Journal article
Large-scale pancreatic cancer detection via non-contrast CT and deep learning
Nature medicine, Vol.29(12), pp.3033-3043
12/01/2023
DOI: 10.1038/s41591-023-02640-w
PMCID: PMC10719100
PMID: 37985692
Abstract
Pancreatic ductal adenocarcinoma (PDAC), the most deadly solid malignancy, is typically detected late and at an inoperable stage. Early or incidental detection is associated with prolonged survival, but screening asymptomatic individuals for PDAC using a single test remains unfeasible due to the low prevalence and potential harms of false positives. Non-contrast computed tomography (CT), routinely performed for clinical indications, offers the potential for large-scale screening, however, identification of PDAC using non-contrast CT has long been considered impossible. Here, we develop a deep learning approach, pancreatic cancer detection with artificial intelligence (PANDA), that can detect and classify pancreatic lesions with high accuracy via non-contrast CT. PANDA is trained on a dataset of 3,208 patients from a single center. PANDA achieves an area under the receiver operating characteristic curve (AUC) of 0.986-0.996 for lesion detection in a multicenter validation involving 6,239 patients across 10 centers, outperforms the mean radiologist performance by 34.1% in sensitivity and 6.3% in specificity for PDAC identification, and achieves a sensitivity of 92.9% and specificity of 99.9% for lesion detection in a real-world multi-scenario validation consisting of 20,530 consecutive patients. Notably, PANDA utilized with non-contrast CT shows non-inferiority to radiology reports (using contrast-enhanced CT) in the differentiation of common pancreatic lesion subtypes. PANDA could potentially serve as a new tool for large-scale pancreatic cancer screening.
A deep learning model provides high accuracy in detecting pancreatic lesions in multicenter data, outperforming radiology specialists.
Details
- Title: Subtitle
- Large-scale pancreatic cancer detection via non-contrast CT and deep learning
- Creators
- Kai Cao - Shanghai Institute of HematologyYingda Xia - Alibaba GroupJiawen Yao - Zhejiang LabXu Han - First Affiliated Hospital Zhejiang UniversityLukas Lambert - General University Hospital in PragueTingting Zhang - XinHua HospitalWei Tang - Fudan University Shanghai Cancer CenterGang Jin - Shanghai Inst Pancreat Dis, Dept Surg, Shanghai, Peoples R ChinaHui Jiang - Shanghai Inst Pancreat Dis, Dept Pathol, Shanghai, Peoples R ChinaXu Fang - Shanghai Institute of HematologyIsabella Nogues - Harvard UniversityXuezhou Li - Shanghai Institute of HematologyWenchao Guo - Zhejiang LabYu Wang - Zhejiang LabWei Fang - Zhejiang LabMingyan Qiu - Alibaba GroupYang Hou - China Medical UniversityTomas Kovarnik - Charles UniversityMichal Vocka - General University Hospital in PragueYimei Lu - Fudan UniversityYingli Chen - Shanghai Inst Pancreat Dis, Dept Surg, Shanghai, Peoples R ChinaXin Chen - Guangdong Provincial People's HospitalZaiyi Liu - Guangdong Provincial People's HospitalJian Zhou - Sun Yat-sen UniversityChuanmiao Xie - Sun Yat-sen UniversityRong Zhang - Sun Yat-sen UniversityHong Lu - Tianjin Medical University Cancer Institute and HospitalGregory D. Hager - Johns Hopkins UniversityAlan L. Yuille - Johns Hopkins UniversityLe Lu - Alibaba GroupChengwei Shao - Shanghai Institute of HematologyYu Shi - China Medical UniversityQi Zhang - Zhejiang UniversityTingbo Liang - First Affiliated Hospital Zhejiang UniversityLing Zhang - Alibaba GroupJianping Lu - Shanghai Institute of Hematology
- Resource Type
- Journal article
- Publication Details
- Nature medicine, Vol.29(12), pp.3033-3043
- Publisher
- NATURE PORTFOLIO
- DOI
- 10.1038/s41591-023-02640-w
- PMID
- 37985692
- PMCID
- PMC10719100
- ISSN
- 1078-8956
- eISSN
- 1546-170X
- Number of pages
- 33
- Grant note
- National Natural Science Foundation of China; National Natural Science Foundation of China (NSFC) The authors acknowledge external clinical validation contributions provided by Chang Gung Memorial Hospital (CGMH), and T.-C. Yen at CGMH for guiding the manuscript's revision. The authors thank R. M. Summers at the National Institutes of Health Clinical C 82372045 / Chang Gung Memorial Hospital (CGMH); Chang Gung Memorial Hospital
- Language
- English
- Date published
- 12/01/2023
- Academic Unit
- Electrical and Computer Engineering
- Record Identifier
- 9984627328202771
Metrics
4 Record Views