Convolutional neural network assistance significantly improves dermatologists' diagnosis of cutaneous tumours using clinical images

被引：15

作者：

Ba, Wei ^{[1
]}

Wu, Huan ^{[2
]}

Chen, Wei W. ^{[3
]}

Wang, Shu H. ^{[4
]}

Zhang, Zi Y. ^{[5
]}

Wei, Xuan J. ^{[1
]}

Wang, Wen J. ^{[1
]}

Yang, Lei ^{[6
]}

Zhou, Dong M. ^{[3
]}

Zhuang, Yi X. ^{[7
]}

Zhong, Qin ^{[7
]}

Song, Zhi G. ^{[8
]}

Li, Cheng X. ^{[1
]}

机构：

[1] Chinese Peoples Liberat Army Gen Hosp, Dept Dermatol, Beijing 100853, Peoples R China

[2] Chinese Peoples Liberat Army Gen Hosp, Res Med Big Data Ctr, Beijing 100853, Peoples R China

[3] Affiliated Capital Med Univ, Beijing Hosp Tradit Chinese Med, Dept Dermatol, Beijing 100010, Peoples R China

[4] Tsinghua Univ, Inst Interdisciplinary Informat Sci, Beijing 100084, Peoples R China

[5] North China Univ Sci & Technol, Affiliated Hosp, Dept Dermatol, Tangshan 063000, Peoples R China

[6] Rocket Force Characterist Med Ctr, Dept Dermatol, Beijing 100088, Peoples R China

[7] Med Sch Chinese PLA Gen Hosp, Beijing 100853, Peoples R China

[8] Chinese Peoples Liberat Army Gen Hosp, Dept Pathol, Beijing 100853, Peoples R China

来源：

EUROPEAN JOURNAL OF CANCER | 2022年 / 169卷

关键词：

Convolutional neural network; Clinical image; Cutaneous tumours; Multi-reader multi-case (MRMC) study; Fully crossed; BASAL-CELL CARCINOMA; SKIN-CANCER; ARTIFICIAL-INTELLIGENCE; CLASSIFICATION; MEDICINE; MELANOMA; GO;

D O I：

10.1016/j.ejca.2022.04.015

中图分类号：

R73 [肿瘤学];

学科分类号：

100214 ;

摘要：

Background: Convolutional neural networks (CNNs) have demonstrated expertlevel performance in cutaneous tumour classification using clinical images, but most previous studies have focused on dermatologist-versus-CNN comparisons rather than their combination. The objective of our study was to evaluate the potential impact of CNN assistance on dermatologists for clinical image interpretation.Methods: A multi-class CNN was trained and validated using a dataset of 25,773 clinical images comprising 10 categories of cutaneous tumours. The CNN's performance was tested on an independent dataset of 2107 images. A total of 400 images (40 per category) were randomly selected from the test dataset. A fully crossed, self-control, multi-reader multi-case (MRMC) study was conducted to compare the performance of 18 board-certified dermatologists

引用

页码：156 / 165

页数：10

共 40 条

[1]

[Anonymous], CLIN PERFORMANCE ASS

[2] An unsupervised feature learning framework for basal cell carcinoma image analysis [J].

Arevalo, John ;

Cruz-Roa, Angel ;

Arias, Viviana ;

Romero, Eduardo ;

Gonzalez, Fabio A. .

ARTIFICIAL INTELLIGENCE IN MEDICINE, 2015, 64 (02) :131-145

[3]

Bray F, 2018, CA-CANCER J CLIN, V68, P394, DOI [10.3322/caac.21492, 10.3322/caac.21609]

[4] A convolutional neural network trained with dermoscopic images performed on par with 145 dermatologists in a clinical melanoma image classification task [J].

Brinker, Titus J. ;

Hekler, Achim ;

Enk, Alexander H. ;

Klode, Joachim ;

Hauschild, Axel ;

Berking, Carola ;

Schilling, Bastian ;

Haferkamp, Sebastian ;

Schadendorf, Dirk ;

Froehling, Stefan ;

Utikal, Jochen S. ;

von Kalle, Christof ;

Ludwig-Peitsch, Wiebke ;

Sirokay, Judith ;

Heinzerling, Lucie ;

Albrecht, Magarete ;

Baratella, Katharina ;

Bischof, Lena ;

Chorti, Eleftheria ;

Dith, Anna ;

Drusio, Christina ;

Giese, Nina ;

Gratsias, Emmanouil ;

Griewank, Klaus ;

Hallasch, Sandra ;

Hanhart, Zdenka ;

Herz, Saskia ;

Hohaus, Katja ;

Jansen, Philipp ;

Jockenhoefer, Finja ;

Kanaki, Theodora ;

Knispel, Sarah ;

Leonhard, Katja ;

Martaki, Anna ;

Matei, Liliana ;

Matull, Johanna ;

Olischewski, Alexandra ;

Petri, Maximilian ;

Placke, Jan-Malte ;

Raub, Simon ;

Salva, Katrin ;

Schlott, Swantje ;

Sody, Elsa ;

Steingrube, Nadine ;

Stoffels, Ingo ;

Ugurel, Selma ;

Sondermann, Wiebke ;

Zaremba, Anne ;

Gebhardt, Christoffer ;

Booken, Nina .

EUROPEAN JOURNAL OF CANCER, 2019, 111 :148-154

[5]

Cesana BM, 2008, MINERVA ANESTESIOL, V74, P431

[6] A guide to deep learning in healthcare [J].

Esteva, Andre ;

Robicquet, Alexandre ;

Ramsundar, Bharath ;

Kuleshov, Volodymyr ;

DePristo, Mark ;

Chou, Katherine ;

Cui, Claire ;

Corrado, Greg ;

Thrun, Sebastian ;

Dean, Jeff .

NATURE MEDICINE, 2019, 25 (01) :24-29

[7] Dermatologist-level classification of skin cancer with deep neural networks [J].

Esteva, Andre ;

Kuprel, Brett ;

Novoa, Roberto A. ;

Ko, Justin ;

Swetter, Susan M. ;

Blau, Helen M. ;

Thrun, Sebastian .

NATURE, 2017, 542 (7639) :115-+

[8]

Finlayson GD, 2004, 12TH COLOR IMAGING CONFERENCE: COLOR SCIENCE AND ENGINEERING SYSTEMS, TECHNOLOGIES, APPLICATIONS, P37

[9] Deep-learning-based, computer-aided classifier developed with a small dataset of clinical images surpasses board-certified dermatologists in skin tumour diagnosis [J].

Fujisawa, Y. ;

Otomo, Y. ;

Ogata, Y. ;

Nakamura, Y. ;

Fujita, R. ;

Ishitsuka, Y. ;

Watanabe, R. ;

Okiyama, N. ;

Ohara, K. ;

Fujimoto, M. .

BRITISH JOURNAL OF DERMATOLOGY, 2019, 180 (02) :373-381

[10] Man against machine: diagnostic performance of a deep learning convolutional neural network for dermoscopic melanoma recognition in comparison to 58 dermatologists [J].

Haenssle, H. A. ;

Fink, C. ;

Schneiderbauer, R. ;

Toberer, F. ;

Buhl, T. ;

Blum, A. ;

Kalloo, A. ;

Hassens, A. Ben Hadj ;

Thomas, L. ;

Enk, A. ;

Uhlmann, L. .

ANNALS OF ONCOLOGY, 2018, 29 (08) :1836-1842

← 1 2 3 4 →