Artificial intelligence links CT images to pathologic features and survival outcomes of renal masses

Treatment decisions for an incidental renal mass are mostly made with pathologic uncertainty. Improving the diagnosis of benign renal masses and distinguishing aggressive cancers from indolent ones is key to better treatment selection. We analyze 13261 pre-operative computed computed tomography (CT) volumes of 4557 patients. Two multi-phase convolutional neural networks are developed to predict the malignancy and aggressiveness of renal masses. The first diagnostic model designed to predict the malignancy of renal masses achieves area under the curve (AUC) of 0.871 in the prospective test set. This model surpasses the average performance of seven seasoned radiologists. The second diagnostic model differentiating aggressive from indolent tumors has AUC of 0.783 in the prospective test set. Both models outperform corresponding radiomics models and the nephrometry score nomogram. Here we show that the deep learning models can non-invasively predict the likelihood of malignant and aggressive pathology of a renal mass based on preoperative multi-phase CT images.

Authors

Institutions

Publication Details

Journal
Nature Communications
Published
2025-02-06
DOI
https://doi.org/10.1038/s41467-025-56784-z
Citations
20
Primary Topic
Renal cell carcinoma treatment
Type
article
Field-Weighted Citation Impact
30.37

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Artificial intelligence links CT images to pathologic features and survival outcomes of renal masses

Jianjun Zhou, Pingyi Zhu, Jinglai Lin, Jiaxi Yao et al.
20 citations
Nature Communications
Renal cell carcinoma treatment
30.37
article

Artificial intelligence links CT images to pathologic features and survival outcomes of renal masses

Jianjun Zhou, Pingyi Zhu, Jinglai Lin, Jiaxi Yao, Yuan Huang, Chenchen Dai, Qi Bai, Jianming Guo, Kang Wang, Risheng Huang, Haoran Wang, Feng Chen, Li Liu, Xue Zhang, Run Wang, Shuo Wang, Ying Xiong, Yu Qi, Linpeng Yao
article en
20 citations

Abstract

Treatment decisions for an incidental renal mass are mostly made with pathologic uncertainty. Improving the diagnosis of benign renal masses and distinguishing aggressive cancers from indolent ones is key to better treatment selection. We analyze 13261 pre-operative computed computed tomography (CT) volumes of 4557 patients. Two multi-phase convolutional neural networks are developed to predict the malignancy and aggressiveness of renal masses. The first diagnostic model designed to predict the malignancy of renal masses achieves area under the curve (AUC) of 0.871 in the prospective test set. This model surpasses the average performance of seven seasoned radiologists. The second diagnostic model differentiating aggressive from indolent tumors has AUC of 0.783 in the prospective test set. Both models outperform corresponding radiomics models and the nephrometry score nomogram. Here we show that the deep learning models can non-invasively predict the likelihood of malignant and aggressive pathology of a renal mass based on preoperative multi-phase CT images.

Nature CommunicationsVol. 16(1)
Fujian Medical University (CN), Shanghai Medical College of Fudan University (CN), Sun Yat-sen University (CN), Xiamen University (CN), University of Cambridge (GB), Fudan University (CN), Sir Run Run Shaw Hospital (CN), Xiamen Municipal Bureau of Science and Technology (CN), Hexi University (CN), Zhongshan Hospital of Xiamen University (CN), Zhongshan Hospital (CN), The First Affiliated Hospital, Sun Yat-sen University (CN), The First People’s Hospital of Lianyungang (CN), Shanghai Institute of Hematology (CN), First Affiliated Hospital Zhejiang University (CN)
Hexi University, National Natural Science Foundation of China, Fudan University, Universidade de Macau, Shanghai International Science and Technology, Medical Innovation Project of Fujian Province
Zero hunger
Openalex Percentile: Top 0%
Renal cell carcinoma treatment
30.37
DOIOpen Access PDFOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.