On April 27, the "Frontier Achievements in the Field of Medical Artificial Intelligence" was released at the 2024 Zhongguancun Forum Annual Conference Artificial Intelligence Theme Day Medical AI Industry Integration and Development Conference.
At the meeting, Wan Suiren, vice president of the Chinese Society of Biomedical Engineering and professor of Southeast University, released the "Nine Frontier Achievements in the Field of Medical Artificial Intelligence in 2023".
"In the field of medicine, artificial intelligence technology has injected new momentum into the new quality productivity of empowering the development of medical health, promoting the development of disciplines, accelerating the construction of smart medical hospitals, and improving clinical diagnosis and treatment services. We must not only make good use of 'artificial intelligence + medical care' to improve people's health and happiness, but also explore artificial 'intelligence + scenarios' and sneak into the deep sea of medical artificial intelligence, and at the same time manage 'artificial intelligence + ethics' and adhere to people-oriented, intelligent and good. Hu Shengshou, academician of the Chinese Academy of Engineering and president of Fuwai Hospital of the Chinese Academy of Medical Sciences, said that accurately grasping the frontier and development trend of artificial intelligence in the medical field is an effective way to help the high-quality development of the industry. The "Nine Frontier Achievements in the Field of Medical Artificial Intelligence in 2023" are as follows:
1. Ultra-large-scale cross-species and three-dimensional morphological database and diversity research of human brain neurons
Applicant: Institute of Brain Science and Intelligent Technology, Southeast University
It has developed a set of high-precision digital reconstruction platform and analysis technology, established the world's largest international cooperation network for neuronal structure research, and launched the multi-species benchmarking platform BigNeuron and the corresponding open-source database. The findings were published in Nature Methods and Science in 2023.
In addition, the research team also innovatively proposed ACTomography (adaptive single-cell tomography) technology. Promote international cooperation in the field of brain science, and provide an important theoretical and practical foundation for future brain science research.
2. Ultrasound auxiliary diagnosis software for thyroid nodules
Reporting unit: Zhejiang Deshang Yunxing Medical Technology Co., Ltd
In view of the non-standard and real-time characteristics of ultrasound images, Deshang Yunxing has independently developed the artificial intelligence technology platform DE-Light, which solves the core technical problems of ultrasound AI from the underlying technology and improves the accuracy, speed and robustness of the system. Based on the DE-Light platform, the AI-SONIC system developed by Deshang Yunxing is industry-leading, which can automatically detect lesions, automatically assist in diagnosis, and automatically generate structured reports.
The "Thyroid Nodule Ultrasound Auxiliary Diagnosis Software" developed by the company has an automatic classification accuracy rate of 93.34%, and in November 2023, it obtained the country's first ultrasound AI Class III medical device registration certificate, which has taken an important step in the market access and commercialization process of ultrasound AI products.
3. The first domestic unicompartmental joint replacement robot
Reporting unit: Beijing Natong Intelligent Technology Co., Ltd
Unicompartmental joint replacement robot is a domestic auxiliary orthopedic surgery system, which has obtained the Class III registration certificate of the National Medical Products Administration (NMPA), filling the gap in the field of domestic robot unicompartmental replacement, and its key technologies have applied for 20 patents and 4 software copyrights.
The robot realizes real-time detection, active tracking and positioning and real-time kinematic analysis through intelligent vision algorithms, and realizes automatic preoperative 360° three-dimensional planning, digital modeling, automatic matching of the optimal prosthesis model and type, as well as the placement position of the prosthesis through intelligent algorithms, and also realizes accurate osteotomy, all-round protection and effective cutting technology of complex bone surface with sub-millimeter accuracy through intelligent operation control algorithms. The system can make the processing capacity of the patient's bone data reach the second level, the accuracy rate reaches more than 90%, and the accuracy rate of its automatic planning algorithm also exceeds 90%, reaching the international advanced level.
4. Joint replacement surgery simulation software
Reporting unit: Beijing Changmugu Medical Technology Co., Ltd
The "Joint Replacement Surgery Simulation Software" developed by the company is China's first artificial intelligence-assisted adult hip and total knee replacement surgery, which can realize accurate segmentation and three-dimensional reconstruction of different types of joint diseases based on the self-established CT image database of various orthopedic diseases and deep learning neural network models, thereby effectively reducing the error of preoperative measurement. AI 3D simulation software can achieve high-precision preoperative planning, thereby improving the prosthesis matching rate, and has great potential for clinical application.
