Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2025 May 20;25(1):902.
doi: 10.1186/s12885-025-14222-9.

Rapid identification of tumor patients with PG-SGA ≥ 4 based on machine learning: a prospective study

Gui Qian  1 , Huang Jiaxin  2 , Cong Minghua  2 , Liu Beijia  1 , Li Yinfeng  3 , Huang Guiyu  4 , Yang Mingxue  3 , Tang Xiaoli  5 , Yan Hongyan  6
Affiliations

Rapid identification of tumor patients with PG-SGA ≥ 4 based on machine learning: a prospective study

Gui Qian et al. BMC Cancer. .

Abstract

Background: Malnutrition is common in cancer patients and worsens treatment and prognosis. The Patient-Generated Subjective Global Assessment (PG-SGA) is the best tool to evaluate malnutrition, but it is complicated has limited its routine clinical use.

Methods: We reviewed 798 records from 416 cancer patients treated at our hospital from July 2022 to March 2024. We used machine learning methods like XGBoost and Random Forest to find important factors linked to PG-SGA scores of 4 or higher. We confirmed the most important factors with logistic regression analysis.

Results: Among all models, XGBoost and Random Forest models perform the best, with the area under the curve (AUC) reaching of 0.75 and 0.77. Multivariate logistic regression analysis identified body mass index (BMI) (OR = 0.82, 95%CI 0.66-0.99; P = 0.045), handgrip strength (HGS) (OR = 0.89, 95%CI 0.82-0.96; P = 0.004), fat-free mass index (FFMI) (OR = 1.36, 95%CI 1.01-1.88; P = 0.045), and bedridden status (OR = 3.16, 95%CI 1.17-9.14; P = 0.026) as key predictors for PG-SGA scores of ≥ 4.

Conclusion: BMI, HGS, FFMI, and bedridden status were identified as practical indicators to efficiently screen patients likely to have PG-SGA scores ≥ 4.

Keywords: Cancer; Machine Learning; Nutritional Assessment; Patient-Generated Subjective Global Assessment.

PubMed Disclaimer

Conflict of interest statement

Declarations. Ethics approval and consent to participate: All experiments involving human participants and/or human tissue samples in this study were performed in accordance with the Declaration of Helsinki and relevant institutional guidelines and regulations. Ethical approval was obtained from the Ethics Committee of Sichuan Provincial Cancer Hospital (Approval Number: SCCHEC-02–2022-066). Written informed consent was obtained from all participants prior to enrollment. Consent for publication: We confirm our consent for the manuscript titled “Rapid Identification of Tumor Patients with PG-SGA ≥ 4 Based on Machine Learning: A prospective Study” to be submitted for publication in BMC cancer. We agree to the submission of the manuscript and its contents to be published in the journal if accepted. Competing interests: The authors declare no competing interests.

Figures

Fig. 1
Fig. 1
Research flow chart. Abbreviation: ML, machine learning; SHAP, Shapley Additive Explanations; SVM, Support Vector Machine
Fig. 2
Fig. 2
Comparison of calibration curves for different machine learning models
Fig. 3
Fig. 3
Comparison of Receiver Operating Characteristic (ROC) curves of different machine learning models on the test set
Fig. 4
Fig. 4
Shapley Additive Explanations (SHAP) analysis for two models
Fig. 5
Fig. 5
The restrict cubic spline plots for continuous variables. The horizontal dashed line represents the reference OR of 1.0. 95% CI 95% confidence interval
See this image and copyright information in PMC

Similar articles

See all similar articles

References

    1. Powrózek T, Dziwota J, Małecka-massalska T. Nutritional Deficiencies in Radiotherapy-Treated Head and Neck Cancer Patients [J]. J Clin Med, 2021, 10(4). - PMC - PubMed
    1. Liu C, Lu Z, Li Z, et al. Influence of Malnutrition According to the GLIM Criteria on the Clinical Outcomes of Hospitalized Patients With Cancer [J]. Front Nutr. 2021;8: 774636. - PMC - PubMed
    1. Steer B, Loeliger J, Edbrooke L, et al. Malnutrition Prevalence according to the GLIM Criteria in Head and Neck Cancer Patients Undergoing Cancer Treatment [J]. Nutrients, 2020, 12(11). - PMC - PubMed
    1. Yu H, Chen M, Zhang J. Nutritional assessment for 891 patients with common cancer in a cancer hospital of southwest China [J]. J clin oncol, 2024, 42(16_suppl): e23280-e.
    1. Kaźmierczak-siedlecka K, Skonieczna-żydecka K, Folwarski M, et al. Influence of malnutrition stage according to GLIM 2019 criteria and SGA on the quality of life of patients with advanced cancer [J]. Nutr hosp. 2020;37(6):1179–85. - PubMed

LinkOut - more resources