Ping An Technology Unveils "Big Healthcare Ecosystem" at the First International Medical AI Conference
On September 22–23, 2017, Ping An Technology drew particular attention at the “2017 Inaugural International Conference on Medical Artificial Intelligence” held in Shanghai. At this prestigious event, which brought together more than 60 leading AI experts from China and abroad, prominent medical industry figures, government representatives, entrepreneurs, and investors, Ping An Technology, as a leader in the fintech sector, not only contributed unique perspectives but also showcased Ping An’s ambitions in the healthcare field, as well as its explorations and practical achievements in healthcare big data.
“Ping An Brain” Intelligent Engine Powers the Construction of a “Grand Medical and Health Ecosystem”
The application of artificial intelligence (AI) technology in the healthcare industry has long drawn significant attention from both the medical and technological communities. The persistent challenges in China’s healthcare sector—such as difficulties in accessing medical care, inconsistent service quality, and inefficient allocation of medical resources—not only provide numerous valuable scenarios for AI implementation but also test the technical capabilities of healthcare AI enterprises.
In his address in the interim report, Ma Mingzhe, Chairman of Ping An Group, stated: “Ping An will gradually transform from a capital-driven company into a technology-driven one.” In this process, Ping An has established the strategic goal of “AI + Health,” and leveraging its years of deep experience in the healthcare industry, it has built an integrated, end-to-end, online-to-offline (O2O) smart service platform as both a healthcare service provider and health insurance provider, enabling users to efficiently and comfortably access intelligent healthcare services.
Dr. Xu Liang of Ping An Technology Introduced Ping An’s “Intelligence + Healthcare” Big Data Analytics Method to Attendees
Behind the realization of intelligent healthcare services lies robust technological support. Ping An Technology has developed the “Ping An Brain” intelligent engine, which leverages Ping An’s big data in “intelligence + finance + healthcare” to establish four major application sectors: public health and disease prediction, medical fund management and risk control, intelligent precision medicine, and intelligent electronic health records. By employing data-driven analogs of traditional Chinese diagnostic methods—inspection, auscultation/olfaction, inquiry, and palpation—it significantly enhances the efficiency of medical services and successfully builds a comprehensive “healthcare ecosystem.”
In the public health and disease prediction sector, "Ping An Brain" has built various disease risk prediction models, as well as nationwide risk maps showing the spread of epidemics and trends in chronic and critical diseases. In the medical fund management and risk control sector, it employs intelligent methods such as data mining and machine learning to automatically identify high-risk medical fraud. In the intelligent precision medicine sector, "Ping An Brain" supports precision medicine and provides clinical decision support based on analysis of clinical, genetic, and lifestyle data. In the intelligent electronic health records sector, "Ping An Brain" achieves intelligent pre-diagnosis by constructing a medical and health knowledge base covering diseases, drugs, symptoms, treatments, and more.
In the future, these four major sectors will continue to expand and strengthen, with their data mutually complementary, truly enabling technology implementation in real-world scenarios and building an intelligent ecosystem for healthcare big data.
Ping An Technology's Explorations and Practices in Various Healthcare Fields
Currently, AI companies in China’s healthcare sector are largely constrained by technological capabilities and funding conditions, leading most to focus their efforts on a single medical specialty. Leveraging its leading advantages in business scenarios and technical expertise, Ping An Technology has explored and piloted applications across diverse medical fields, achieving significant results in disease prediction, medical imaging, and healthcare big data.
In collaboration with the Chongqing Municipal Health and Family Planning Commission, Ping An Technology developed the world’s first artificial intelligence (AI) and big data-based influenza prediction model. By comprehensively collecting influencing factors and constructing multi-dimensional models, Ping An Technology applied a series of machine learning algorithms from internet finance to disease prediction. At the macro level, the system mined historical data by integrating environmental and meteorological factors from hundreds of cities across China, along with regional lifestyle and behavioral patterns. At the micro level, it predicted disease risk by incorporating comprehensive, multi-dimensional individual predictive factors and information. At a deeper level, both macro and micro factors were integrated into various deep learning and AI methodologies.
Currently, the influenza prediction model can forecast outbreaks one week in advance. In addition, based on the current data landscape in Chongqing, Ping An Technology has jointly developed a chronic obstructive pulmonary disease (COPD) prediction model with the Chongqing Municipal Health Commission. Further research into predictive models for other chronic diseases will also be conducted to support policy formulation and epidemic prevention and control efforts.
Ping An Technology applies deep learning and image recognition technologies to the field of medical imaging, helping radiologists improve their efficiency in interpreting images. Compared with imaging interpretation products developed by other companies, Ping An Technology utilizes diverse data from hospitals at various levels during model training, enabling the system to be compatible with the majority of hospitals. Currently, the sensitivity of Ping An Technology’s pulmonary nodule detection system has exceeded 95%, and research on diseases such as gastric cancer and cervical cancer is progressing steadily.
In the field of medical insurance, phenomena such as medical fraud, collusive swiping of insurance cards, and resale of medications for profit occur from time to time. Ping An Technology leverages multi-source user healthcare behavior data, integrating temporal and spatial dimensions, to construct a healthcare utilization relationship network across multiple dimensions including patients, diseases, diagnosis and treatment, physicians, and hospitals. By employing machine learning and related algorithms, it identifies fraudulent activities and associated groups.
The “Similar Cases” service leverages semantic recognition technology to retrieve the most relevant medical cases and physicians’ responses based on symptoms input by users. This technology enhances the diagnostic and treatment capabilities of primary care physicians, enabling them to fulfill the roles of general practitioners. Furthermore, it serves end-users by reducing unnecessary medical visits.
In the future, the application of artificial intelligence (AI) in healthcare will become more extensive and profound. To enable AI to better serve clinical diagnosis and truly benefit the medical industry, it is essential to continuously refine the technology in real-world scenarios, thereby ensuring its practical value. Ping An Technology will adhere to the strategic direction of “technology-driven success, scenario-first,” activate the “Ping An Brain” intelligent engine, strengthen its technological foundation, meet practical needs, improve the healthcare ecosystem, and safeguard users’ healthy lives with professionalism and dedication.