Comparing with the manual extraction, a semiautomatic ROI detection method is used based on the modified target contour. First, a smallest enclosing rectangle of the target area is calculated through the pixels on the contour. Then, the rectangle is expanded by 3 mm to ensure all the needle information within the ROI. If any needle point candidates are located outside the enlarged rectangle, the expansion distance will be increased manually. In the end, the image surrounded by the updated rectangle is cropped for subsequent processing. A ROI for one of the clinical CT images is shown in Fig. 1.
Do you have any questions about this protocol?
Post your question to gather feedback from the community. We will also invite the authors of this article to respond.