Unmasking bias in artificial intelligence: a systematic review of bias detection and mitigation strategies in electronic health record-based models.

OBJECTIVES: Leveraging artificial intelligence (AI) in conjunction with electronic health records (EHRs) holds transformative potential to improve healthcare. However, addressing bias in AI, which risks worsening healthcare disparities, cannot be overlooked. This study reviews methods to handle various biases in AI models developed using EHR data. MATERIALS AND METHODS: We conducted a systematic review following the Preferred Reporting Items for Systematic Reviews and Meta-analyses guidelines, analyzing articles from PubMed, Web of Science, and IEEE published between January 01, 2010 and December 17, 2023. The review identified key biases, outlined strategies for detecting and mitigating bias throughout the AI model development, and analyzed metrics for bias assessment. RESULTS: Of the 450 articles retrieved, 20 met our criteria, revealing 6 major bias types: algorithmic, confounding, implicit, measurement, selection, and temporal. The AI models were primarily developed for predictive tasks, yet none have been deployed in real-world healthcare settings. Five studies concentrated on the detection of implicit and algorithmic biases employing fairness metrics like statistical parity, equal opportunity, and predictive equity. Fifteen studies proposed strategies for mitigating biases, especially targeting implicit and selection biases. These strategies, evaluated through both performance and fairness metrics, predominantly involved data collection and preprocessing techniques like resampling and reweighting. DISCUSSION: This review highlights evolving strategies to mitigate bias in EHR-based AI models, emphasizing the urgent need for both standardized and detailed reporting of the methodologies and systematic real-world testing and evaluation. Such measures are essential for gauging models' practical impact and fostering ethical AI that ensures fairness and equity in healthcare.

Moving Forward in the Next Decade: Radiation Oncology Sciences for Patient-Centered Cancer Care.

In a time of rapid advances in science and technology, the opportunities for radiation oncology are undergoing transformational change. The linkage between and understanding of the physical dose and induced biological perturbations are opening entirely new areas of application. The ability to define anatomic extent of disease and the elucidation of the biology of metastases has brought a key role for radiation oncology for treating metastatic disease. That radiation can stimulate and suppress subpopulations of the immune response makes radiation a key participant in cancer immunotherapy. Targeted radiopharmaceutical therapy delivers radiation systemically with radionuclides and carrier molecules selected for their physical, chemical, and biochemical properties. Radiation oncology usage of "big data" and machine learning and artificial intelligence adds the opportunity to markedly change the workflow for clinical practice while physically targeting and adapting radiation fields in real time. Future precision targeting requires multidimensional understanding of the imaging, underlying biology, and anatomical relationship among tissues for radiation as spatial and temporal "focused biology." Other means of energy delivery are available as are agents that can be activated by radiation with increasing ability to target treatments. With broad applicability of radiation in cancer treatment, radiation therapy is a necessity for effective cancer care, opening a career path for global health serving the medically underserved in geographically isolated populations as a substantial societal contribution addressing health disparities. Understanding risk and mitigation of radiation injury make it an important discipline for and beyond cancer care including energy policy, space exploration, national security, and global partnerships.