Use of Artificial Intelligence in the Prediction of Malignant Potential of Gastric Gastrointestinal Stromal Tumors

Abstract

Background and Aims

This study aimed to investigate whether AI via a deep learning algorithm using endoscopic ultrasonography (EUS) images could predict the malignant potential of gastric gastrointestinal stromal tumors (GISTs).

Methods

A series of patients who underwent EUS before surgical resection for gastric GISTs were included. A total of 685 images of GISTs from 55 retrospectively included patients were used as the training data set for the AI system. Convolutional neural networks were constructed to build a deep learning model. After applying the synthetic minority oversampling technique, 70% of the generated images were used for AI training and 30% were used to test AI diagnoses. Next, validation was performed using 153 EUS images of 15 patients with GISTs. In addition, conventional EUS features of 55 patients in the training cohort were evaluated to predict the malignant potential of GISTs and mitotic index.

Results

The overall sensitivity, specificity, and accuracy of the AI system for predicting malignancy risk were 83%, 94%, and 82% in the training dataset, and 75%, 73%, and 66% in the validation cohort, respectively. When patients were divided into low-risk and high-risk groups, sensitivity, specificity, and accuracy increased to 99% in the training dataset and 99.7%, 99.7%, and 99.6%, respectively, in the validation cohort. No conventional EUS features were found to be associated with either malignant potential or mitotic index (P > 0.05).

Conclusions

AI via a deep learning algorithm using EUS images could predict the malignant potential of gastric GISTs with high accuracy.

Graphic Abstract

This is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2
Fig. 3

Abbreviations

AFIP:

Armed Forces Institute of Pathology

AI:

Artificial intelligence

ANN:

Artificial neural network

CNN:

Convolutional neural network

EUS:

Endoscopic ultrasonography

FNA:

Fine needle aspiration

GIST:

Gastrointestinal stromal tumor

HPF:

High-power field

NPV:

Negative predictive value

PPV:

Positive predictive value

ROC:

Receiver Operating Characteristic

SD:

Standard deviation

SMOTE:

Synthetic minority oversampling technique

References

  1. 1.

    Miettinen M, Sobin LH, Lasota J. Gastrointestinal stromal tumors of the stomach: a clinicopathologic, immunohistochemical, and molecular genetic study of 1765 cases with long-term follow-up. Am J Surg Pathol 2005;29:52–68. https://doi.org/10.1097/01.pas.0000146010.92933.de.

    Article  PubMed  Google Scholar 

  2. 2.

    Søreide K, Sandvik OM, Søreide JA, Giljaca V, Jureckova A, Bulusu VR. Global epidemiology of gastrointestinal stromal tumours (GIST): a systematic review of population-based cohort studies. Cancer Epidemiol 2016;40:39–46. https://doi.org/10.1016/j.canep.2015.10.031.

    Article  PubMed  Google Scholar 

  3. 3.

    Baysal B, Masri OA, Eloubeidi MA, Senturk H. The role of EUS and EUS-guided FNA in the management of subepithelial lesions of the esophagus: a large, single-center experience. Endosc Ultrasound 2017;6:308–316. https://doi.org/10.4103/2303-9027.155772.

    Article  PubMed  PubMed Central  Google Scholar 

  4. 4.

    Chandrasekhara V, Ginsberg GG. Endoscopic management of gastrointestinal stromal tumors. Curr Gastroenterol Rep 2011;13:532–539. https://doi.org/10.1007/s11894-011-0224-6.

    Article  PubMed  Google Scholar 

  5. 5.

    Fletcher CD, Berman JJ, Corless C, Gorstein F, Lasota J, Longley BJ et al. Diagnosis of gastrointestinal stromal tumors: a consensus approach. Hum Pathol 2002;33:459–465. https://doi.org/10.1053/hupa.2002.123545.

    Article  PubMed  Google Scholar 

  6. 6.

    Miettinen M, Lasota J. Gastrointestinal stromal tumors: pathology and prognosis at different sites. Semin Diagn Pathol 2006;23:70–83. https://doi.org/10.1053/j.semdp.2006.09.001.

    Article  PubMed  Google Scholar 

  7. 7.

    Joensuu H. Risk stratification of patients diagnosed with gastrointestinal stromal tumor. Hum Pathol 2008;39:1411–1419. https://doi.org/10.1016/j.humpath.2008.06.025.

    Article  PubMed  Google Scholar 

  8. 8.

    Miettinen M, Makhlouf H, Sobin LH, Lasota J. Gastrointestinal stromal tumors of the jejunum and ileum: a clinicopathologic, immunohistochemical, and molecular genetic study of 906 cases before imatinib with long-term follow-up. Am J Surg Pathol 2006;30:477–489. https://doi.org/10.1097/00000478-200604000-00008.

