  \nWenting Jin 1\\#, Ying Shao2\\#, Jue Pan 1, Meixia Wang3, Tongjie Gu4, Wei Shen5, Xi Ouyang2, Zhi Qiao6, Dongdong Gu2, Zhen Qian6, Yaozong Gao2*, Bijie Hu1,7*  \n1Department of Infectious Diseases, Zhongshan Hospital, Fudan University, Shanghai, China; 2R&D Department, Shanghai United Imaging Intelligence Co., Ltd., Shanghai, China; 3Zhongshan Hospital (Xiamen), Fudan University, Xiamen, China; 4Department of Respiratory Medicine, Ningbo No. 2 Hospital, Ningbo, China; 5Department of Respiratory Medicine, Cixi No. 3 Hospital, Ningbo, China; 6Institute of Intelligent Diagnostics, Beijing United-Imaging Research Institute of Intelligent Imaging, Beijing, China; 7Shanghai Institute of Infectious Disease and Biosecurity, Shanghai, China  \nContributions: (I) Conception and design: B Hu, Y Gao, W Jin, J Pan; (II) Administrative support: B Hu, J Pan, Y Gao; (III) Provision of study materials or patients: W Jin; (IV) Collection and assembly of data: W Jin, Y Shao; (V) Data analysis and interpretation: M Wang, Y Shao, W Jin; (VI)  \nManuscript writing: All authors; (VII) Final approval of manuscript: All authors.  \n\\#These authors contributed equally to this work as co-first authors.  \n*These authors contributed equally to this work.  \nCorrespondence to: Yaozong Gao, PhD. R&D Department, Shanghai United Imaging Intelligence Co., Ltd., No. 701, Yunjin Road, Xuhui District, Shanghai 200232, China. Email: [yaozong.gao@uii-ai.com](yaozong.gao@uii-ai.com); Bijie Hu, PhD. Department of Infectious Diseases, Zhongshan Hospital, Fudan University, 180 Fenglin Road, Xuhui District, Shanghai 200032, China; Shanghai Institute of Infectious Disease and Biosecurity, Shanghai, China. [Email: hu.bijie@zs-hospital.sh.cn](Email: hu.bijie@zs-hospital.sh.cn).  \nBackground: Rapid and precise aetiological diagnosis is crucial for managing pneumonia. We aimed to develop and validate deep learning (DL) models for differentiating ten pneumonia aetiologies on chest  \ncomputed tomography images.  \nMethods: We enrolled 1,091 pneumonia patients with 1 of 10 definite aetiological diagnoses between October  \n1st, 2015 and June 30th, 2022 in this retrospective study. We trained and validated two DL models: a classic 3D-DenseNet model (DenseNet) and a novel large vision model (LVM) . The models were tested on a data from 183 nonoverlapping patients for external dataset. Model performance was assessed using the area under the curve (AUC) of the Top1 diagnosis and the accuracy of the Top1, Top2, and Top3 diagnoses. Comparisons  \nwere also performed between the DL models and eight experienced radiologists and pulmonologists.  \nResults: The LVM combined with non-imaging model (LVM+) had a greater average prediction performance than DenseNet combined with non-imaging model (DenseNet+), radiologists’ results with nonimaging data (radiologists+) and pulmonologists’ results with non-imaging data (pulmonologists+), with Top1 AUCs of 0.872, 0.851, 0.643 and 0.644, respectively. The Top1, Top2, and Top3 accuracies of LVM+ were 0.527, 0.701 and 0.820, respectively, similarly outperforming DenseNet+, radiologists+ and pulmonologists+ .  \nThe two models performed similarly in the external test sets, with the Top1 AUCs of 0.743 for DenseNet and 0.775 for LVM. The classification-related confusion matrix of LVM/DenseNet with or without nonimaging model showed a significant advantage in identifying pulmonary non-tuberculous mycobacterium  \npulmonary disease (PNTM), pulmonary tuberculosis (PTB) and Pneumocystis jirovecii pneumonia (PJP) .  \nConclusions: This study presents a comprehensive classification closely aligned with pneumonia diagnosis in realistic clinical settings. We expect this method to be applied clinically to foster novel approaches to  \n© AME Publishing Company. Quant Imaging Med Surg 2026;16(1):","cbCaiaqjK5UYJyrq","https://ap.wps.com/l/cbCaiaqjK5UYJyrq","pdf",3229914,"English","# Introduction\n## Background and clinical challenge\n## Role of chest CT and AI in pneumonia diagnosis\n# Methods\n## Study design and participants\n## Deep learning models and evaluation\n# Results\n## Performance vs baselines and clinicians\n## External validation and confusion-matrix findings\n# Conclusions","[{\"question\":\"What was the main goal of the study?\",\"answer\":\"To develop and validate deep learning models that differentiate ten pneumonia aetiologies using chest computed tomography images, enabling rapid and accurate diagnosis.\"},{\"question\":\"How were the deep learning models trained and evaluated?\",\"answer\":\"The study retrospectively enrolled 1,091 pneumonia patients with 1 of 10 definite aetiological diagnoses and trained two models, then validated internally and tested them on an external dataset of 183 non-overlapping patients. Performance used AUC and Top-1/Top-2/Top-3 accuracy.\"},{\"question\":\"How did the AI models compare with experienced clinicians?\",\"answer\":\"The LVM combined with a non-imaging model (LVM+) showed greater average prediction performance than the DenseNet+ approach and outperformed radiologists’ and pulmonologists’ results using non-imaging data, with reported Top-1 AUC and accuracy metrics.\"}]","Artificial intelligence model outperformed experienced clinicians in differentiating the aetiology of pneumonia on chest computed tomography - a retrospective study | PDF",1790748582,48]