Xia, P., Lv, Z., Wen, Y., Zhang, B., Zhao, X., Zhang, B., Wang, Y., Cui, H., Wang, C., Zheng, H., Qin, Y., Sun, L., Ye, N., Cheng, H., Yao, L., Zhou, H., Zhen, J., Hu, Z., Zhu, W., ... Chen, L. (2023). Development of a multiple convolutional neural network–facilitated diagnostic screening program for immunofluorescence images of IgA nephropathy and idiopathic membranous nephropathy. CKJ: Clinical Kidney Journal, 16(12), 2503-2513. https://doi.org/10.1093/ckj/sfad153
Xia, Peng ; Lv, Zhilong ; Wen, Yubing 等. / Development of a multiple convolutional neural network–facilitated diagnostic screening program for immunofluorescence images of IgA nephropathy and idiopathic membranous nephropathy. 在: CKJ: Clinical Kidney Journal. 2023 ; 卷 16, 号码 12. 页码 2503-2513.
@article{c8f1d3a4acc24a1ea2f62937310fa861,
title = "Development of a multiple convolutional neural network–facilitated diagnostic screening program for immunofluorescence images of IgA nephropathy and idiopathic membranous nephropathy",
abstract = "Background. Immunoglobulin A nephropathy (IgAN) and idiopathic membranous nephropathy (IMN) are the most common glomerular diseases. Immunofluorescence (IF) tests of renal tissues are crucial for the diagnosis. We developed a multiple convolutional neural network (CNN) -facilitated diagnostic program to assist the IF diagnosis of IgAN and IMN. Methods. The diagnostic program consisted of four parts: a CNN trained as a glomeruli detection module, an IF intensity comparator, dual-CNN (D-CNN) trained as a deposition appearance and location classifier and a post-processing module. A total of 1573 glomerular IF images from 1009 patients with glomerular diseases were used for the training and validation of the diagnostic program. A total of 1610 images of 426 patients from different hospitals were used as test datasets. The performance of the diagnostic program was compared with nephropathologists. Results. In > 90% of the tested images, the glomerulus location module achieved an intersection over union > 0.8. The accuracy of the D-CNN in recognizing irregular granular mesangial deposition and fine granular deposition along the glomerular basement membrane was 96.1% and 93.3%, respectively. As for the diagnostic program, the accuracy, sensitivity and specificity of diagnosing suspected IgAN were 97.6%, 94.4% and 96.0%, respectively. The accuracy, sensitivity and specificity of diagnosing suspected IMN were 91.7%, 88.9% and 95.8%, respectively. The corresponding areas under the curve (AUCs) were 0.983 and 0.935. When tested with images from the outside hospital, the diagnostic program showed stable performance. The AUCs for diagnosing suspected IgAN and IMN were 0.972 and 0.948, respectively. Compared with inexperienced nephropathologists, the program showed better performance. Conclusion. The proposed diagnostic program could assist the IF diagnosis of IgAN and IMN.",
keywords = "IgA nephropathy, convolutional neural network, idiopathic membranous nephropathy, immunofluorescence",
author = "Peng Xia and Zhilong Lv and Yubing Wen and Baichuan Zhang and Xuesong Zhao and Boyao Zhang and Ying Wang and Haoyuan Cui and Chuanpeng Wang and Hua Zheng and Yan Qin and Lijun Sun and Nan Ye and Hong Cheng and Li Yao and Hua Zhou and Junhui Zhen and Zhao Hu and Weiguo Zhu and Fa Zhang and Xuemei Li and Fei Ren and Limeng Chen",
note = "Publisher Copyright: {\textcopyright} The Author (s) 2023. Published by Oxford University Press on behalf of the ERA.",
year = "2023",
month = dec,
day = "1",
doi = "10.1093/ckj/sfad153",
language = "English",
volume = "16",
pages = "2503--2513",
journal = "CKJ: Clinical Kidney Journal",
issn = "2048-8505",
publisher = "Oxford University Press",
number = "12",
}
Xia, P, Lv, Z, Wen, Y, Zhang, B, Zhao, X, Zhang, B, Wang, Y, Cui, H, Wang, C, Zheng, H, Qin, Y, Sun, L, Ye, N, Cheng, H, Yao, L, Zhou, H, Zhen, J, Hu, Z, Zhu, W, Zhang, F, Li, X, Ren, F & Chen, L 2023, 'Development of a multiple convolutional neural network–facilitated diagnostic screening program for immunofluorescence images of IgA nephropathy and idiopathic membranous nephropathy', CKJ: Clinical Kidney Journal, 卷 16, 号码 12, 页码 2503-2513. https://doi.org/10.1093/ckj/sfad153
Development of a multiple convolutional neural network–facilitated diagnostic screening program for immunofluorescence images of IgA nephropathy and idiopathic membranous nephropathy. / Xia, Peng; Lv, Zhilong; Wen, Yubing 等.
在:
CKJ: Clinical Kidney Journal, 卷 16, 号码 12, 01.12.2023, 页码 2503-2513.
