Q: 68 years old male is transferred to ICU with septic shock. Patient has advanced metastatic lung cancer. The family decides on palliative care. EnMed student rotating in the ICU inquired about using machine learning technology to determine survival estimates in advanced cancer. Machine learning technology shown to provide better prognostication when compared to conventional logistic regression regarding survival estimate in advanced terminal cancer?
A) True
B) False
Answer: A
Unfortunately, most of the existing integrated prognostication tools do not accurately identify most patients who will die beyond 6 months. Newer data is showing that machine learning technology may allow better prognostication by modeling both linear and nonlinear interactions among many variables i.e., age, sex, comorbidity, stage of metastasis, solid vs. nonsolid tumor, labs, EKGs, and other variables. Artificial Intelligence (AI) based algorithms can differentiate between patients into high and low prognostic groups. Some of these algorithms are internally validated (single-center) in the studies.
Although not yet accepted as a standard of care in practice, their role in the near future would be highly enhanced regarding prognosis in advanced oncology patients.
#oncology
#AI
References:
1. Elfiky AA, Pany MJ, Parikh RB, Obermeyer Z. Development and Application of a Machine Learning Approach to Assess Short-term Mortality Risk Among Patients With Cancer Starting Chemotherapy. JAMA Netw Open 2018; 1:e180926.
2. Bertsimas D, Dunn J, Pawlowski C, et al. Applied Informatics Decision Support Tool for Mortality Predictions in Patients With Cancer. JCO Clin Cancer Inform 2018; 2:1.
3. Parikh RB, Manz C, Chivers C, et al. Machine Learning Approaches to Predict 6-Month Mortality Among Patients With Cancer. JAMA Netw Open 2019; 2:e1915997.
4.
Manz CR, Chen J, Liu M, et al. Validation of a Machine Learning Algorithm to Predict 180-Day Mortality for Outpatients With Cancer. JAMA Oncol 2020; 6:1723.
No comments:
Post a Comment