Nearly every day exciting articles are published about how AI is being used in pathology and radiology to improve the quality of reporting and the follow-up of salient findings. A recent article discussed such AI software (see: How University of Rochester uses AI to reduce risk of failed follow-up). Below is an excerpt from it:
The University of Rochester Medical Center needed a better way to ensure that its many patients with incidental radiology findings received their recommended follow-up care in a timely manner. Failure to follow up happens for a number of reasons, including inconsistent communications during care transitions, not notifying patients of actionable test results, and inadequate systems for managing and tracking incidental findings.....In 2015, the provider organization piloted a recommendation tracking system it calls “Backstop”....The goal of the system was to serve as a safety net for patients for whom clinicians had identified a potential malignancy or aneurysm and offered an actionable recommendation....[However,] the manual process required to flag recommendations was a significant barrier to widespread adoption of the program. The Backstop program depended on the radiologist manually adding patient cases to a central database for tracking at the time of dictation.
....Backstop needed to take advantage of recent advances in natural language processing algorithms to help identify and track more of the recommendations coming out of the department (see: A Backstop that reduces risk of delayed diagnosis).....[The radiology department] partnered with Nuance to integrate mPower, their NLP-based clinical analytics solution, into...[the] Backstop system. mPower [a Nuance product,] contains radiologist-designed and -validated algorithms that automatically identify actionable recommendations from unstructured radiology report text. Without disruption to the provider organization’s radiologists’ workflow, the NLP-based clinical analytics provided another layer of protection for patients.
To summarize, the University of Rochester radiology department is using their Backstop program with added Nuance software to ensure that actionable recommendations for follow-up in its reports are being attended to. Such followup recommendations relate mainly to "possible" malignancies or aneurysms. Nuance natural language programming (NLP) was added to Backstop program to avoid the necessity of manually adding cases to a central database that includes followup recommendations. Instead, such follow-up recommendations are now extracted from the natural language in the radiology report.
Until recently, recommendations for followup in radiology and pathology reports would have been covered by case coding by the pathologist and radiologist. This would allow such recommendations to be identified in the future but, of course, such cases would need to be identified by running the relevant search program. It may be possible in the near future for coding of pathology reports to be totally avoided by the use of ever more sophisticated NLP software. In fact, automating this coding process has the potential to be accurate than the dictated codes for a case. This is part of the ongoing process of removing manual steps from diagnostic report generation and thus improvising pathologist and radiologist work efficiency.
Comments