Residents' structured reports get boost from software assistance
Software-assisted structured reporting can give radiology residents approaching multiparametric prostate MRI exams a significant boost when it comes to interpretation times.
Structured reporting templates are known to have numerous clinical benefits, such as increased consistency and improved communication between providers, but less is known about how these reports can be utilized as training tools for residents. A new analysis in the European Journal of Radiology details how training dedicated to structured reporting specifically can improve residents’ skills at this task.
“Traditionally, these reports have been written predominantly in a narrative style, relying heavily on free text,” corresponding author Andrea Ponsiglione, fromt he Department of Advanced Biomedical Sciences at the University of Naples Federico in Italy, and colleagues noted. “However, radiologists may sometimes omit key details due to a lack of understanding of what referring clinicians specifically require, potentially failing to fully meet the needs of patients and physicians. With the increasing complexity and volume of imaging studies, there is growing evidence supporting a shift in preference towards structured reporting among both referring clinicians and radiologists in different fields.”
For the analysis, the team utilized software specific to structured reporting that converts radiological and clinical findings into interactive decision trees. They centered their focus on how the assistance could improve residents’ skills at reporting multiparametric MRI (mpMRI) performed for prostate cancer detection and staging—a challenging task with known variabilities in results, even among experienced readers.
The group first underwent six hours of intensive training on interpreting prostate mpMRI studies, before reading a set of 100 exams. The dataset was split into two, with all residents first reporting on the first half using basic narrative text and half the residents utilizing software assistance for the remaining 50 cases. Report completeness was assessed using a PI-RADS-based checklist.
Residents who were able to access software assistance displayed an increase in word counts but a decrease in reporting time compared to their first batch of narrative style reports. Those same residents also showed significant improvements in concordance with an expert radiologist who reviewed the scans.
The residents’ narrative reports were notably missing important details, including image quality and final impressions, both of which are crucial elements, especially when describing and staging cancerous lesions.
“It is essential that referring physicians are informed of the true diagnostic value of a scan, as inadequate image quality can severely compromise clinical decision-making and patient care,” the authors cautioned. “The omission of final impressions from all generated NR is likewise alarming. Final impressions play a crucial role in summarizing the key findings for the referring physician while also providing additional context that may influence the overall level of suspicion.”
The findings highlight a need for solutions that can address these critical shortcomings. The group suggested that their software-assisted structured reporting training strategy represents a viable means of doing so without disrupting clinical workflows, but rather improving them.
“This approach not only enhances the efficiency of the reporting process by requiring the reader to focus on all relevant findings, but also positively impacts diagnostic accuracy," the authors concluded.
Learn more about the analysis here.
In addition to her background in journalism, Hannah also has patient-facing experience in clinical settings, having spent more than 12 years working as a registered rad tech. She joined Innovate Healthcare in 2021 and has since put her unique expertise to use in her editorial role with Health Imaging.