RADLogics Showcases Virtual-Resident™ — Image Analysis Clinical Decision Support Solution for Radiologists—Achieving Up to 44% Improvement in Radiologists Productivity

Share Article

At RSNA 2018 RADLogics demonstrates how its Virtual-Resident machine learning image analysis platform can boost radiologists productivity and improve quality.

RADLogics' Virtual-Resident Clinical Decision Support Solution Achieves Up to 44% Increase in Radiologists Productivity"

RADLogics, the pioneer in Machine-Learning driven medical imaging analysis solutions, showcases results from a first-of-its-kind clinical study evaluating the effect of an integrated detection and report creation environment, a study which used its Virtual Resident software—a clinical decision support solution that helps radiologists achieve long-awaited leaps in productivity and quality.

In a UCLA study led by Matt Brown, PhD, recently published in Academic Radiology, a team of researchers and radiologists evaluated the potential for improved efficiency and report completion time for chest CTs. RADLogics’ Virtual Resident software platform--running numerous machine-learning image analysis algorithms--was integrated into the PACS and configured to automatically incorporate results and measurement data into a standardized radiology report (Nuance Powerscribe 360). Results indicate that using this integrated system of automated detection and report generation decreased radiologists’ time to evaluate a chest CT and create a final report by up to 44%. Dr. Brown’s study also indicates that the overall findings detection rate of the system was comparable to radiologists who did not use the software. All measurements were consistent between the radiologists as well as the software, with a variation of 1mm or less.

The UCLA study also found that additional potentially valuable information, typically not measured or included in the report by an overworked radiologist (e.g., nodule volume, aortic diameter, free fluid/free air volume) was consistently quantified, and the numeric data automatically included in the final report, resulting in a more informative and valuable report, at no additional time cost to the radiologist.

“This UCLA study by Dr. Brown and his team showed that using automated CAD assistance and an integrated report generation system can significantly decrease interpretation and final report creation times, while creating a more clinically valuable report,” says Patrick Browning, MD, Chief Medical Officer, RADLogics. “When consistently used and integrated into the radiologist’s work environment, the RADLogics Virtual Resident is capable of enabling the harried radiologist to evaluate studies and create reports of equal or greater quality in significantly less time, enabling increased productivity and/or decreased stress. An fully integrated solution also preserves the existing workflow, minimizing disruption while improving quality of care radiologists deliver.”

The RADLogics cloud-based Virtual-Resident enables machine learning image analysis algorithms to search and analyze imaging data associated with CTs, MRIs and X-rays. In addition to the speed and productivity improvements, the cloud-based solution can improve interpretation consistency among radiologists. RADLogics Virtual Resident-powered algorithms provide consistent objective measurements and numeric characterization of findings, lowering variability when comparing studies for the same patient and tracking findings over time. In the end, this results in a more valuable report, and improves the quality of the healthcare radiologists provide.

About RADLogics
RADLogics’ mission is to create clinical decision support solutions that help radiologists provide higher value reports to better serve referring physicians and patients. RADLogics refers to this capability as Virtual Resident.

Contact Information
info at RADLogics.com
RSNA North Hall, Booth #6361

Share article on social media or email:

View article via:

Pdf Print

Contact Author

Moshe Becker
+1 (408) 905-8054
Email >
Follow >
Visit website