    Article  PubMed  Google Scholar 

  9. 9.

    Larghi A, Fuccio L, Chiarello G, Attili F, Vanella G, Paliani GB et al. Fine-needle tissue acquisition from subepithelial lesions using a forward-viewing linear echoendoscope. Endoscopy 2014;46:39–45. https://doi.org/10.1055/s-0033-1344895.

    Article  PubMed  Google Scholar 

  10. 10.

    Polkowski M, Gerke W, Jarosz D, Nasierowska-Guttmejer A, Rutkowski P, Nowecki ZI et al. Diagnostic yield and safety of endoscopic ultrasound-guided trucut [corrected] biopsy in patients with gastric submucosal tumors: a prospective study. Endoscopy 2009;41:329–334. https://doi.org/10.1055/s-0029-1214447.

    CAS  Article  PubMed  Google Scholar 

  11. 11.

    Ricci R, Chiarello G, Attili F, Fuccio L, Alfieri S, Persiani R et al. Endoscopic ultrasound-guided fine needle tissue acquisition biopsy samples do not allow a reliable proliferation assessment of gastrointestinal stromal tumours. Dig Liver Dis 2015;47:291–295. https://doi.org/10.1016/j.dld.2014.12.011.

    Article  PubMed  Google Scholar 

  12. 12.

    Ando N, Goto H, Niwa Y, Hirooka Y, Ohmiya N, Nagasaka T et al. The diagnosis of GI stromal tumors with EUS-guided fine needle aspiration with immunohistochemical analysis. Gastrointest Endosc 2002;55:37–43. https://doi.org/10.1067/mge.2002.120323.

    Article  PubMed  Google Scholar 

  13. 13.

    Okubo K, Yamao K, Nakamura T, Tajika M, Sawaki A, Hara K et al. Endoscopic ultrasound-guided fine-needle aspiration biopsy for the diagnosis of gastrointestinal stromal tumors in the stomach. J Gastroenterol 2004;39:747–753. https://doi.org/10.1007/s00535-004-1383-0.

    Article  PubMed  Google Scholar 

  14. 14.

    Seven G, Kochan K, Caglar E, Kiremitci S, Koker IH, Senturk H. Evaluation of Ki67 index in endoscopic ultrasound-guided fine needle aspiration samples for the assessment of malignancy risk in gastric gastrointestinal stromal tumors. Dig Dis. 2020. https://doi.org/10.1159/000511994.

    Article  PubMed  Google Scholar 

  15. 15.

    Cağlar E, Hatemi I, Atasoy D, Sişman G, Sentürk H. Concordance of endoscopic ultrasonography-guided fine needle aspiration diagnosis with the final diagnosis in subepithelial lesions. Clin Endosc 2013;46:379–383. https://doi.org/10.5946/ce.2013.46.4.379.

    Article  PubMed  PubMed Central  Google Scholar 

  16. 16.

    Wani S, Muthusamy VR, Komanduri S. EUS-guided tissue acquisition: an evidence-based approach (with videos). Gastrointest Endosc 2014;80:939–59.e7. https://doi.org/10.1016/j.gie.2014.07.066.

    Article  PubMed  Google Scholar 

  17. 17.

    Jeon SW, Park YD, Chung YJ, Cho CM, Tak WY, Kweon YO et al. Gastrointestinal stromal tumors of the stomach: endosonographic differentiation in relation to histological risk. J Gastroenterol Hepatol 2007;22:2069–2075. https://doi.org/10.1111/j.1440-1746.2006.04767.x.

    Article  PubMed  Google Scholar 

  18. 18.

    Shah P, Gao F, Edmundowicz SA, Azar RR, Early DS. Predicting malignant potential of gastrointestinal stromal tumors using endoscopic ultrasound. Dig Dis Sci 2009;54:1265–1269. https://doi.org/10.1007/s10620-008-0484-7.

    Article  PubMed  Google Scholar 

  19. 19.

    Chen TH, Hsu CM, Chu YY, Wu CH, Chen TC, Hsu JT et al. Association of endoscopic ultrasonographic parameters and gastrointestinal stromal tumors (GISTs): can endoscopic ultrasonography be used to screen gastric GISTs for potential malignancy? Scand J Gastroenterol 2016;51:374–377. https://doi.org/10.3109/00365521.2015.1095350.

    CAS  Article  PubMed  Google Scholar 

  20. 20.

    Chen T, Xu L, Dong X, Li Y, Yu J, Xiong W et al. The roles of CT and EUS in the preoperative evaluation of gastric gastrointestinal stromal tumors larger than 2 cm. Eur Radiol 2019;29:2481–2489. https://doi.org/10.1007/s00330-018-5945-6.

    Article  PubMed  Google Scholar 

  21. 21.

