Tóm tắt
Background: Oral candidiasis is a common opportunistic infection in cancer patients, particularly those undergoing chemotherapy. Multiple clinical and hematological factors contribute to infection risk, but their complex interactions remain poorly understood using conventional statistical methods.
Objectives: To identify associated factors, and develop machine learning models to predict infection risk of oral candidiasis among cancer patients receiving and not yet receiving chemotherapy.
Materials and Methods: This cross-sectional study enrolled 69 cancer patients at Hue University of Medicine and Pharmacy Hospital between October 2024 and May 2025. Patients underwent clinical examinations, laboratory testing, direct oral swab microscopy and cultivation for candidiasis diagnosis. Sparse Partial Least Squares Discriminant Analysis (sPLS-DA) was used to select relevant features. eXtreme Gradient Boosting (XGBoost) models were developed for each patient group (chemotherapy and non-chemotherapy) and interpreted using SHapley Additive exPlanations (SHAP) value method.
Results: Oral candidiasis was detected in 36.8% of chemotherapy patients and 35.4% of non-chemotherapy patients. Key associated factors included dry mouth, taste change, white patches on mucosa, low lymphocyte or red blood cell counts, poor oral hygiene, and antibiotic use. XGBoost models achieved high performance in both groups (AUC-ROC: 0.9093 for chemotherapy; 0.8758 for non-chemotherapy). SHAP analysis revealed feature-specific contributions aligned with clinical relevance, confirming the model’s interpretability and consistency.
Conclusion: Oral candidiasis is highly prevalent among cancer patients, with distinct risk profiles between those with and without chemotherapy. Machine learning methods such as sPLS-DA and XGBoost effectively identified and interpreted predictive factors, offering valuable tools for clinical risk stratification and early prevention in oncology care.
| Đã xuất bản | 30-12-2025 | |
| Toàn văn |
|
|
| Ngôn ngữ |
|
|
| Số tạp chí | Tập 15 Số 6 (2025) | |
| Phân mục | Nghiên cứu | |
| DOI | 10.34071/jmp.2025.6.889 | |
| Từ khóa | Oral Candidiasis, Cancer, Chemotherapy, Machine Learning, sPLS-DA, XGBoost |
công trình này được cấp phép theo Creative Commons Attribution-phi thương mại-NoDerivatives 4.0 License International .
Bản quyền (c) 2026 Tạp chí Y Dược Huế
Akpan A, Morgan R. Oral candidiasis. Postgraduate medical journal. 2002;78(922):455-459.
Lalla RV, Latortue MC, Hong CH, Ariyawardana A, D’Amato-Palumbo S, et al. A systematic review of oral fungal infections in patients receiving cancer therapy. Supportive care in cancer. 2010;18(8):985-992.
Bray F, Laversanne M, Sung H, Ferlay J, Siegel RL, et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA: a cancer journal for clinicians. 2024;74(3):229-263.
Chitapanarux I, Wongsrita S, Sripan P, Kongsupapsiri P, Phakoetsuk P, et al. An underestimated pitfall of oral candidiasis in head and neck cancer patients undergoing radiotherapy: an observation study. BMC oral health. 2021;21(1):353.
Diaz PI, Hong B-Y, Dupuy AK, Choquette L, Thompson A, et al. Integrated analysis of clinical and microbiome risk factors associated with the development of oral candidiasis during cancer chemotherapy. Journal of Fungi. 2019;5(2):49.
Mayer LM, Strich JR, Kadri SS, Lionakis MS, Evans NG, et al., editors. Machine learning in infectious disease for risk factor identification and hypothesis generation: Proof of concept using invasive candidiasis. Proceedings of the Open forum infectious diseases; 2022: Oxford University Press.
Rajula HSR, Verlato G, Manchia M, Antonucci N, Fanos V. Comparison of conventional statistical methods with machine learning in medicine: diagnosis, drug development, and treatment. Medicina. 2020;56(9):455.
Châu NTM, Hòa PTN. Tỷ lệ nhiễm nấm Candida niêm mạc miệng và yếu tố liên quan ở bệnh nhân điều trị bệnh nội khoa tại Bệnh viện Trường Đại học Y-Dược Huế. Tạp chí Y Dược học - Trường Đại học Y Dược Huế. 2023;3(13):126132.
Nguyen BV, Nguyen HH, Vo T-H, Le M-T, Tran-Nguyen V-K, et al. Prevalence and drug susceptibility of clinical Candida species in nasopharyngeal cancer patients in Vietnam. One Health. 2024;18:100659.
Pulito C, Cristaudo A, Porta CL, Zapperi S, Blandino G, et al. Oral mucositis: the hidden side of cancer therapy. Journal of experimental & clinical cancer research. 2020;39:1-15.
Epstein JB, Thariat J, Bensadoun RJ, Barasch A, Murphy BA, et al. Oral complications of cancer and cancer therapy: from cancer treatment to survivorship. CA: a cancer journal for clinicians. 2012;62(6):400-422.
Patil S, Rao RS, Majumdar B, Anil S. Clinical appearance of oral Candida infection and therapeutic strategies. Frontiers in microbiology. 2015;6:1391.
Fidel Jr P. Candida-host interactions in HIV disease: implications for oropharyngeal candidiasis. Advances in dental research. 2011;23(1):45-49.
Dinarello CA. Interleukin-1 in the pathogenesis and treatment of inflammatory diseases. Blood, The Journal of the American Society of Hematology. 2011;117(14):37203732.
Kubo M. Mast cells and basophils in allergic inflammation. Current opinion in immunology. 2018;54:7479.
Tonzetich J. Production and origin of oral malodor: a review of mechanisms and methods of analysis. Journal of periodontology. 1977;48(1):13-20.
Li Z, Li J, Fu R, Liu Ja, Wen X, Zhang L. Halitosis: etiology, prevention, and the role of microbiota. Clinical oral investigations. 2023;27(11):6383-6393.
Coronado-Castellote L, Jiménez-Soriano Y. Clinical and microbiological diagnosis of oral candidiasis. Journal of clinical and experimental dentistry. 2013;5(5):e279.
Lu S-Y. Perception of iron deficiency from oral mucosa alterations that show a high prevalence of Candida infection. Journal of the Formosan Medical Association. 2016;115(8):619-627.
Deo RC. Machine learning in medicine. Circulation. 2015;132(20):1920-1930.
Rajkomar A, Dean J, Kohane I. Machine learning in medicine. New England Journal of Medicine. 2019;380(14):1347-1358.
Shi J, Chen L, Yuan X, Yang J, Xu Y, et al. A potential XGBoost Diagnostic Score for Staphylococcus aureus bloodstream infection. Frontiers in Immunology. 2025;16:1574003.
Gu Y, Su S, Wang X, Mao J, Ni X, et al. Comparative study of XGBoost and logistic regression for predicting sarcopenia in postsurgical gastric cancer patients. Scientific Reports. 2025;15(1):12808.
Lundberg SM, Lee S-I. A unified approach to interpreting model predictions. Advances in neural information processing systems. 2017;30.