Can't understand the inspection report? AI empowers smart healthcare to make medical treatment more convenient for the public
2023-09-28
With the rapid development of technology, how can we provide patients with more accurate, efficient, and convenient medical services in the medical industry? On the 26th, the Runda Medical Big Model Good Doctor Xiaohui Product Launch, co hosted by Runda Medical and Huawei Cloud, was held in Shanghai. The first AI vertical big model in the medical industry, "Good Doctor Xiaohui," was released, which will accelerate the construction of smart healthcare and help the high-quality development of the medical industry. The result of the red blood cell count test is 4.44, and the platelet distribution width is 13.5... In the medical industry, test reports are one of the important basis for doctors to diagnose and treat. However, for patients, the numbers and indicators in these test reports are often difficult to understand, and it is also easy to feel confused and uneasy. For doctors, the frequent submission of dozens or hundreds of test reports every day greatly increases their workload. In order to enable doctors to make good use of tests and help patients understand and alleviate anxiety, with the support of the Huawei Cloud Pangu big model, Runda Medical has developed the medical big model "Good Doctor Xiaohui" by combining the inspection knowledge graph of Huijian and Huawei Cloud's artificial intelligence technology, and constructed the largest medical inspection knowledge graph in the world. After training up to 1 billion times, the good doctor Xiaohui can explain over 4500 test items and 2800 diseases, with a comprehensive accuracy of 87.74%. Zhang Nan, CTO of Runda Medical, stated that in the comparison process with industry-leading excellent large model tools, Liangyi Xiaohui has shown significant advantages in patient report analysis, comparison, interpretation, and other aspects. It can comprehensively consider the patient's medical history, symptoms, and other related factors, providing a more comprehensive and detailed assessment for patients, and thus providing more accurate diagnosis and treatment plans. At the meeting, You Peng, the Minister of Huawei Cloud EI Service Product Department, stated that Huawei Cloud is fully leveraging its technological accumulation and engineering capabilities in the field of large models, allowing AI to reshape its business. Focusing on the medical field and addressing issues such as long drug research and development cycles, high manual experimental costs, and long time consumption, we have developed the Pangu Drug Molecular Large Model, which allows the large model to learn small molecule compounds like humans, fully understand the molecular structure, and ultimately improve the accuracy of drug prediction by 20%. In practice, based on the Pangu drug molecular model, we collaborated with the First Affiliated Hospital of Xi'an Jiaotong University to develop antibacterial drugs. The lead drug development cycle was shortened from one year to one month, greatly improving the drug development efficiency. At present, Huawei Cloud is based on the computing power base of the Shengteng AI cloud service, providing developers with comprehensive tools and resources to support the efficient migration of large models, ensure the surge of computing power supply for model training, and ensure the stability and reliability of the environment through native incubation and adaptation to mainstream industry models. In addition, Huawei Cloud is also overflowing the capabilities of the Pangu large model. The formation of Xiaohui, a good doctor, is precisely based on the Huawei Cloud Pan Gu big model, and Runda Medical has established its own specialized data, ultimately creating its own application model of L1 model. In the future, with the update and upgrade of Huawei's cloud pangu big model technology, the specialized disease models included in Good Doctor Xiaohui will also be more abundant, and will be promoted and applied from the laboratory to more hospitals. Its powerful interpretation ability and accuracy can help doctors better understand and analyze
Edit：GuoGuo Responsible editor：FangZhiYou
Source：gmw.cn
Special statement: if the pictures and texts reproduced or quoted on this site infringe your legitimate rights and interests, please contact this site, and this site will correct and delete them in time. For copyright issues and website cooperation, please contact through outlook new era email：lwxsd@liaowanghn.com
Recommended Reading Change it 
Home
