A Clinician's Guide to Running Custom Machine-Learning Models in an Electronic Health Record Environment

We recently brought an internally developed machine-learning model for predicting which patients in the emergency department would require hospital admission into the live electronic health record environment. Doing so involved navigating several engineering challenges that required the expertise of multiple parties across our institution. Our team of physician data scientists developed, validated, and implemented the model. We recognize a broad interest and need to adopt machine-learning models into clinical practice and seek to share our experience to enable other clinician-led initiatives. This Brief Report covers the entire model deployment process, starting once a team has trained and validated a model they wish to deploy in live clinical operations.

Comparison of the distribution of tissue kallikrein and esterase A, a kallikrein-like enzyme, in rat kidney using specific monoclonal antibodies

Tissue kallikrein (E.C. 3.4.21.35) and arginine esterase A, another closely related, kinin-generating serine protease, have been localized by immunocytochemistry in rat kidney, using monoclonal antibodies that do not crossreact with other kallikrein-related enzymes or with tonin. Kallikrein was present primarily in the apical cytoplasm of the connecting tubule and the cortical collecting duct. Esterase A, on the other hand, was present primarily in the basolateral region of both proximal and distal straight tubules in the outer medulla and medullary rays. In addition, esterase A was demonstrable in distal convoluted tubules and, to a lesser extent, in proximal convoluted tubules. The presence of different kinin-generating enzymes at these sites would permit the formation of kinins from appropriate substrates on both the vascular and luminal poles of separate segments of the kidney tubule.