Clinical Predictive Keyboard using Statistical and Neural Language Modeling

A language model can be used to predict the next word during authoring, to\ncorrect spelling or to accelerate writing (e.g., in sms or emails). Language\nmodels, however, have only been applied in a very small scale to assist\nphysicians during authoring (e.g., discharge summaries or radiology reports).\nBut along with the assistance to the physician, computer-based systems which\nexpedite the patient's exit also assist in decreasing the hospital infections.\nWe employed statistical and neural language modeling to predict the next word\nof a clinical text and assess all the models in terms of accuracy and keystroke\ndiscount in two datasets with radiology reports. We show that a neural language\nmodel can achieve as high as 51.3% accuracy in radiology reports (one out of\ntwo words predicted correctly). We also show that even when the models are\nemployed only for frequent words, the physician can save valuable time.\n

Paper

Similar papers

© 2026 NYSGPT2525 LLC