การพัฒนารูปแบบการพยาบาลเพื่อทำนายโอกาสเกิดภาวะภูมิไวเกินโดยใช้ปัญญาประดิษฐ์ ในผู้ป่วยมะเร็งที่ได้รับยาเคมีบำบัด Paclitaxel

Nalinee  Jantajit; Somsri Thipprasopchock; Ekapob Sangariyavanich

Authors

Nalinee Jantajit National Cancer Institute
Somsri Thipprasopchock National Cancer Institute
Ekapob Sangariyavanich National Cancer Institute

Keywords:

hypersensitivity reaction, paclitaxel, artificial intelligence

Abstract

Introduction:Hypersensitivity reactions (HSRs) to paclitaxel chemotherapy are potential complications that may occur at any treatment cycle, ranging from mild symptoms to severe,
life-threatening events.

Research objectives: To develop an artificial intelligence (AI)-assisted nursing model to predict hypersensitivity reaction in cancer patients at high risk of experiencing HSRs during paclitaxel treatment.

Research methodology: This research employed a research and development design. Identify key variables in prediction. Using the Hyperparameter optimization, each model was created 1,000 times. Retrospective data were collected from electronic medical records of 173 patients.
The dataset was divided into a training dataset (91 patients), a validation dataset (40 patients), and a testing dataset (43 patients). The model was applied in a clinical setting with an independent dataset (43 patients) to evaluate its practical performance.

Results: Among the machine learning approaches tested, the Support Vector Machine (SVM) model demonstrated the highest predictive performance. In the testing dataset, the SVM achieved an area under the curve (AUC) of .80, with a positive predictive value (PPV) of 69.50% and a negative predictive value (NPV) of 96%. In real situation, the model maintained robust predictive capability with an AUC of .72. The patients identified by the AI model as high-risk and subsequently managed according to the clinical protocol, none experienced HSRs greater than grade 2, and nurses’ overall satisfaction was at a high level (M = 4.65, SD = .29)

Conclusions: The AI-assisted nursing model effectively predicts hypersensitivity reactions. The model supports early risk identification, reduces the severity of hypersensitivity reactions, and enhances patient safety.

Implications: The model should be implemented across multiple hospitals and within diverse patient populations.

Downloads

Download data is not yet available.

References

Strategy and Planning, Ministry of Public Health. Public health statistics. [Internet].2024 [cited 2024 Jun 17]. Available from: https://spd.moph.go.th/wp-content/uploads/2023/11/Hstatistic65.pdf.

Common Terminology Criteria for Adverse Events (CTCAE) Version 5.0 [Internet]. 2017 [cited 2024 Sep 18]. Available from: https://dctd.cancer.gov/research/ctep-trials/for-sites/adverse-events/ctcae-v5-8x11.pdf.

Thangwonglers T, Santimaleeworagun W, Therasakvichya S, Saengsukkasemsak N, Pimsi P. Characteristics of immediate hypersensitivity reaction to paclitaxel-based chemotherapy in gynecologic cancer patients. Asian Pacific Journal of Allergy and Immunology 2023;41(4):340-6.

Topol EJ. High-performance medicine: the convergence of human and artificial intelligence. Nature Medicine. 2019;25(1):44–56.

Jiang F, Jiang Y, Zhi H, Dong Y, Li H, Ma S, et al. Artificial intelligence in healthcare: past, present, and future. Stroke and Vascular Neurology 2017;2(4):230–43.

Islam KR, Prithula J, Kumar J, Tan TL, Reaz MBI, Sumon MSI, et al. Machine learning‑based early prediction of sepsis using electronic health records: A systematic review. Journal of Clinical Medicine 2023;12(17):5658.

Liu T, Krentz A, Lu L, Curcin V. Machine learning based prediction models for cardiovascular disease risk using electronic health records data: systematic review and meta-analysis. European heart journal Digital health 2024;6(1):7-12.

Sa-Nguansai S, Sukphinetkul R. Development and validation of a clinical prediction model for paclitaxel hypersensitivity reaction on the basis of real-world data: Pac-HSR score. JCO Global Oncology2024;10:e2400318

Harrell FE Jr, Lee KL, Mark DB. Multivariable prognostic models: Issues in developing models,evaluating assumptions and adequacy, and measuring and reducing errors. Statistics in Medicine 1996;15(4):361–87.

Vapnik VN. Statistical learning theory. New York: Wiley; 1998.

Foongfaung S. Nursing management of hypersensitivity and infusion reactions from intravenously administered chemotherapeutic agents. Thai Cancer Journal 2018;38(1):29-41. (in Thai)

Daniel WW, Cross CL. Determination of sample size for estimating proportions. Biostatistics: A Foundation for Analysis in the Health Sciences. 1999;8:189-90.

Jantajit N, Pornpitak. Thipprasopchock S. Early symptoms and factors on hypersensitivity reaction of cancer patient with receiving paclitaxel and carboplatin chemotherapy. Thai Cancer Journal 2025;45(1):24-14. (in Thai)

Tsao LR, Young FD. Hypersensitivity reactions to platinum agents and taxanes. Clin Rev Allergy Immunol 2022;63(1):123-46.

Park HJ, Ko M, Kim I-W, Oh JM. Personalized risk assessment for taxane-induced hypersensitivity reactions: A systematic review and meta-analysis. Journal of Personalized Medicine 2025;15(1):2.

Taha HA, Zeilani RS, Haddad RH, Abdalrahim MS. Artificial intelligence and machine learning techniques for predicting neuropathic pain in patients with cancer: A systematic review. Digital Health 2025;11:20552076251358315.

O'Connor S, Vercell A, Wong D, Yorke J, Fallatah FA, Cave L, Anny Chen LY. The application and use of artificial intelligence in cancer nursing: A systematic review. European Journal of Oncology Nursing 2024;68:102510.

Hosmer Jr, DW, Lemeshow S, Sturdivant RX. Applied logistic regression. 3rd ed, John Wiley & Sons; 2013.

Hu Q, Chen Y, Zou D, He Z, Xu T. Predicting adverse drug event using machine learning based on electronic health records: a systematic review and meta-analysis. Frontiers in Pharmacol. 2024;15:1497397.