On July 5, Peking Union Medical College Hospital and Tencent AI Lab jointly released a portable and intelligent surgical navigation system with completely independent intellectual property rights. The initial clinical application was successful.
The system consists of a navigation tablet, a navigation stick, a depth camera and other equipment. It is small and portable and does not occupy the space of the operating room. It is a neurosurgical navigation system that can be located on the head frame. It has the characteristics of high precision, portability, and low cost. It is like adding GPS to clinicians’ operations to assist doctors to accurately locate lesions hidden in brain tissue.
At present, more than 30 clinical trials of this product have been carried out in the Department of Neurosurgery of Peking Union Medical College Hospital, involving cerebral hemorrhage, pituitary adenoma, glioma, meningioma, craniopharyngioma and other diseases.
Intraoperative usage scenarios & core components of the system
Professor Wang Renzhi, director of the Peking Union Medical College Neurosurgery Teaching and Research Office and doctoral supervisor, said: “During surgery, neurosurgeons not only have high requirements for the accuracy of lesion location, but also need to understand the relationship between the lesion and surrounding blood vessels and nerves, so as to avoid damage to these important organs during surgery. Organization. In the past, we only re-integrated the patient’s CT or MRI two-dimensional images into the brain to form three-dimensional images based on experience. In fact, this learning process is very long and requires a lot of long-term experience accumulation. We and Tencent The jointly developed surgical navigation is simple, convenient, and easy to use, and can help clinicians solve many practical problems.”
In my country, the utilization rate of traditional surgical navigation is very low. This is because: on the one hand, the core technology and components of traditional surgical navigation systems rely on imports, and it is difficult for this technology to truly benefit primary hospitals at a cost of millions. On the one hand, traditional surgical navigation equipment is bulky, has a high threshold for learning and operation, and takes a long time to use. It is difficult to really apply it in practical scenarios such as emergency.
In order to fundamentally solve these problems and make it easier for patients to receive more accurate surgical treatment at home, the R&D team has carried out a subversive and innovative design of surgical navigation for three years, which not only greatly reduces the cost, but also saves the preparation time. Compared with the products, they have the same degree of accuracy, and have completed the localized replacement of the surgical navigation system. Under the premise of high precision, the new type of navigation reduces its cost to one-tenth of that of traditional surgical navigation products, filling the gap in the domestic market.
Before surgery, based on AI and imaging technology, the surgical navigation system can automatically segment and reconstruct MRI, CT and other images and perform 3D visualization processing. Through the algorithm capabilities of Tencent AI Lab, a customized 3D “transparent brain” can be generated with one click. It is convenient for doctors to formulate surgical plans and explain the surgical plan to patients and their families.
Preoperative “clear brain” generation
The entire system integrates optical positioning and SLAM tracking. Through optical positioning, the surgical navigation system can track surgical instruments in real time; through real-time positioning technology (SLAM), the system can observe lesions from multiple angles without the need for a head frame and a reference frame. Location.
Intraoperatively, SLAM tracking enables augmented reality navigation without the need for a head frame and reference frame. Taking a glioma patient as an example, the doctor moves the camera, and the system can display the positional relationship of the three-dimensional structure in the patient’s skull relative to the scalp in real time, assisting the doctor to quickly make an incision.
Preoperative incision making
With the help of a depth camera, combined with deep learning and visual algorithms, the system can achieve high-precision one-click registration between virtual space and real space, and match the generated 3D “transparent brain” with the actual situation of the patient 1:1, and perform intraoperative navigation. .
Intraoperative Navigation
Test data show that the registration accuracy can reach sub-millimeter level. The improvement of algorithm ability can shorten the registration time of traditional surgical navigation from the traditional nearly half an hour to a few seconds, and seize precious treatment opportunities for patients.
In practical application scenarios, the system can also observe the brain from multiple perspectives by matching with AR glasses according to the needs of doctors, and eliminate blind spots in the field of vision.
In the future, in addition to neurosurgery, ENT, and emergency departments can all use surgical navigation systems. The system can also be used in classroom teaching to improve primary medical service capabilities.
Tencent has a number of layouts in the fields of medical and life sciences. Leifeng.com also previously reported that “Zhong Nanshan’s team worked with Tencent to develop AI prediction for severe new crowns, and the results were published in the “Nature” sub-issue. When the outbreak broke out in 2020, Tencent AI Lab and Zhong Nanshan’s team jointly released a prediction using AI. The research results of the probability of COVID-19 patients progressing to critical condition were published in the sub-journal of “Nature”.
With the help of the cutting-edge algorithm capabilities of Tencent AI Lab and the clinical accumulation of neurosurgery of Peking Union Medical College Hospital for many years, the new domestic surgical navigation system released this time will assist doctors in improving the efficiency of diagnosis and treatment in preoperative and intraoperative links.
This article is reproduced from: https://www.leiphone.com/category/aihealth/jK2hHIaCahYx9Pg8.html
This site is for inclusion only, and the copyright belongs to the original author.