Modern healthcare systems increasingly recognize patient perspectives as essential for improving the quality of cancer care. Consequently, AI in oncology research facilitates the automated identification of patient-reported outcome measures (PROMs) and patient-reported experience measures (PREMs). These tools help clinicians address individual variability across diverse patient backgrounds by capturing real-time feedback on symptoms and treatment satisfaction. Traditional keyword-based methods often struggle to detect these nuanced data points within clinical databases.

Advanced Methodology for AI in Oncology Research

A retrospective cross-sectional study recently analyzed 24,491 oncology trials on ClinicalTrials.gov from 2012 to 2022. Researchers compared a traditional expert-based keyword search against an AI-enriched model using Bidirectional Encoder Representations from Transformers (BERT). The AI-enriched method identified that 33% of oncology studies utilized patient-centered measures, whereas the traditional method identified only 31%. This improvement highlights the superior ability of artificial intelligence to process complex medical language and natural text descriptions.

Accuracy and Clinical Impact

The AI algorithm demonstrated a remarkable 90% accuracy rate in identifying these measures, significantly outperforming the expert-based method at 84%. Furthermore, data trends show a consistent increase in the use of PROMs and PREMs over the last decade. Breast and digestive cancers accounted for nearly half of the trials incorporating these endpoints. Specifically, the EORTC QLQ-C30 questionnaire emerged as the most frequently used instrument. By accelerating the identification of these measures, AI allows medical professionals to prioritize patient-centric outcomes more effectively in future research.

Frequently Asked Questions

What are PROMs and PREMs in oncology?

PROMs are standardized questionnaires that collect health outcome data directly from patients, such as symptom severity or functional status. PREMs focus on the patient's subjective experience with the healthcare delivery process and environment.

Why is AI used for outcome identification?

AI models like BERT utilize natural language processing to understand the context within trial descriptions. This allows researchers to identify patient-centered endpoints that traditional keyword searches might miss due to linguistic variations.

Disclaimer: This content is for informational and educational purposes only. It does not constitute medical advice or a substitute for professional healthcare consultation. Refer to the latest local and national guidelines for clinical practice.

References

Soyer J et al. Artificial Intelligence for Identifying Patient-Reported Outcome and Experience Measures in Oncology: Retrospective Cross-Sectional Study Using ClinicalTrials.gov. J Med Internet Res. 2026 Apr 16. doi: 10.2196/84533. PMID: 41989827.

Wintner L et al. Patient-reported outcomes data enhances how cancer treatment toxicities are evaluated. Oncology Central. 2026.

Basch E et al. The Importance of Patient Reported Outcomes in Oncology Clinical Trials and Clinical Practice to Inform Regulatory and Healthcare Decision-Making. PMC. 2024.