科研成果: 期刊稿件 › 文章 › 同行评审
TY - JOUR
T1 - Development of a multiple convolutional neural network–facilitated diagnostic screening program for immunofluorescence images of IgA nephropathy and idiopathic membranous nephropathy
AU - Xia, Peng
AU - Lv, Zhilong
AU - Wen, Yubing
AU - Zhang, Baichuan
AU - Zhao, Xuesong
AU - Zhang, Boyao
AU - Wang, Ying
AU - Cui, Haoyuan
AU - Wang, Chuanpeng
AU - Zheng, Hua
AU - Qin, Yan
AU - Sun, Lijun
AU - Ye, Nan
AU - Cheng, Hong
AU - Yao, Li
AU - Zhou, Hua
AU - Zhen, Junhui
AU - Hu, Zhao
AU - Zhu, Weiguo
AU - Zhang, Fa
AU - Li, Xuemei
AU - Ren, Fei
AU - Chen, Limeng
N1 - Publisher Copyright:
© The Author (s) 2023. Published by Oxford University Press on behalf of the ERA.
PY - 2023/12/1
Y1 - 2023/12/1
N2 - Background. Immunoglobulin A nephropathy (IgAN) and idiopathic membranous nephropathy (IMN) are the most common glomerular diseases. Immunofluorescence (IF) tests of renal tissues are crucial for the diagnosis. We developed a multiple convolutional neural network (CNN) -facilitated diagnostic program to assist the IF diagnosis of IgAN and IMN. Methods. The diagnostic program consisted of four parts: a CNN trained as a glomeruli detection module, an IF intensity comparator, dual-CNN (D-CNN) trained as a deposition appearance and location classifier and a post-processing module. A total of 1573 glomerular IF images from 1009 patients with glomerular diseases were used for the training and validation of the diagnostic program. A total of 1610 images of 426 patients from different hospitals were used as test datasets. The performance of the diagnostic program was compared with nephropathologists. Results. In > 90% of the tested images, the glomerulus location module achieved an intersection over union > 0.8. The accuracy of the D-CNN in recognizing irregular granular mesangial deposition and fine granular deposition along the glomerular basement membrane was 96.1% and 93.3%, respectively. As for the diagnostic program, the accuracy, sensitivity and specificity of diagnosing suspected IgAN were 97.6%, 94.4% and 96.0%, respectively. The accuracy, sensitivity and specificity of diagnosing suspected IMN were 91.7%, 88.9% and 95.8%, respectively. The corresponding areas under the curve (AUCs) were 0.983 and 0.935. When tested with images from the outside hospital, the diagnostic program showed stable performance. The AUCs for diagnosing suspected IgAN and IMN were 0.972 and 0.948, respectively. Compared with inexperienced nephropathologists, the program showed better performance. Conclusion. The proposed diagnostic program could assist the IF diagnosis of IgAN and IMN.
AB - Background. Immunoglobulin A nephropathy (IgAN) and idiopathic membranous nephropathy (IMN) are the most common glomerular diseases. Immunofluorescence (IF) tests of renal tissues are crucial for the diagnosis. We developed a multiple convolutional neural network (CNN) -facilitated diagnostic program to assist the IF diagnosis of IgAN and IMN. Methods. The diagnostic program consisted of four parts: a CNN trained as a glomeruli detection module, an IF intensity comparator, dual-CNN (D-CNN) trained as a deposition appearance and location classifier and a post-processing module. A total of 1573 glomerular IF images from 1009 patients with glomerular diseases were used for the training and validation of the diagnostic program. A total of 1610 images of 426 patients from different hospitals were used as test datasets. The performance of the diagnostic program was compared with nephropathologists. Results. In > 90% of the tested images, the glomerulus location module achieved an intersection over union > 0.8. The accuracy of the D-CNN in recognizing irregular granular mesangial deposition and fine granular deposition along the glomerular basement membrane was 96.1% and 93.3%, respectively. As for the diagnostic program, the accuracy, sensitivity and specificity of diagnosing suspected IgAN were 97.6%, 94.4% and 96.0%, respectively. The accuracy, sensitivity and specificity of diagnosing suspected IMN were 91.7%, 88.9% and 95.8%, respectively. The corresponding areas under the curve (AUCs) were 0.983 and 0.935. When tested with images from the outside hospital, the diagnostic program showed stable performance. The AUCs for diagnosing suspected IgAN and IMN were 0.972 and 0.948, respectively. Compared with inexperienced nephropathologists, the program showed better performance. Conclusion. The proposed diagnostic program could assist the IF diagnosis of IgAN and IMN.
KW - IgA nephropathy
KW - convolutional neural network
KW - idiopathic membranous nephropathy
KW - immunofluorescence
UR - http://www.scopus.com/inward/record.url?scp=85184751960&partnerID=8YFLogxK
U2 - 10.1093/ckj/sfad153
DO - 10.1093/ckj/sfad153
M3 - Article
AN - SCOPUS:85184751960
SN - 2048-8505
VL - 16
SP - 2503
EP - 2513
JO - CKJ: Clinical Kidney Journal
JF - CKJ: Clinical Kidney Journal
IS - 12
ER -
Xia P, Lv Z, Wen Y, Zhang B, Zhao X, Zhang B 等. Development of a multiple convolutional neural network–facilitated diagnostic screening program for immunofluorescence images of IgA nephropathy and idiopathic membranous nephropathy. CKJ: Clinical Kidney Journal. 2023 12月 1;16(12):2503-2513. doi: 10.1093/ckj/sfad153
