Length of Stay Prediction in Neurosurgery with Russian GPT-3 Language Model Compared to Human Expectations
Gleb Danilov, Konstantin Kotik, E V Shevchenko, Dmitriy Usachev, Michael Shifrin, Yulia Strunina, Tatyana Tsukanova, Timur Ishankulov, Vasiliy Lukshin, Potapov Aa
Abstract
Patients, relatives, doctors, and healthcare providers anticipate the evidence-based length of stay (LOS) prediction in neurosurgery. This study aimed to assess the quality of LOS prediction with the GPT3 language model upon the narrative medical records in neurosurgery comparing to doctors' and patients' expectations. We found no significant difference (p = 0.109) between doctors', patients', and model's predictions with neurosurgeons tending to be more accurate in prognosis. The modern neural network language models demonstrate feasibility in LOS prediction.
Topics & Concepts
NeurosurgeryNarrativeMedical recordArtificial neural networkQuality (philosophy)MedicineComputer scienceMedical physicsPsychologyMedical emergencyInternal medicineArtificial intelligenceRadiologyLinguisticsEpistemologyPhilosophyCardiac, Anesthesia and Surgical OutcomesSurgical Simulation and TrainingArtificial Intelligence in Healthcare and Education