When AI augments medical images, scientific research reaches beyond the laboratory, and the world’s top academic minds begin to share ideas, the results are expected to be far-reaching and profound. On March 5th, influential experts on the forefront of medical imaging gathered at Neusoft Medical Systems headquarters in Shenyang, China, to discuss the direction of evolving medical imaging technology. The peers from Germany, France, the United States, and China explored the application of AI software for more powerful imaging and brainstormed various ways to overcome pervasive difficulties in clinical diagnosis with CT and MRI.
Wu Shaojie, CEO of Neusoft Medical Systems welcomed the arrival of these influential experts, sharing a vision in which Neusoft Medical Systems strengthens its software and hardware advantages while fostering on-going dialogue with academic leaders. In relation to the prospects of the next generation of medical imaging technology, Wu Shaojie said, “We will fully integrate the platform, internal research team, and the resources of Northeastern University and Neusoft Group. We will also obtain in-depth understanding of clinical needs to make unremitting efforts to realize the dream of value creators in the medical industry.”
Among the foreign experts participating was Professor Jürgen Hennig, Senior expert in magnetic resonance, Scientific Director of the Department of Diagnostic Radiology, University of Freiburg, Germany, “Einstein Professor” of the Chinese Academy of Science, Former President of the International Society for Magnetic Resonance in Medicine. Professor Hennig expressed his hopes for the future of MRI development, saying, “The future of magnetic resonance is not merely in the field strength that is constantly improving; more importantly, it lies in the continuous improvement of scanning speed through technological innovation based on the existing 1.5T and 3.0T and the broadening of MRI’s clinical scope as well as accurate and quantitative clinical applications.”
French Professor Jean-Pierre Laissy, a top international expert in cardiovascular imaging and intervention, Professor of Department of Radiology, Lariboisière Hospital, Paris, France, and Secretary General of French Society for Cardiovascular Imaging Diagnosis and Intervention. “CT and MRI boast their own strengths and can complement each other. In the future, with the help of AI technology and other advanced post-processing technologies, CT enjoys a broader prospect for the application in the field of cardiac diagnosis,” observed Professor Laissy.
Professor Ron Schilling, from Stanford University contributed to the conversation saying, “With the boom of application of the advanced technologies such as AI and Big Data into the medical field, it can greatly reduce communication costs and ultimately benefit patients by showing users more intuitive and near-realistic results; the typical application scenario is that by providing a wealth of complementary information and precise implementations for surgery through mixed reality technology, it combine AI, big data and clinical perception to further enhance efficiency of surgeons.”
Leadership from within China expounded on the topic of AI included Professor Dai Jianping, the Honorary Chairman of China International Medical Foundation, Foreign Associates of the Institute of Medicine, and Honorary Member of RSNA and Professor Dong Di, Associate Research Fellow of Institute of Automation, Chinese Academy of Sciences, Head of Image Grouping, Key Laboratory of Molecular Imaging Chinese Academy of Sciences, and IEEE member. Professor Dong predicted that, “Future imaging-omics will develop into a comprehensive system platform integrating artificial intelligence methods, data resource platforms, auxiliary diagnostic systems and shared communication platforms; future doctors of imaging should be experts in imaging informatics. Doctors who know how to use of AI technology will definitely replace one who do not.”
Professor Zhou Shaohua, an international expert in medical image processing and Researcher at the Institute of Computing Technology in Chinese Academy of Sciences, Associate Editor of IEEE TMI and Medical Image Analysis, Area Chair of CVPR and MICCAI, Co-editor of The Vision Seeker, and Fellow of the American Institute for Medical and Biological Engineering, joined in the conversation saying, “Deep learning technology and medical imaging knowledge model can be combined in multiple stages such as data input, output and algorithm implementation. This combination can significantly improve the accuracy and stability of the machine learning model, while reducing the dependence on deep learning model on medical annotated data and thus effectively improve clinical efficiency.”
Renown Chinese expert in medical imaging, Professor Kong Dexing, was also in attendance. He is the Distinguished Visiting Professor of Zhejiang University, Director of the Image Processing and R&D Center of the Faculty of Science in Zhejiang University, Visiting Professor of the Chinese PLA General Hospital (Beijing 301 Hospital), and Member of the Expert Committee of China Medical Equipment Artificial Intelligence Alliance. Commenting on hopes for China’s healthcare sector, Professor Kong said, “By building a national medical image database, and promoting the advancement and implementation of artificial intelligence technology and its application in the medical field, the standard of diagnosis and treatment will be further improved and medical resources will be more balanced to effectively solve the knotty problem of difficult and expensive medical treatment.”
On that topic, Professor Cao Peng, an expert in computer-aided diagnosis who works for a key laboratory of the Ministry of Education for Intelligent Computing of Medical Imaging in Northeastern University, also added his thoughts. He is visiting scholar at the University of Alberta and has published his thesis on TKDD, Neurocomputing, Pattern Recognition, Neuroinformatics, MICCAI, PAKDD and other authoritative international key periodicals and conferences. “Machine learning methods and medical data experience full of difficulties and challenges in the process from their encountering to integration, such as high data dimensions, uneven distribution, multimodal, multi-domain data, lack of annotated information, etc. the combination of multiple machine learning and regularization methods can play a positive role in predicting Alzheimer,” said Professor Cao.
Professor DONG Di, as Associate Research Fellow of Institute of Automation, Chinese Academy of Sciences, Head of Image Grouping, Key Laboratory of Molecular Imaging Chinese Academy of Sciences, Member of the Youth Innovation Promotion Association of the Chinese Academy of Sciences, Member of the Radiology Professional Youth Committee of the Chinese Research Hospital Association, and IEEE member, said that “Future imaging-omics will develop into a comprehensive system platform integrating artificial intelligence methods, data resource platforms, auxiliary diagnostic systems and shared communication platforms; future doctors of imaging should be experts in imaging informatics: ‘Doctors who know how to use of AI technology will definitely replace one who do not’.”
Professor ZHOU Shaohua, as International Top Expert in Medical Image Processing, Researcher of Institute of Computing Technology in Chinese Academy of Sciences, Associate Editor of IEEE TMI and Medical Image Analysis, Area Chair of CVPR and MICCAI, Co-editor of The Vision Seeker, Fellow of the American Institute for Medical and Biological Engineering, said that “Deep learning technology and medical imaging knowledge model can be combined in multiple stages such as data input, output and algorithm implementation. This combination can significantly improve the accuracy and stability of the machine learning model, while reduce the dependence of deep learning model on a large number of medical annotated data and thus effectively improve clinical efficiency.”
Lastly, Professor LI Qince, an expert of cardiovascular system modeling, discussed the increasingly blurred boundaries between physics and the digital world and how digital technology will impact clinical treatment. “Comprehensive use the knowledge of physiology, pathology and pharmacology, modeling based on mathematical approaches, and the establishment of virtual physiological heart model, will be of great significance for studying the law of cardiac activity in both health and diseased states, improving the accuracy and efficiency of disease diagnosis, and accelerating drug research and development,” said the winner of the Young TopNotch Talents of Harbin Institute of Technology, winner of the British ORSAS, scientific researcher with the Natural Science Foundation China, and author to many papers published in leading international journals.
By engaging experts, Neusoft Medical Systems demonstrates its focus on global technology development and accelerates the movement of forward-looking technologies from laboratories into clinical practice to realize “beyond imaging” benefits such as making medical assistance available to more people around the world and ensuring faster and more accurate diagnosis to better guide patient treatment.