The software has obtained the registration certificate of the National Medical Products Administration (NMPA). Since the establishment of the software project, 5 articles have been published, 56 patents have been applied, including 55 invention patents and 1 design patent, as well as 5 software copyrights and 9 PCT international patents. These results mark a major breakthrough in the clinical application of artificial intelligence in the field of orthopedic surgery. It has been applied in 598 tertiary hospitals and more than 1,000 county-level hospitals in 31 provinces, municipalities and autonomous regions in China, serving 159,000 patients.
5. "Non-contrast CT+AI" may change the guidelines for pancreatic cancer screening and diagnosis
Reporting unit: Shanghai Changhai Hospital
An innovative study conducted by a research team from Shanghai Changhai Hospital has shown that the "non-contrast CT+AI" technology has great potential in the early screening of large-scale pancreatic cancer. The research results have been published in the internationally renowned medical journal Nature Medicine.
The research team has developed a deep learning framework called PANDA for the detection of pancreatic cancer. Based on the actual clinical needs, this study not only improved the success rate of severe acute pancreatitis, but also improved the postoperative quality of life of patients with chronic pancreatitis. The PANDA model was proposed to realize the early screening of pancreatic cancer on a large scale for the first time with "non-scan CT combined with artificial intelligence".
6. The similarity between the AI speech model and the human brain's auditory pathway
Applicant: School of Biomedical Engineering, ShanghaiTech University
A research team from the School of Biomedical Engineering at ShanghaiTech University, together with collaborators at the University of California, San Francisco, and Fudan University, has conducted an innovative study that explores the similarities between artificial intelligence (AI) speech models and auditory pathways in the human brain in terms of neural computation and representation.
The study reveals significant similarities between the hierarchical structure of the end-to-end speech pre-training network and the hierarchical structure of the auditory circuit. The results have been published in important academic journals such as Nature Neuroscience, providing an important scientific basis for understanding the connection between AI speech models and the human brain.
7. Individualized and dynamic decision-making of tumor clinical diagnosis and treatment based on ultrasound AI
Reporting unit: Institute of Automation, Chinese Academy of Sciences, Key Laboratory of Molecular Imaging, Chinese Academy of Sciences
Through the deep integration of ultrasound real-time medical imaging and artificial intelligence technology, new analytical methods and technologies have been developed, which have significantly improved the accuracy of tumor diagnosis and the ability to predict treatment effects. The team's research results have been published in several international journals, including Medical Image Analysis, IEEE TMI, etc., and have published articles in important journals in the medical field such as Gut, Liver Cancer, BMC Medicine, EJNMMI, and European Radiology.
The team members have published more than 30 academic papers in the past five years and won the first prize of the 2023 Huang Jiasi Biomedical Engineering Award of the Chinese Society of Biomedical Engineering.
8. Comprehensive service platform for medical knowledge
Applicant: Institute of Medical Information, Chinese Academy of Medical Sciences
The "Medical Knowledge Integrated Service Platform" developed by the company is a national-level innovation project that aims to build a knowledge system for medical artificial intelligence services. As part of the National Science and Technology Innovation 2030 New Generation Artificial Intelligence Major Project, the platform has passed the project acceptance in December 2023 and has been included in the Achievement Transformation Catalog of the High Technology Center.
The platform not only builds the first million-scale terminology system of traditional Chinese and Western medicine concept relationship in China, but also explores the full-chain service model from knowledge base to application scenarios, providing effective empowerment for the intelligent development of medical and health care. 9. Application unit for the construction of a medical artificial intelligence standard system with Chinese characteristics: Puyang Institute of Big Data and Artificial Intelligence has made significant progress in the formulation of standards in the field of medical artificial intelligence in the mainland, and has initially established a multi-level standard system in the field of artificial intelligence medical devices in the mainland. Since 2019, the State Food and Drug Administration and other institutions have actively promoted the research on standards and specifications for safety and efficacy, which has greatly promoted the development of the industry. As of 2023, the release and formulation of industry standards for artificial intelligence medical devices have been steadily promoted, and basic standards such as "YY/T 1833.4-2023" and product testing standards such as "YY/T 1907-2023" have been released. Breakthroughs have been made in international standards, led by the China Institute for Food and Drug Control, the first international standard for the testing of artificial intelligence medical device products was organized in the IEC organization, and standards such as IEEE2801 and 2802 were released. These achievements in the construction of standards show that the mainland has reached an international leading position in this field, which is of great significance for promoting technological progress and protecting public health. Source | Biomedical Alliance
Join the community
Dots to share
Point Collection