Development and external validation of the multichannel deep learning model based on unenhanced CT for differentiating fat-poor angiomyolipoma from renal cell carcinoma: a two-center retrospective study

被引:5
|
作者
Yao, Haohua [1 ,2 ]
Tian, Li [3 ]
Liu, Xi [1 ]
Li, Shurong [4 ]
Chen, Yuhang [1 ]
Cao, Jiazheng [5 ]
Zhang, Zhiling [6 ]
Chen, Zhenhua [1 ]
Feng, Zihao [1 ]
Xu, Quanhui [1 ]
Zhu, Jiangquan [1 ]
Wang, Yinghan [1 ]
Guo, Yan [4 ]
Chen, Wei [1 ]
Li, Caixia [7 ]
Li, Peixing [7 ]
Wang, Huanjun [4 ]
Luo, Junhang [1 ]
机构
[1] Sun Yat Sen Univ, Affiliated Hosp 1, Dept Urol, Guangzhou, Peoples R China
[2] Guangdong Acad Med Sci, Guangdong Prov Peoples Hosp, Dept Urol, Guangzhou, Peoples R China
[3] Sun Yat Sen Univ, Collaborat Innovat Ctr Canc Med, Dept Med Imaging, State Key Lab Oncol South China,Canc Ctr, Guangzhou, Peoples R China
[4] Sun Yat Sen Univ, Affiliated Hosp 1, Dept Radiol, Guangzhou, Peoples R China
[5] Jiangmen Cent Hosp, Dept Urol, Jiangmen, Peoples R China
[6] Sun Yat Sen Univ, Collaborat Innovat Ctr Canc Med, Dept Urol, State Key Lab Oncol South China,Canc Ctr, Guangzhou, Peoples R China
[7] Sun Yat Sen Univ, Sch Math & Computat Sci, Guangzhou, Peoples R China
基金
中国国家自然科学基金;
关键词
Renal cell carcinoma; Fat-poor angiomyolipoma; Urology; Deep learning; Computed tomography; TEXTURE ANALYSIS; CORTICAL TUMORS; MINIMAL FAT; CLASSIFICATION; DIAGNOSIS; FEATURES; SUBTYPES; IMAGES; MASSES;
D O I
10.1007/s00432-023-05339-0
中图分类号
R73 [肿瘤学];
学科分类号
100214 ;
摘要
PurposeThere are undetectable levels of fat in fat-poor angiomyolipoma. Thus, it is often misdiagnosed as renal cell carcinoma. We aimed to develop and evaluate a multichannel deep learning model for differentiating fat-poor angiomyolipoma (fp-AML) from renal cell carcinoma (RCC).MethodsThis two-center retrospective study included 320 patients from the First Affiliated Hospital of Sun Yat-Sen University (FAHSYSU) and 132 patients from the Sun Yat-Sen University Cancer Center (SYSUCC). Data from patients at FAHSYSU were divided into a development dataset (n = 267) and a hold-out dataset (n = 53). The development dataset was used to obtain the optimal combination of CT modality and input channel. The hold-out dataset and SYSUCC dataset were used for independent internal and external validation, respectively.ResultsIn the development phase, models trained on unenhanced CT images performed significantly better than those trained on enhanced CT images based on the fivefold cross-validation. The best patient-level performance, with an average area under the receiver operating characteristic curve (AUC) of 0.951 & PLUSMN; 0.026 (mean & PLUSMN; SD), was achieved using the "unenhanced CT and 7-channel" model, which was finally selected as the optimal model. In the independent internal and external validation, AUCs of 0.966 (95% CI 0.919-1.000) and 0.898 (95% CI 0.824-0.972), respectively, were obtained using the optimal model. In addition, the performance of this model was better on large tumors (& GE; 40 mm) in both internal and external validation.ConclusionThe promising results suggest that our multichannel deep learning classifier based on unenhanced whole-tumor CT images is a highly useful tool for differentiating fp-AML from RCC.
