Exploration and application of attention mechanism in survival analysis of competitive events in oral cancer

doi:10.19439/j.sjos.2025.04.004

Abstract

Abstract: PURPOSE: This study constructed a model of OSAA (oral survival analysis with attention) for survival analysis of competitive events in oral cancer based on attention mechanism, and explored its application value in oral auxiliary diagnosis and treatment of oral cancer. METHODS: Eligible data of oral cancer patient from Surveillance, Epidemiology, and End Results Program(SEER) database were selected as research subjects. Cox proportional hazards models, deep learning-based survival analysis models (such as DeepSurv, DeepHit), and OSAA models were established and trained for prediction. The predictive performance of each model was tested through concordance index (C-index) and integrated Brier score (IBS) test. The model's discriminative ability was evaluated using the Kaplan-Meier survival curve and the time-dependent receiver operating characteristic (ROC) curve. RESULTS: OSAA demonstrated ahigher C-index and a lower IBS on different datasets, with more distinct survival and ROC curves compared to other models. CONCLUSIONS: The OSAA model exhibits superior predictive performance compared to other models, with better robustness and generalization ability under different datasets and tasks. It has a certain value for establishing auxiliary diagnosis and treatment models for oral diseases represented by oral cancer.

Key words: Attention mechanism, Oral cancer, Competing events, Survival analysis, Auxiliary diagnosis and treatment

CLC Number:

R-05
R739.8

Jin Lu, Zhang Rui, Si Misi, Gao Shang, Chen Qianming. Exploration and application of attention mechanism in survival analysis of competitive events in oral cancer[J]. Shanghai Journal of Stomatology, 2025, 34(4): 362-368.

References

[1] Mehrtash H, Duncan K, Parascandola M, et al.Defining a global research and policy agenda for betel quid and areca nut[J]. Lancet Oncol, 2017, 18(12): e767-e775.
[2] Zhang S, Zheng R, Chen Q, et al.Oral cancer incidence and mortality in China, 2011[J].Chin J Cancer Res, 2015, 27(1): 44-51.
[3] Chen W, Zheng R, Baade PD, et al.Cancer statistics in China, 2015[J]. CA Cancer J Clin, 2016, 66(2): 115-132.
[4] Lei L, Zheng R, Peng K, et al.Incidence and mortality of oral and oropharyngeal cancer in China, 2015[J]. Chin J Cancer Res, 2020, 32(1): 1-9.
[5] Kumar M, Nanavati R, Modi TG, et al.Oral cancer: etiology and risk factors: a review[J]. J Cancer Res Ther, 2016, 12(2): 458-463.
[6] Zeng H, Chen W, Zheng R, et al.Changing cancer survival in China during 2003-15: a pooled analysis of 17 population-based cancer registries[J]. Lancet Glob Health, 2018, 6(5): e555-e567.
[7] Ronneberger O, Fischer P, Brox T.U-net: convolutional networks for biomedical image segmentation[C]//Medical Image Computing and Computer-Assisted Intervention-MICCAI 2015: 18th International Conference. Munich: Springer International Publishing, 2015: 234-241.
[8] Malamateniou C, McFadden S, McQuinlan Y, et al. Artificial intelligence: guidance for clinical imaging and therapeutic radiography professionals, a summary by the Society of Radiographers AI working group[J]. Radiography(Lond), 2021, 27(4): 1192-1202.
[9] Bodner J, Wykypiel H, Wetscher G, et al.First experiences with the da Vinci^TM operating robot in thoracic surgery[J]. Eur J Cardiothorac Surg, 2004, 25(5): 844-851.
[10] Dissanayaka WL, Pitiyage G, Kumarasiri PVR, et al.Clinical and histopathologic parameters in survival of oral squamous cell carcinoma[J]. Oral Surg Oral Med Oral Pathol Oral Radiol, 2012, 113(4): 518-525.
[11] Katzman JL, Shaham U, Cloninger A, et al.DeepSurv: personalized treatment recommender system using a Cox proportional hazards deep neural network[J]. BMC Med Res Methodol, 2018, 18(1): 24-35.
[12] Anderson Cancer Center Head and Neck Quantitative Imaging Working Group. Matched computed tomography segmentation and demographic data for oropharyngeal cancer radiomics challenges[J]. Sci Data, 2017, 4: 170077.
[13] Surveillance, Epidemiology,End Results (SEER) Program (www.seer.cancer.gov) SEER*Stat Database: Incidence - SEER Research Data, 8 Registries, Nov2021 Sub (1975-2020) - Linked To County Attributes - Time Dependent (1990-2020) Income/Rurality, 1969-2020 Counties, National Cancer Institute, DCCPS, Surveillance Research Program, released April 2023, based on the November 2022 submission.
[14] Lee C, Zame W, Yoon J, et al.Deephit: a deep learning approach to survival analysis with competing risks[C]//Proceedings of the AAAI conference on artificial intelligence. 2018.
[15] Nagpal C, Li X, Dubrawski A.Deep survival machines: fully parametric survival regression and representation learning for censored data with competing risks[J]. IEEE J Biomed Health Inform, 2021, 25(8): 3163-3175.
[16] Wang Z, Sun J.SurvTRACE: Transformers for survival analysis with competing events[C]//Proceedings of the 13th ACM International Conference on Bioinformatics, Computational Biology and Health Informatics. 2022: 1-9.
[17] AlTuwaijri AA, Alessa MA, Abuhaimed AA, et al. Lymph node yield as a prognostic factor in clinically node negative oral cavity squamous cell carcinoma[J]. Saudi Med J, 2021, 42(12): 1357-1361.
[18] Sahoo AK, Pradhan C, Barik RK, et al.DeepReco: deep learning based health recommender system using collaborative filtering[J]. Computation, 2019, 7(2): 25.
[19] Eraslan G, Avsec Ž, Gagneur J, et al.Deep learning: new computational modelling techniques for genomics[J]. Nat Rev Genet, 2019, 20(7): 389-403.