A Natural Language Processing Model for COVID-19 Detection Based on Dutch General Practice Electronic Health Records by Using Bidirectional Encoder Representations From Transformers: Development and Validation Study

Maarten Homburg   ¹ , MSc, MD ;   Eline Meijer   ^{1, 2} , MSc ;   Matthijs Berends   ^{1, 3, 4} , PhD ;   Thijmen Kupers   ^{1, 2} , MSc ;   Tim Olde Hartman   ⁵ , MD, PhD ;   Jean Muris   ⁶ , MD, Prof Dr ;   Evelien de Schepper   ⁷ , MD, PhD ;   Premysl Velek   ⁷ , MSc ;   Jeroen Kuiper   ⁸ , MSc ;   Marjolein Berger   ¹ , MD, Prof Dr ;   Lilian Peters   ^{1, 2, 9} , PhD

¹ Department of Primary- and Long-Term Care, University Medical Center Groningen, Groningen, The Netherlands

² Data Science Center in Health, University Medical Center Groningen, Groningen, The Netherlands

³ Department of Medical Microbiology and Infection Prevention, University Medical Center Groningen, Groningen, The Netherlands

⁴ Department of Medical Epidemiology, Certe Foundation, Groningen, The Netherlands

⁵ Department of Primary and Community Care, Radboud University Nijmegen Medical Center, Nijmegen, The Netherlands

⁶ Care and Public Health Research Institute, Department of Family Medicine, Maastricht University Medical Center, Maastricht, The Netherlands

⁷ Department of General Practice, Erasmus Medical Center, Rotterdam, The Netherlands

⁸ Municipal Health Service Groningen, Groningen, The Netherlands

⁹ Midwifery Science, Amsterdam Public Health, Vrije Universiteit Amsterdam, Amsterdam University Medical Center, Amsterdam, The Netherlands