Design of prostate puncture system guided by 3D MRI and TRUS images

The incidence rate of prostate cancer in the world is increasing year by year. At present, the mainstream diagnostic method is puncture biopsy, the doctor holds an ultrasonic probe for imaging and manual puncture. There are many uncertain factors, such as ultrasound imaging effect, doctor's clinical level and psychological state. This paper designs a robot system including control system, image acquisition and processing system, probe feeding mechanism and mechanical arm. An automatic segmentation algorithm for prostate 3D MRI (Magnetic Resonance Imaging) and TRUS (Trans rectal Ultrasonography) images is proposed, which takes into account both real time and accuracy. A novel convolution module that integrates residual connection, dense connection and deep separable convolution is proposed for multi-scale feature extraction and fusion. At the same time, deep supervision mechanism and multiple attention mechanisms are integrated to improve training efficiency. The experimental results show that the real time and accuracy of the segmentation method proposed in this paper are better than the existing methods. The image-guided puncture experiment also shows that the puncture accuracy of the system designed in this paper is less than 1.5mm, The system and network designed in this paper have important clinical significance and practical value for image-guided biopsy.