Prostate nuclear magnetism ultrasonic image registration fusion method

Abstract

The invention provides a prostate nuclear magnetism ultrasonic image registration fusion method, and belongs to the field of computer-aided diagnosis. The prostate nuclear magnetic ultrasonic image registration and fusion method comprises the following steps: firstly, training an active appearance model of a prostate ultrasonic image for an image segmentation task, and establishing a mathematicalmodel of a boundary driving function based on a random forest to realize automatic segmentation of the ultrasonic image; secondly, extracting a pre-operation segmented nuclear magnetic image and an automatically segmented ultrasonic image to establish a shape feature vector of a contour, and carrying out feature matching and image registration. The prostate nuclear magnetic ultrasonic image registration fusion method provided by the invention can adapt to imaging parameters of different equipment manufacturers, the image registration puncture efficiency is improved, and the problem of low manual segmentation registration efficiency caused by low imaging quality of the ultrasonic image can be effectively solved.

Description

Technical field [0001] The invention belongs to the field of computer-aided diagnosis, and specifically relates to a method for registration and fusion of prostate nuclear magnetic ultrasound images, which divides ultrasound images and registers ultrasound nuclear magnetic images. Background technique [0002] At present, the diagnosis of prostate cancer mainly relies on the prostate-specific antigen PSA (Prostate Specific Antigen, PSA), followed by needle biopsy. Due to the poor specificity of PSA, needle biopsy is clinically used as the gold standard for diagnosing prostate cancer. Transrectal ultrasound-guided prostate biopsy has become the most common clinical method for diagnosing prostate cancer due to its advantages of real-time, low cost, and easy operation. However, due to the low imaging quality of rectal ultrasound images, it is difficult to accurately locate the malignant tumor area from the image. The false negative rate of 6-needle needle biopsy of the prostate und...

AI Technical Summary

Problems solved by technology

[0005] Aiming at the low efficiency of manual segmentation and registration of ultrasound images due to low imaging quality, the present invention provides a method for automatic segmentation of prostate MR / TRUS images based on supervised learning, which performs non-rigid body registration with preoperative MRI images

Images