Research And Application Of Medical Image Intelligent Analysis System Based On VR

Computer aided diagnosis technology is through medical image processing technology,combined with computer analysis and calculation,auxiliary detection of lesions.With the development of computer science and technology and the expansion of application fields,computer aided diagnosis technology,which combines computer and medical images,has become research hotspots.As an important medium of transmitting the main ideas,displaying the subjective information,expressing the rich emotion and recording the effective information,images has a certain preservation meaning and research value.In the field of human medicine,taking medical images is an important way to obtain the information of the patient’s life and signs directly.It is of great significance to the field of medical engineering to study the computer aided technology of combining medical images with the aid of automated testing to reduce the workload of doctors and even improve the diagnostic rate.This thesis combines medical images with VR technology,applies thyroid twodimensional DICOM images to VR environment after processing,and then performs corresponding operations and analysis,thus designing and implementing a medical image intelligent analysis system.Through the intelligent medical image analysis system,the doctor’s workload can be reduced or the diagnostic accuracy can be improved.The main research results are as follows:First of all,aiming at the specific characteristics of two-dimensional DICOM images of thyroid,i studied the active contour model segmentation technology based on level set.The C-V model was used to segment the thyroid successfully for following identification.Next,the thyroid image that had been segmented in the previous step was subjected to feature extraction,and then by using the SVM classifier,the thyroid image was classified as the normal image and the image of lesions.Based on this,a GA-SVM classifier based on genetic algorithm was proposed to optimize the relevant parameters in the SVM algorithm thus improving the classification accuracy.Finally,i studied the medical image 3D reconstruction technology based on VTK.The three-dimensional images of thyroid were successfully drawn by using surface rendering and volume rendering techniques.At the same time,i studied the HTC vive virtual reality device.The 3D thyroid model was imported into the VR environment by using the Unity3 D engine and the SteamVR plug-in,and related VR interaction operations were performed.