Artificial intelligence technology-based pituitary tumor operation planning system

Abstract

The invention discloses a pituitary tumor operation planning system based on an artificial intelligence technology. The system comprises the following steps: 1, acquiring a vascular medical image sequence, a skeleton medical image sequence and a pituitary tumor medical image sequence; 2, if the medical image sequences are the image sequences shot at different moments or the image sequences shot at the same moment are not in the same coordinate system, performing registration, and placing the image sequences in the same coordinate system; 3, using a deep neural network model to take each medical image sequence in the same coordinate system as input, and outputting a blood vessel image segmentation result, a skeleton image segmentation result and a pituitary tumor image segmentation result; and 4, performing three-dimensional reconstruction on the blood vessel image segmentation result, the skeleton image segmentation result and the pituitary tumor image segmentation result to obtain a three-dimensional model for pituitary tumor preoperative planning. The method is high in processing speed and high in reliability, so that preoperative evaluation and surgical planning of clinicians become simple, accurate and rapid.

Description

technical field [0001] The invention relates to the fields of biomedical engineering and medical images, in particular to a pituitary tumor surgery planning system based on artificial intelligence technology. Background technique [0002] In the past 20 years, neuroendoscopy technology has developed rapidly and its application scope has been continuously expanded. Among them, the development of skull base endoscopy is particularly rapid. The complex anatomy of the skull base is one of the most challenging areas of neurosurgery. Neuroendoscopy has been widely used in the treatment of various lesions of the skull base due to its wide field of view, clear imaging, and small surgical trauma. [0003] In pituitary aneurysm resection endoscopic skull base surgery, due to the complexity of structures such as optic nerve, cavernous sinus, petrosal bone segment, and cavernous sinus internal carotid artery, the operation is difficult. Injury to these structures during surgery may ca...

AI Technical Summary

Problems solved by technology

Patent CN202010907454 proposes an artificial intelligence-based method and device for positioning and diagnosing pituitary adenomas in magnetic resonance images. The method of tumor location, this patent can assist clinicians and radiologists in rapid diagnosis, however, the selection of pituitary adenomas cannot be reconstructed in 3D, and the 3D reconstruction of pituitary tumors needs to segment the pituitary tumor area pixel by pixel

At the same time, this method does not propose methods for segmenting blood vessels, nerves, etc., so it cannot provide support for surgical planning, simulation and navigation

[0007] In the article [Wang Lei, Wang Hongfei, Wang Yan, et al. The application progress of artificial intelligence in the diagnosis, treatment and prognosis prediction of pituitary adenoma [J]. Journal of Clinical Neurosurgery, 2021, 18(5):4.] mentioned the application of The application of artificial intelligence technology and surgical robotic arm to complete the precise resection of pituitary tumors does not mention what technology is used to ensure the safety of surgery and the accuracy of resection

[0008] Therefore, in order to solve the problems of long preoperative planning time and difficult image segmentation, the present invention proposes for the first time a fully automatic surgical planning system based on artificial intelligence technology, which assists clinicians to quickly complete surgical planning and evaluation, and can use the system The generated 3D model is used as the input of the surgical navigation system to ensure the safe implementation of the operation

Images