Value of artificial intelligence with novel tumor tracking technology in the diagnosis of gastric submucosal tumors by contrast-enhanced harmonic endoscopic ultrasonography.
artificial intelligences
contrast-enhanced harmonic
endoscopic ultrasonography
gastrointestinal stromal tumor
neural network
submucosal tumor
Journal
Journal of gastroenterology and hepatology
ISSN: 1440-1746
Titre abrégé: J Gastroenterol Hepatol
Pays: Australia
ID NLM: 8607909
Informations de publication
Date de publication:
May 2022
May 2022
Historique:
revised:
01
12
2021
received:
16
09
2021
accepted:
12
01
2022
pubmed:
20
1
2022
medline:
12
5
2022
entrez:
19
1
2022
Statut:
ppublish
Résumé
Contrast-enhanced harmonic endoscopic ultrasonography (CH-EUS) is useful for the diagnosis of lesions inside and outside the digestive tract. This study evaluated the value of artificial intelligence (AI) in the diagnosis of gastric submucosal tumors by CH-EUS. This retrospective study included 53 patients with gastrointestinal stromal tumors (GISTs) and leiomyomas, all of whom underwent CH-EUS between June 2015 and February 2020. A novel technology, SiamMask, was used to track and trim the lesions in CH-EUS videos. CH-EUS was evaluated by AI using deep learning involving a residual neural network and leave-one-out cross-validation. The diagnostic accuracy of AI in discriminating between GISTs and leiomyomas was assessed and compared with that of blind reading by two expert endosonographers. Of the 53 patients, 42 had GISTs and 11 had leiomyomas. Mean tumor size was 26.4 mm. The consistency rate of the segment range of the tumor image extracted by SiamMask and marked by the endosonographer was 96% with a Dice coefficient. The sensitivity, specificity, and accuracy of AI in diagnosing GIST were 90.5%, 90.9%, and 90.6%, respectively, whereas those of blind reading were 90.5%, 81.8%, and 88.7%, respectively (P = 0.683). The κ coefficient between the two reviewers was 0.713. The diagnostic ability of CH-EUS results evaluated by AI to distinguish between GISTs and leiomyomas was comparable with that of blind reading by expert endosonographers.
Sections du résumé
BACKGROUND AND AIM
OBJECTIVE
Contrast-enhanced harmonic endoscopic ultrasonography (CH-EUS) is useful for the diagnosis of lesions inside and outside the digestive tract. This study evaluated the value of artificial intelligence (AI) in the diagnosis of gastric submucosal tumors by CH-EUS.
METHODS
METHODS
This retrospective study included 53 patients with gastrointestinal stromal tumors (GISTs) and leiomyomas, all of whom underwent CH-EUS between June 2015 and February 2020. A novel technology, SiamMask, was used to track and trim the lesions in CH-EUS videos. CH-EUS was evaluated by AI using deep learning involving a residual neural network and leave-one-out cross-validation. The diagnostic accuracy of AI in discriminating between GISTs and leiomyomas was assessed and compared with that of blind reading by two expert endosonographers.
RESULTS
RESULTS
Of the 53 patients, 42 had GISTs and 11 had leiomyomas. Mean tumor size was 26.4 mm. The consistency rate of the segment range of the tumor image extracted by SiamMask and marked by the endosonographer was 96% with a Dice coefficient. The sensitivity, specificity, and accuracy of AI in diagnosing GIST were 90.5%, 90.9%, and 90.6%, respectively, whereas those of blind reading were 90.5%, 81.8%, and 88.7%, respectively (P = 0.683). The κ coefficient between the two reviewers was 0.713.
CONCLUSIONS
CONCLUSIONS
The diagnostic ability of CH-EUS results evaluated by AI to distinguish between GISTs and leiomyomas was comparable with that of blind reading by expert endosonographers.
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
841-846Informations de copyright
© 2022 Journal of Gastroenterology and Hepatology Foundation and John Wiley & Sons Australia, Ltd.
Références
Akahoshi K, Oya M, Koga T et al. Current clinical management of gastrointestinal stromal tumor. World J. Gastroenterol. 2018; 24: 2806-2817.
Nishida T, Hirota S. Biological and clinical review of stromal tumors in the gastrointestinal tract. Histol. Histopathol. 2000; 15: 1293-1301.
Kamata K, Takenaka M, Kitano M et al. Contrast-enhanced harmonic endoscopic ultrasonography for differential diagnosis of submucosal tumors of the upper gastrointestinal tract. J. Gastroenterol. Hepatol. 2017; 32: 1686-1692.
Sakamoto H, Kitano M, Matsui S et al. Estimation of malignant potential of GI stromal tumors by contrast-enhanced harmonic EUS (with videos). Gastrointest. Endosc. 2011; 73: 227-237.
Akahoshi K, Oya M, Koga T et al. Clinical usefulness of endoscopic ultrasound-guided fine needle aspiration for gastric subepithelial lesions smaller than 2 cm. J. Gastrointestin. Liver Dis. 2014; 23: 405-412.
