The annual meeting of the North American Society of Radiology (RSNA 2018) opened in Chicago on November 25. It is the largest radiology event in the world, attracting tens of thousands of Radiology experts, scholars and enterprise representatives from more than 130 countries around the world. As a medical imaging AI company committed to the global market, Huiyi Huiying, together with two full cycle AI products – “breast cancer artificial intelligence full cycle health management platform” and “aortist2.0 aortic artificial intelligence cloud platform”, once again appeared at the North American annual radiology conference, which was the first time that China’s ai+ medical full cycle products went to sea. Different from other AI products, these two products extend AI from pre diagnosis to the whole process of treatment decision-making and prognosis prediction, so as to realize more intelligent auxiliary treatment decision-making and more accurate prognosis prediction.

Patient centered AI medical full cycle products are favored

“Breast cancer AI full cycle health management platform” and “aortist2.0 aorta AI cloud platform” had previously made their debut at the 2018 China radiology academic conference, which was the first time that China’s AI medical full cycle products appeared on the international stage. Aortist2.0 aortic artificial intelligence cloud platform is an intelligent stent placement solution for type B aortic dissection. It was jointly released by the General Hospital of the people’s Liberation Army and Huiyi Huiying in Beijing, China in April this year. According to Guo Wei, director of vascular surgery of the General Hospital of the people’s Liberation Army, this is the first artificial intelligence automatic segmentation method for type B aortic dissection developed worldwide. Another “breast cancer artificial intelligence full cycle health management platform”, jointly developed by Huiyi Huiying and Intel, is committed to providing breast cancer artificial intelligence full cycle medical solutions, integrating breast molybdenum target, nuclear magnetic resonance, pathology and other data, quantifying the diagnosis results, and combining neoadjuvant chemotherapy to reduce the clinical stages of tumors and improve the breast preservation rate and the success rate of breast preservation surgery. In addition, combined with imaging omics, accurate prognosis prediction and personalized follow-up plan are realized.

Compared with ai1.0 medical imaging products that only carry out disease screening and auxiliary diagnosis, AI medical full cycle products have three obvious characteristics: Based on the fusion of multimodal data such as CT and MRI, as well as multidimensional data such as clinical data, pathological data and gene data, build a full image data chain to realize data fusion; By recommending personalized treatment decision-making schemes and optimizing prognosis, artificial intelligence can create greater value for doctors and patients and realize value integration; Image AI products are seamlessly integrated with doctors’ diagnosis and treatment processes to help doctors improve diagnosis efficiency and realize scene fusion.

Liu Shiyuan, chairman designate of Radiology branch of Chinese Medical Association, chairman of China Medical Imaging AI industry university research innovation alliance and director of imaging medicine and nuclear medicine discipline of Shanghai Changzheng Hospital, once said that Huiyi Huiying AI’s products and concepts of full cycle health management are very good. At present, image AI has the following problems. AI can only solve the problem of single point, but can not solve all the problems of medical image; Light to solve the problem of image can not solve all the problems of medicine. Therefore, only the whole chain of imaging AI and medical AI solutions can finally solve the problems of doctors and patients.

“Some operations performed by many top hospitals are restorative operations, and the proportion of postoperative recurrence is high. For such complex diseases, discharge is not the end point. Prognosis prediction and discharge follow-up are a new starting point for medical treatment. Our goal is to design a patient-centered product to cover the whole medical cycle of patients. In addition to improving the surgical accuracy of doctors, aortist system also integrates the imaging cloud platform developed by us Enter the prognosis prediction model to predict whether adverse events will occur in patients with type B dissection after operation. ” Chai Xiangfei, founder and CEO of Huiyi Huiying, said.

Full data platform drives AI into the whole medical cycle

Since 1898, in the field of medical imaging, we have experienced a physical driven era represented by X-ray, ultrasound and nuclear magnetic resonance, and an application driven era represented by image guidance and treatment planning. After 2010, we have entered a data-driven era of intelligent medicine, which is typically characterized by mining effective information in massive data and optimizing diagnosis and treatment methods.

