Harrisburg University of Science and Technology faculty member Mark Newman and Ph.D. students Bowen Long and Fangya Tan recently published ground-breaking research. Their study, titled “Ensemble DeBERTa Models on USMLE Patient Notes Automatic Scoring using Note-based and Character-based approaches”, introduces an innovative method to enhance the accuracy and efficiency of the United States Medical Licensing Examination (USMLE) Step 2 clinical skills examination scoring process.
The USMLE Step 2 clinical skills exam requires the evaluation of patient notes, which are conventionally scored manually by physician raters. This traditional method can be time-consuming and inconsistent. To address these challenges, the authors, affiliated with Harrisburg University, have proposed an ensemble model that automates the scoring process, ensuring a more accurate and efficient evaluation.
This research showcases the immense potential of ensemble models in Natural Language Understanding tasks, especially when multiple DeBERTa models are combined using a Note-based approach with LSTM. This innovation promises to revolutionize the scoring process for USMLE patient notes by enhancing accuracy and efficiency.
The paper appears in Vol. 6 No. 1 (2023): 2023 International Conference on Automation, Computer Technology and Intelligent Computing (CACTIC 2023）and is found here https://madison-proceedings.com/index.php/aetr/article/view/1229
Moreover, the research reflects the high caliber of academic prowess and innovation at Harrisburg University of Science and Technology. The collaboration between faculty and Ph.D. students in the Data Science program demonstrates the University’s commitment to contributing to cutting-edge healthcare education and evaluation developments. This research is a stepping stone for further exploration into how Artificial Intelligence and advanced ensemble models can optimize critical processes in the healthcare field.
More on the Ph.D. in Data Sciences at Harrisburg University is found at https://www.harrisburgu.edu/programs/phd-data-sciences/
ABOUT HARRISBURG UNIVERSITY
Accredited by the Middle States Commission on Higher Education, Harrisburg University is a private nonprofit university offering bachelor and graduate degree programs in science, technology, and math fields. For more information on the University’s affordable demand-driven undergraduate and graduate programs, call 717-901-5146 or email, Connect@HarrisburgU.edu. Follow on Twitter (@HarrisburgU) and Facebook (Facebook.com/HarrisburgU.