Sepe PS, Moparty B, Pitman MB et al. EUS-guided FNA for the diagnosis of GI stromal cell tumors: sensitivity and cytologic yield. Gastrointest. Endosc. 2009; 70: 254-261.
Yang YT, Shen N, Ao F et al. Diagnostic value of contrast-enhanced harmonic endoscopic ultrasonography in predicting the malignancy potential of submucosal tumors: a systematic review and meta-analysis. Surg. Endosc. 2020; 34: 3754-3765.
Kamata K, Kurita A, Yasukawa S et al. Utility of a 20G needle with a core trap in EUS-guided fine-needle biopsy for gastric submucosal tumors: a multicentric prospective trial. Endosc. Ultrasound 2021; 10: 134-140.
Kuwahara T, Hara K, Mizuno N et al. Current status of artificial intelligence analysis for endoscopic ultrasonography. Dig. Endosc. 2021; 33: 298-305.
Kuwahara T, Hara K, Mizuno N et al. Usefulness of deep learning analysis for the diagnosis of malignancy in intraductal papillary mucinous neoplasms of the pancreas. Clin. Transl. Gastroenterol. 2019; 10: 1-8.
Minoda Y, Ihara E, Komori K et al. Efficacy of endoscopic ultrasound with artificial intelligence for the diagnosis of gastrointestinal stromal tumors. J. Gastroenterol. 2020; 55: 1119-1126.
Xia J, Xia T, Pan J et al. Use of artificial intelligence for detection of gastric lesions by magnetically controlled capsule endoscopy. Gastrointest. Endosc. 2021; 93: 133-139.
He YS, Su JR, Li Z et al. Application of artificial intelligence in gastrointestinal endoscopy. J. Dig. Dis. 2019; 20: 623-630.
Q.L.Z. Wang, L. Bertinetto et al. Fast online object tracking and segmentation: a unifying approach. IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2019.
Santoro A, Bartunov S, Botvinick M et al. Meta-learning with memory-augmented neural networks. International conference on machine learning. PMLR 2016: 1842-1850.
Vanschoren J. Meta-Learning. Automated Machine Learning. Cham: Springer, 2019; 35-61.
Timothy Hospedales AA, Paul Micaelli, Amos Storkey. Meta-learning in neural networks: a survey. ArXiv 2020.
Hospedales T, Antoniou A, Micaelli P et al. Meta-learning in neural networks: a survey. ArXiv 2020.
Vinyals O, Blundell C, Lillicrap T, Wierstra D. Matching networks for one shot learning. Proceedings of the 30th International Conference on Neural Information Processing Systems 2016; 3637-3645.
Liang Z, Liang C, Zhang Y, Mu H, Li G. Tracking of moving target based on SiamMask for video SAR system. IEEE International Conference on Signal, Information and Data Processing (ICSIDP) 2019; 1-4.
Iwasa Y, Iwashita T, Takeuchi Y et al. Automatic segmentation of pancreatic tumors using deep learning on a video image of contrast-enhanced endoscopic ultrasound. J. Clin. Med. 2021; 10: 3589.
Jha D, Smedsrud PH, Riegler MA et al. Resunet++: an advanced architecture for medical image segmentation. IEEE Int. Symp. Multimed. (ISM) 2019: 225-2255.
Hirasawa T, Aoyama K, Tanimoto T et al. Application of artificial intelligence using a convolutional neural network for detecting gastric cancer in endoscopic images. International Gastric Cancer Association and the Japanese Gastric Cancer Association 2018; 21: 653-660.
Komeda Y, Handa H, Watanabe T et al. Computer-aided diagnosis based on convolutional neural network system for colorectal polyp classification: preliminary experience. Oncology 2017; 93: 30-34.
Krizhevsky ASI, Hinton G. ImageNet classification with deep convolutional neural networks. Adv. Neural. Inf. Process Syst. 2012; 25: 1097-1105.
Christian Szegedy, Wei Liu, Yangqing Ji et al. Going deeper with convolutions. ArXiv 2014; 1409: 4842.
Kaiming He, Xiangyu Zhang, Shaoqing Ren et al. Deep residual learning for image recognition.preprint ArXiv 2015; 1512: 3385.
Redmon JDS, Girshick R, Farhadi A. You only look once: unified, real-time object detection. IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2016: 779-788.
Tang JY, Tao KG, Zhang LY et al. Value of contrast-enhanced harmonic endoscopic ultrasonography in differentiating between gastrointestinal stromal tumors: a meta-analysis. J. Dig. Dis. 2019; 20: 127-134.
Kitano M, Yamashita Y, Kamata K et al. The Asian Federation of Societies for Ultrasound in Medicine and Biology (AFSUMB) guidelines for contrast-enhanced endoscopic ultrasound. Ultrasound Med. Biol. 2021; 47: 1433-1447.