Simancek presented on Thursday, March 21st about why “mitigating name recognition errors is essential to minimizing the risk of patient re-identification.” His presentation was titled: “Handling Name Errors of a BERT-Based De-Identification System: Insights from Stratified Sampling and Markov-based Pseudonymization.”
The HILS Ph.D student told DLHS Communications that missed recognition of named entities while de-identifying clinical narratives poses a critical challenge in protecting patient-sensitive health information. His paper highlighted the need for “stratified sampling and enhanced contextual considerations concerning Name tokens using a fine-tuned Longformer BERT model for clinical text de-identification.” Simancek explained that, “experimental results underscore the impact of addressing name recognition challenges in BERT-based de-identification systems for heightened privacy protection in electronic health records.”
Simancek is mentored by VD Vinod Vydiswaran, Ph.D.
