Application of EfficientNet-B0 and GRU-based deep learning on classifying the colposcopy diagnosis of precancerous cervical lesions

被引:28
作者
Chen, Xiaoyue [1 ]
Pu, Xiaowen [1 ]
Chen, Zhirou [1 ]
Li, Lanzhen [2 ,3 ]
Zhao, Kong-Nan [4 ,5 ]
Liu, Haichun [2 ,3 ]
Zhu, Haiyan [1 ]
机构
[1] Tongji Univ, Shanghai Matern & Infant Hosp 1, Dept Gynecol, Sch Med, Shanghai, Peoples R China
[2] Shanghai Jiao Tong Univ, Dept Automat, Shanghai, Peoples R China
[3] Shanghai Jiao Tong Univ, Ningbo Artificial Intelligent Inst, Ningbo, Peoples R China
[4] Wenzhou Med Univ, Sch Basic Med Sci, Wenzhou, Peoples R China
[5] Univ Queensland, Australian Inst Bioengn & Nanotechnol, St Lucia, Qld, Australia
关键词
artificial intelligence (AI); cervical cancer; colposcopy; precancerous cervical lesions; CLASSIFICATION;
D O I
10.1002/cam4.5581
中图分类号
R73 [肿瘤学];
学科分类号
100214 ;
摘要
Background: Colposcopy is indispensable for the diagnosis of cervical lesions. However, its diagnosis accuracy for high-grade squamous intraepithelial lesion (HSIL) is at about 50%, and the accuracy is largely dependent on the skill and experience of colposcopists. The advancement in computational power made it possible for the application of artificial intelligence (AI) to clinical problems. Here, we explored the feasibility and accuracy of the application of AI on precancerous and cancerous cervical colposcopic image recognition and classification. Methods: The images were collected from 6002 colposcopy examinations of normal control, low-grade squamous intraepithelial lesion (LSIL), and HSIL. For each patient, the original, Schiller test, and acetic-acid images were all collected. We built a new neural network classification model based on the hybrid algorithm. EfficientNet-b0 was used as the backbone network for the image feature extraction, and GRU(Gate Recurrent Unit)was applied for feature fusion of the three modes examinations (original, acetic acid, and Schiller test). Results: The connected network classifier achieved an accuracy of 90.61% in distinguishing HSIL from normal and LSIL. Furthermore, the model was applied to "Trichotomy", which reached an accuracy of 91.18% in distinguishing the HSIL, LSIL and normal control at the same time. Conclusion: Our results revealed that as shown by the high accuracy of AI in the classification of colposcopic images, AI exhibited great potential to be an effective tool for the accurate diagnosis of cervical disease and for early therapeutic intervention in cervical precancer.
引用
收藏
页码:8690 / 8699
页数:10
相关论文
共 39 条
[1]   Aceto-white temporal pattern classification using k-NN to identify precancerous cervical lesion in colposcopic images [J].
Acosta-Mesa, Hector-Gabriel ;
Cruz-Ramirez, Nicandro ;
Hernandez-Jimenez, Rodolfo .
COMPUTERS IN BIOLOGY AND MEDICINE, 2009, 39 (09) :778-784
[2]   A Python']Python Clustering Analysis Protocol of Genes Expression Data Sets [J].
Agapito, Giuseppe ;
Milano, Marianna ;
Cannataro, Mario .
GENES, 2022, 13 (10)
[3]   An Efficient CNN-Based Deep Learning Model to Detect Malware Attacks (CNN-DMA) in 5G-IoT Healthcare Applications [J].
Anand, Ankita ;
Rani, Shalli ;
Anand, Divya ;
Aljahdali, Hani Moaiteq ;
Kerr, Dermot .
SENSORS, 2021, 21 (19)
[4]   Recurrence rate after loop electrosurgical excision procedure (LEEP) and laser Conization: A 5-year follow-up study [J].
