Keep it simple: Designing a tool to automatically extract ROI values

Automatically extracting region of interest (ROI) values from a PACS viewer can be easily and accurately achieved with a module based on open-source software, and vendors should include such functionality in future versions of their PACS, according to a group of Korean researchers.

Drawing ROIs and then manually inputting the related statistical values is “one of the most uninteresting and tedious tasks in radiologic reading and research,” argued Young Han Lee, MD, PhD, of the Research Institute of Radiological Science at Yonsei University College of Medicine in Seoul, and colleagues.

To ease this repetitive and error-prone task, Lee and colleagues sought to use optical character recognition (OCR) technology to convert imaged values into text data. They detailed their work in a study published Aug. 12 in Academic Radiology.

The module they developed was based on open-source software including GOCR, for the OCR method, and AutoHotkey, for the macro function. “By combining an OCR module with a macro program, we can extract ROI values automatically and without error,” wrote the authors. “Manual readings and repetitive typewriting can be minimized, and a high degree of accuracy can be achieved.”

The process, tested using MR images, was described as follows:

1.       ROI values are captured as a graphic file

2.       OCR software recognizes the text

3.       Error-correction

4.       Values including area, average, standard deviation and more are extracted from the text

5.       Values are reformatted into temporary strings with tabs

6.       Temporary strings are pasted into a spreadsheet

After evaluating this process with 1,040 recognitions from 280 randomly selected ROIs from the MR images, Lee and colleagues reported 100 percent accuracy. Moreover, inputting the values with the module took roughly one-fifth the time of manual entry. Average input times with conventional methods averaged 34.97 seconds compared with 7.87 seconds for the module-assisted method.

Lee and colleagues wondered why commercially available PACS viewers don’t already make it easy to save or copy ROI values, and hope this and other beneficial functionality can be included in the next generation of PACS.

“We expect further potential applications in radiologic reading, such as ROI subtraction on in-phase and opposed-phase MR of the adrenal gland, ROI comparison of the fat fraction map of an MR image, and a comparison of the Hounsfield unit in CT scans,” wrote the authors.

Evan Godt
Evan Godt, Writer

Evan joined TriMed in 2011, writing primarily for Health Imaging. Prior to diving into medical journalism, Evan worked for the Nine Network of Public Media in St. Louis. He also has worked in public relations and education. Evan studied journalism at the University of Missouri, with an emphasis on broadcast media.

Trimed Popup
Trimed Popup