27Nov 2022
Empirical comparison of routinely collected electronic health record data for head and neck cancer‐specific survival in machine‐learnt prognostic models

Empirical comparison of routinely collected electronic health record data for head and neck cancer‐specific survival in machine‐learnt prognostic models

Cancer-specific survival (CSS) is a critical oncological outcome measure used to understand the impact of cancer treatment. This study has been prepared in accordance with the TRIPOD statement for the transparent reporting of a multivariable prediction model for individual prognosis or diagnosis. Datasets extracted from an OIS for routine clinical practice and reporting, and separately for research purposes, have demonstrated utility for informing clinical practice and 2-year CSS outcomes for patients with HNC. Data collected for research purposes identified more prognostic factors for 2-year head and neck CSS compared to data collected for routine clinical practice and reporting. However, in the absence of research datasets, institutions can use routinely collected clinical information stored in an OIS for ML prediction of HNC survival outcomes to evaluate clinical practice and develop a better understanding of the prognostic information for each new patient.

  • #oncology

Like

Save

Share