دورية أكاديمية
Evaluating High-Dimensional Machine Learning Models to Predict Hospital Mortality Among Older Patients With Cancer.
| العنوان: | Evaluating High-Dimensional Machine Learning Models to Predict Hospital Mortality Among Older Patients With Cancer. |
|---|---|
| المؤلفون: | Qiao EM; Department of Radiation Medicine and Applied Sciences, University of California, San Diego, La Jolla, CA., Qian AS; Department of Radiation Medicine and Applied Sciences, University of California, San Diego, La Jolla, CA., Nalawade V; Department of Radiation Medicine and Applied Sciences, University of California, San Diego, La Jolla, CA., Voora RS; Department of Radiation Medicine and Applied Sciences, University of California, San Diego, La Jolla, CA., Kotha NV; Department of Radiation Medicine and Applied Sciences, University of California, San Diego, La Jolla, CA., Vitzthum LK; Department of Radiation Oncology, Stanford University, Stanford, CA., Murphy JD; Department of Radiation Medicine and Applied Sciences, University of California, San Diego, La Jolla, CA. |
| المصدر: | JCO clinical cancer informatics [JCO Clin Cancer Inform] 2022 Jun; Vol. 6, pp. e2100186. |
| نوع المنشور: | Journal Article |
| اللغة: | English |
| بيانات الدورية: | Publisher: American Society of Clinical Oncology Country of Publication: United States NLM ID: 101708809 Publication Model: Print Cited Medium: Internet ISSN: 2473-4276 (Electronic) Linking ISSN: 24734276 NLM ISO Abbreviation: JCO Clin Cancer Inform Subsets: MEDLINE |
| أسماء مطبوعة: | Original Publication: Alexandria, VA : American Society of Clinical Oncology, [2017]- |
| مواضيع طبية MeSH: | Frailty* , Neoplasms*/diagnosis , Neoplasms*/therapy, Hospital Mortality ; Humans ; Machine Learning ; Retrospective Studies |
| مستخلص: | Purpose: Older hospitalized cancer patients face high risks of hospital mortality. Improved risk stratification could help identify high-risk patients who may benefit from future interventions, although we lack validated tools to predict in-hospital mortality for patients with cancer. We evaluated the ability of a high-dimensional machine learning prediction model to predict inpatient mortality and compared the performance of this model to existing prediction indices. Methods: We identified patients with cancer older than 75 years from the National Emergency Department Sample between 2016 and 2018. We constructed a high-dimensional predictive model called Cancer Frailty Assessment Tool (cFAST), which used an extreme gradient boosting algorithm to predict in-hospital mortality. cFAST model inputs included patient demographic, hospital variables, and diagnosis codes. Model performance was assessed with an area under the curve (AUC) from receiver operating characteristic curves, with an AUC of 1.0 indicating perfect prediction. We compared model performance to existing indices including the Modified 5-Item Frailty Index, Charlson comorbidity index, and Hospital Frailty Risk Score. Results: We identified 2,723,330 weighted emergency department visits among older patients with cancer, of whom 144,653 (5.3%) died in the hospital. Our cFAST model included 240 features and demonstrated an AUC of 0.92. Comparator models including the Modified 5-Item Frailty Index, Charlson comorbidity index, and Hospital Frailty Risk Score achieved AUCs of 0.58, 0.62, and 0.71, respectively. Predictive features of the cFAST model included acute conditions (respiratory failure and shock), chronic conditions (lipidemia and hypertension), patient demographics (age and sex), and cancer and treatment characteristics (metastasis and palliative care). Conclusion: High-dimensional machine learning models enabled accurate prediction of in-hospital mortality among older patients with cancer, outperforming existing prediction indices. These models show promise in identifying patients at risk of severe adverse outcomes, although additional validation and research studying clinical implementation of these tools is needed. Competing Interests: Lucas K. VitzthumResearch Funding: RefleXion Medical (Inst) James D. MurphyConsulting or Advisory Role: Boston Consulting GroupResearch Funding: eContourNo other potential conflicts of interest were reported. |
| References: | JAMA Surg. 2017 Feb 1;152(2):175-182. (PMID: 27893030) CA Cancer J Clin. 2017 Sep;67(5):362-377. (PMID: 28731537) J Clin Oncol. 2008 Dec 20;26(36):5863-9. (PMID: 19018082) Ann Oncol. 2015 Jun;26(6):1091-1101. (PMID: 25403592) JCO Clin Cancer Inform. 2018 Dec;2:1-11. (PMID: 30652595) Cancer Med. 2021 Dec;10(24):9129-9138. (PMID: 34821051) J Clin Oncol. 2020 Nov 1;38(31):3652-3661. (PMID: 32886536) Cancer. 1997 Nov 15;80(10):1973-80. (PMID: 9366301) Ann Surg Oncol. 2010 Jul;17(7):1802-7. (PMID: 20155401) JAMA Netw Open. 2019 Oct 2;2(10):e1915997. (PMID: 31651973) J Crit Care. 2011 Aug;26(4):388-94. (PMID: 21195582) Breast J. 2021 Jun;27(6):509-513. (PMID: 33650221) Lancet. 2018 May 5;391(10132):1775-1782. (PMID: 29706364) J Oncol Pract. 2018 Mar;14(3):e168-e175. (PMID: 29206553) Surg Endosc. 2017 Jun;31(6):2509-2519. (PMID: 27699515) JAMA Oncol. 2017 Oct 12;3(10):e172450. (PMID: 28859189) Neurosurgery. 2020 Dec 15;88(1):147-154. (PMID: 32803222) Future Healthc J. 2019 Jun;6(2):94-98. (PMID: 31363513) J Chronic Dis. 1987;40(5):373-83. (PMID: 3558716) JAMA Surg. 2014 Jul;149(7):633-40. (PMID: 24804971) J Palliat Med. 2012 Aug;15(8):902-9. (PMID: 22663175) J Natl Compr Canc Netw. 2016 Aug;14(8):979-87. (PMID: 27496114) Cancer. 2014 Jan 1;120(1):134-41. (PMID: 24122413) |
| تواريخ الأحداث: | Date Created: 20220607 Date Completed: 20220609 Latest Revision: 20230608 |
| رمز التحديث: | 20240829 |
| مُعرف محوري في PubMed: | PMC9225681 |
| DOI: | 10.1200/CCI.21.00186 |
| PMID: | 35671416 |
| قاعدة البيانات: | MEDLINE |
| تدمد: | 2473-4276 |
|---|---|
| DOI: | 10.1200/CCI.21.00186 |