In the era of data-driven smart medicine, data is the key. In the medical field, medical big data is special. Medical big data is not “big”. Even the image data is very limited. Especially for a single disease, everyone can’t take a film a year on average, such as interstitial pneumonia or a fracture in a certain part. There may be only tens of thousands of patients in the country every year, and they are scattered in various hospitals, so it is very difficult to obtain the data. In addition, the data collection standards among hospitals are not unified, including a large amount of unstructured data.

The development of AI depends on the development of data. The data center integrating image data, clinical data, test data, pathological data and even genetic data will become the password to open the health of patients. At present, Huiyi Huiying has built a patient-centered full data platform to provide a full stack of medical AI solutions from diagnosis and treatment to scientific research. The whole process of artificial intelligence is covered in a single disease, including the whole disease cycle from intelligent screening to intelligent decision-making and prognosis prediction. Combined with the full data of patients, doctors can be given effective help in many links of diagnosis and treatment, such as tumor staging, adaptive radiotherapy, prognosis follow-up and so on.

Chief scientist of Huiyi Huiying Dr. Xing Lei, a lifelong professor at Stanford University, believes that: “Ai plays a greater role in medical treatment. It also needs to build a multi-dimensional database. Using AI to integrate multi-dimensional data such as imaging, genetics, pathology and clinic can provide personalized medical plans for patients, recommend surgical plans for clinicians and provide medication guidance. Using AI can provide patients with reasonable inspection, treatment, follow-up and rehabilitation plans, provide intelligent monitoring and management of the whole course of disease, optimize diagnosis and treatment processes and save diagnosis and treatment costs Expenses. “

“Our big data intelligent analysis system based on icomics brings together different topics from more than 500 hospitals. At present, we have a large number of fine labeled image data, and combined with a large amount of clinical data, test data and pathological data to build a full-scale data center, which is the basis of realizing full-cycle artificial intelligence medical treatment.” Chai Xiangfei said.

Huiyi Huiying’s patient-centered and data-driven product logic coincides with the spirit of this RSNA conference. In his keynote speech on “how emerging technology will empower tomorrow’s radiologists to provide better patient care”, Vijay M. Rao, President of the 2018 RSNA conference, said that in the future, methods such as AI and machine learning will be of great benefit to the radiology department. These technologies will make the work of radiologists more efficient and transfer more time to the care of patients. Technological innovation will make image technology develop in the direction of “fast, safe, quantitative, precise and affordable”. The information center established with important image data, clinical information, genetic characteristics and risk factors will provide great help for patients to implement individualized treatment.

In addition, Huiyi Huiying also launched a new product on RSNA: tuberculosis intelligent screening. This product can screen on X-ray to detect whether the patient has tuberculosis, and quantify the texture, position, shape and other characteristics of tuberculosis in combination with CT analysis to diagnose, assess the risk and assist doctors in treatment decision-making.

Previously, Huiyi Huiying signed a business cooperation agreement with nuance, the world’s largest company specializing in the R & D and sales of speech recognition software and the supporter of Apple Siri product speech recognition technology. Huiyi Huiying’s four mature AI algorithms have successfully landed on nuance’s AI algorithm store platform. During the RSNA conference, the two sides jointly participated in the exhibition on the cooperation results.

At present, Huiyi Huiying AI products include biased screening products such as pulmonary nodule screening, fracture screening, chest X-ray screening and pulmonary tuberculosis screening, as well as two full cycle products. They have been implemented in more than 800 hospitals of different sizes in China and have been applied in more than 50 hospitals such as Japan, the United States, Kazakhstan and Brazil.

Guo Na, co-founder and chief operating officer of Huiyi Huiying, said: “the application of multimodal image data and the full cycle coverage of disease diagnosis and treatment are the development trend of the industry. In the future, more diseases will realize the whole process AI application. Man machine cooperation will become the normal work of doctors, and computers will become the most powerful assistant of doctors.”

Leave a Reply

Your email address will not be published.