Bogani, Giorgio ;
Di Donato, Violante ;
Sopracordevole, Francesco ;
Ciavattini, Andrea ;
Ghelardi, Alessandro ;
Lopez, Salvatore ;
Simoncini, Tommaso ;
Plotti, Francesco ;
Casarin, Jvan ;
Serati, Maurizio ;
Pinelli, Ciro ;
Valenti, Gaetano ;
Bergamini, Alice ;
Gardella, Barbara ;
Dell'Acqua, Andrea ;
Monti, Ermelinda ;
Vercellini, Paolo ;
Fischetti, Margherita ;
D'Ippolito, Giovanni ;
Aguzzoli, Lorenzo ;
Mandato, Vincenzo D. ;
Carunchio, Paola ;
Carlinfante, Gabriele ;
Giannella, Luca ;
Scaffa, Cono ;
Falcone, Francesca ;
Borghi, Chiara ;
Ditto, Antonino ;
Malzoni, Mario ;
Giannini, Andrea ;
Salerno, Maria Giovanna ;
Liberale, Viola ;
Contino, Biagio ;
Donfrancesco, Cristina ;
Desiato, Michele ;
Perrone, Anna Myriam ;
Dondi, Giulia ;
De Iaco, Pierandrea ;
Maggiore, Umberto Leone Roberti ;
Signorelli, Mauro ;
Chiappa, Valentina ;
Ferrero, Simone ;
Sarpietro, Giuseppe ;
Matarazzo, Maria G. ;
Cianci, Antonio ;
Bosio, Sara ;
Ruisi, Simona ;
Guerrisi, Rocco ;
Brusadelli, Claudia ;
Mosca, Lavinia .
GYNECOLOGIC ONCOLOGY, 2020, 159 (03) :636-641
[5]   Colposcopy: a closer look into its past, present and future [J].
Bokil, Mugdha ;
Lim, Boon .
BJOG-AN INTERNATIONAL JOURNAL OF OBSTETRICS AND GYNAECOLOGY, 2019, 126 (04) :543-543
[6]   Global elimination of cervical cancer as a public health problem [J].
Brisson, Marc ;
Drolet, Melanie .
LANCET ONCOLOGY, 2019, 20 (03) :319-321
[7]   The diagnostic accuracy of colposcopy - A review of research methodology and impact on the outcomes of quality assurance [J].
Brown, Brian Hilton ;
Tidy, John A. .
EUROPEAN JOURNAL OF OBSTETRICS & GYNECOLOGY AND REPRODUCTIVE BIOLOGY, 2019, 240 :182-186
[8]   Accuracy of combinations of visual inspection using acetic acid or lugol iodine to detect cervical precancer: a meta-analysis [J].
Catarino, R. ;
Schafer, S. ;
Vassilakos, P. ;
Petignat, P. ;
Arbyn, M. .
BJOG-AN INTERNATIONAL JOURNAL OF OBSTETRICS AND GYNAECOLOGY, 2018, 125 (05) :545-553
[9]   Classification of cervical neoplasms on colposcopic photography using deep learning [J].
Cho, Bum-Joo ;
Choi, Youn Jin ;
Lee, Myung-Je ;
Kim, Ju Han ;
Son, Ga-Hyun ;
Park, Sung-Ho ;
Kim, Hong-Bae ;
Joo, Yeon-Ji ;
Cho, Hye-Yon ;
Kyung, Min Sun ;
Park, Young-Han ;
Kang, Byung Soo ;
Hur, Soo Young ;
Lee, Sanha ;
Park, Sung Taek .
SCIENTIFIC REPORTS, 2020, 10 (01)
[10]   Long-term observational approach in women with histological diagnosis of cervical low-grade squamous intraepithelial lesion: an Italian multicentric retrospective cohort study [J].
Ciavattini, Andrea ;
Serri, Matteo ;
Di Giuseppe, Jacopo ;
Liverani, Carlo Antonio ;
Gardella, Barbara ;
Papiccio, Maria ;
Delli Carpini, Giovanni ;
Morini, Stefano ;
Clemente, Nicolo ;
Sopracordevole, Francesco .
BMJ OPEN, 2019, 9 (07)