Clinical coding is a critical task in healthcare, although traditional methods for automating clinical coding may not provide sufficient explicit evidence for coders in production environments. This evidence is crucial, as medical coders have to make sure there exists at least one explicit passage in the input health record that justifies the attribution of a code. We therefore propose to reframe the task as an entity linking problem, in which each document is annotated with its set of codes and respective textual evidence, enabling better human-machine collaboration. By leveraging parameter-efficient fine-tuning of Large Language Models (LLMs), together with constrained decoding, we introduce three approaches to solve this problem that prove effective at disambiguating clinical mentions and that perform well in few-shot scenarios.

The coding of medical diagnosis and treatment has always been a challenging issue. Translating a patient's complex symptoms, and a clinician's efforts to address them, into a clear and unambiguous classification code was difficult even in simpler times. Now, however, hospitals and health insurance companies want very detailed information on what was wrong with a patient and the steps taken to treat them-- for clinical record-keeping, for hospital operations review and planning, and perhaps most importantly, for financial reimbursement purposes. The current international standard for medical coding is ICD-10 (the tenth version of International Classification of Disease codes), from the World Health Organization (WHO). ICD‑10 has over 14,000 codes for diagnoses.

An experiment in which humans and AI augmented each other's strengths demonstrates how leaders can reimagine processes to create greater business value and prepare for the next wave of innovation. This article is part of an MIT SMR initiative exploring how technology is reshaping the practice of management. In the longstanding argument about whether AI will replace or complement human beings, the new watchword is symbiosis. Most recently, Elon Musk used the term to describe how a brain implant might merge human and digital intelligence. But you don't need to go full cyborg to achieve a mutually beneficial relationship between humans and AI.

We recently asked Alexa if she could code a few medical charts for us. "Sorry I don't know that." After all, the U.S. healthcare industry spends billions of dollars on 250,000 medical coders every year to do the job. This way of doing business might be error-prone, inefficient, and bound by constantly changing regulations, but hey, IT IS a solution. But you know what else is a solution?