    Brand B, Oesterhelweg L, Binmoeller KF, Sriram PV, Bohnacker S, Seewald S et al. Impact of endoscopic ultrasound for evaluation of submucosal lesions in gastrointestinal tract. Dig Liver Dis 2002;34:290–297. https://doi.org/10.1016/s1590-8658(02)80150-5.

    CAS  Article  PubMed  Google Scholar 

  22. 22.

    Chak A, Canto MI, Rösch T, Dittler HJ, Hawes RH, Tio TL et al. Endosonographic differentiation of benign and malignant stromal cell tumors. Gastrointest Endosc 1997;45:468–473. https://doi.org/10.1016/s0016-5107(97)70175-5.

    CAS  Article  PubMed  Google Scholar 

  23. 23.

    Le Berre C, Sandborn WJ, Aridhi S, Devignes MD, Fournier L, Smaïl-Tabbone M et al. Application of artificial intelligence to gastroenterology and hepatology. Gastroenterology 2020;158:76-94.e2. https://doi.org/10.1053/j.gastro.2019.08.058.

    Article  PubMed  Google Scholar 

  24. 24.

    Kuwahara T, Hara K, Mizuno N, Okuno N, Matsumoto S, Obata M 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. https://doi.org/10.14309/ctg.0000000000000045.

    CAS  Article  PubMed  Google Scholar 

  25. 25.

    Kurita Y, Kuwahara T, Hara K, Mizuno N, Okuno N, Matsumoto S et al. Diagnostic ability of artificial intelligence using deep learning analysis of cyst fluid in differentiating malignant from benign pancreatic cystic lesions. Sci Rep 2019;9:6893. https://doi.org/10.1038/s41598-019-43314-3.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  26. 26.

    Maeda Y, Kudo SE, Mori Y, Misawa M, Ogata N, Sasanuma S et al. Fully automated diagnostic system with artificial intelligence using endocytoscopy to identify the presence of histologic inflammation associated with ulcerative colitis (with video). Gastrointest Endosc 2019;89:408–415. https://doi.org/10.1016/j.gie.2018.09.024.

    Article  PubMed  Google Scholar 

  27. 27.

    Horie Y, Yoshio T, Aoyama K, Yoshimizu S, Horiuchi Y, Ishiyama A et al. Diagnostic outcomes of esophageal cancer by artificial intelligence using convolutional neural networks. Gastrointest Endosc 2019;89:25–32. https://doi.org/10.1016/j.gie.2018.07.037.

    Article  PubMed  Google Scholar 

  28. 28.

    Zhu Y, Wang QC, Xu MD, Zhang Z, Cheng J, Zhong YS et al. Application of convolutional neural network in the diagnosis of the invasion depth of gastric cancer based on conventional endoscopy. Gastrointest Endosc 2019;89:806–15.e1. https://doi.org/10.1016/j.gie.2018.11.011.

    Article  PubMed  Google Scholar 

  29. 29.

    Goodfellow I. Deep learning. Cambridge: MIT Press; 2016.

    Google Scholar 

  30. 30.

    von Mehren M, Randall RL, Benjamin RS, Boles S, Bui MM, Ganjoo KN et al. Soft Tissue Sarcoma, Version 2.2018, NCCN Clinical Practice Guidelines in Oncology. J Natl Compr Cancer Netw 2018;16:536–563. https://doi.org/10.6004/jnccn.2018.0025.

    Article  Google Scholar 

  31. 31.

    Joo MK, Park JJ, Kim H, Koh JS, Lee BJ, Chun HJ et al. Endoscopic versus surgical resection of GI stromal tumors in the upper GI tract. Gastrointest Endosc 2016;83:318–326. https://doi.org/10.1016/j.gie.2015.07.034.

    Article  PubMed  Google Scholar 

  32. 32.

    Kim GH, Choi KD, Gong CS, Lee IS, Park YS, Han M et al. Comparison of the treatment outcomes of endoscopic and surgical resection of GI stromal tumors in the stomach: a propensity score-matched case–control study. Gastrointest Endosc 2020;91:527–536. https://doi.org/10.1016/j.gie.2019.10.020.

    Article  PubMed  Google Scholar 

Download references

Author information

Affiliations

Authors

Corresponding author

Correspondence to Hakan Senturk.

Ethics declarations

Conflict of interest

The authors have no conflicts of interest to disclose.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Seven, G., Silahtaroglu, G., Kochan, K. et al. Use of Artificial Intelligence in the Prediction of Malignant Potential of Gastric Gastrointestinal Stromal Tumors. Dig Dis Sci (2021). https://doi.org/10.1007/s10620-021-06830-9

Download citation

Keywords

  • Artificial intelligence
  • Gastric gastrointestinal stromal tumors
  • Mitotic index
  • Risk classification