引用
收藏
页码:15827 / 15838
页数:12
相关论文
共 49 条
  • [1] Development and external validation of the multichannel deep learning model based on unenhanced CT for differentiating fat-poor angiomyolipoma from renal cell carcinoma: a two-center retrospective study
    Haohua Yao
    Li Tian
    Xi Liu
    Shurong Li
    Yuhang Chen
    Jiazheng Cao
    Zhiling Zhang
    Zhenhua Chen
    Zihao Feng
    Quanhui Xu
    Jiangquan Zhu
    Yinghan Wang
    Yan Guo
    Wei Chen
    Caixia Li
    Peixing Li
    Huanjun Wang
    Junhang Luo
    Journal of Cancer Research and Clinical Oncology, 2023, 149 : 15827 - 15838
  • [2] A convention-radiomics CT nomogram for differentiating fat-poor angiomyolipoma from clear cell renal cell carcinoma
    Ma, Yanqing
    Ma, Weijun
    Xu, Xiren
    Guan, Zheng
    Pang, Peipei
    SCIENTIFIC REPORTS, 2021, 11 (01)
  • [3] A convention-radiomics CT nomogram for differentiating fat-poor angiomyolipoma from clear cell renal cell carcinoma
    Yanqing Ma
    Weijun Ma
    Xiren Xu
    Zheng Guan
    Peipei Pang
    Scientific Reports, 11
  • [4] Review of Value of CT Texture Analysis and Machine Learning in Differentiating Fat-Poor Renal Angiomyolipoma fromRenal Cell Carcinoma
    Zhang, Yuhan
    Li, Xu
    Lv, Yang
    Gu, Xinquan
    TOMOGRAPHY, 2020, 6 (04) : 325 - 332
  • [5] Can Quantitative CT Texture Analysis be Used to Differentiate Fat-poor Renal Angiomyolipoma from Renal Cell Carcinoma on Unenhanced CT Images?
    Hodgdon, Taryn
    McInnes, Matthew D. F.
    Schieda, Nicola
    Flood, Trevor A.
    Lamb, Leslie
    Thornhill, Rebecca E.
    RADIOLOGY, 2015, 276 (03) : 787 - 796
  • [6] Re: Can Quantitative CT Texture Analysis be Used to Differentiate Fat-Poor Renal Angiomyolipoma from Renal Cell Carcinoma on Unenhanced CT Images?
    Siegel, Cary
    JOURNAL OF UROLOGY, 2016, 196 (06): : 1636 - 1636
  • [7] Utility of radiomics features of diffusion-weighted magnetic resonance imaging for differentiation of fat-poor angiomyolipoma from clear cell renal cell carcinoma: model development and external validation
    Shunya Matsumoto
    Yuki Arita
    Soichiro Yoshida
    Hiroshi Fukushima
    Koichiro Kimura
    Ichiro Yamada
    Hajime Tanaka
    Fumiko Yagi
    Minato Yokoyama
    Yoh Matsuoka
    Mototsugu Oya
    Ukihide Tateishi
    Masahiro Jinzaki
    Yasuhisa Fujii
    Abdominal Radiology, 2022, 47 : 2178 - 2186
  • [8] Utility of radiomics features of diffusion-weighted magnetic resonance imaging for differentiation of fat-poor angiomyolipoma from clear cell renal cell carcinoma: model development and external validation
    Matsumoto, Shunya
    Arita, Yuki
    Yoshida, Soichiro
    Fukushima, Hiroshi
    Kimura, Koichiro
    Yamada, Ichiro
    Tanaka, Hajime
    Yagi, Fumiko
    Yokoyama, Minato
    Matsuoka, Yoh
    Oya, Mototsugu
    Tateishi, Ukihide
    Jinzaki, Masahiro
    Fujii, Yasuhisa
    ABDOMINAL RADIOLOGY, 2022, 47 (06) : 2178 - 2186
  • [9] An Effective Radiomics Model for Noninvasive Discrimination of Fat-poor Angiomyolipoma from Clear Cell Renal Cell Carcinoma
    Yang, Yi
    Zhu, Jianbing
    Zhou, Zhiyong
    Dai, Yakang
    Qian, Xusheng
    2019 IEEE SYMPOSIUM SERIES ON COMPUTATIONAL INTELLIGENCE (IEEE SSCI 2019), 2019, : 1551 - 1558
  • [10] Alkaline phosphatase combines with CT factors for differentiating small (≤ 4 cm) fat-poor angiomyolipoma from renal cell carcinoma: a multiple quantitative tool
    Tianming Peng
    Junhong Fan
    Binyang Xie
    Qianqian Wang
    Yuchun Chen
    Yong Li
    Kunlin Wu
    Chunxiang Feng
    Teng Li
    Hanzhong Chen
    Xiaoyong Pu
    Jiumin Liu
    World Journal of Urology, 2023, 41 : 1345 - 1351