Super fungus is coming, academicians develop AI to help rapid detection

Forum site

Yesterday (April 28), the 2019 China Integrative Medicine Conference was held in Xi'an. Reporters learned from the conference that Academician Liao Wanqing from the Shanghai Key Laboratory of Molecular Biology of Medical Fungi and Yiku Cloud have successfully created an AI (artificial intelligence) for fungal identification. This will help doctors identify pathogenic medical fungi and enable auxiliary diagnosis of medical fungal diseases wherever there is a mobile phone or network signal.

High misdiagnosis rate of fungal infection and lack of talent are the shortcomings of treatment

Data show that there are about 2 million species of fungi in nature, and about 560 species of human pathogenic fungi. The annual cost of fungal diagnosis and treatment is about 2.6 billion US dollars. Superficial fungal diseases are widespread, mainly invading the skin, hair, and nails, with a prevalence rate of about 47.6%. Deep fungi are serious, mainly invading internal organs such as the heart, liver, spleen, lungs, kidneys, and brain, with a mortality rate of more than 30% to 90%.

Every year, tens of millions of people are infected with fungi in the world. Deep fungal infections caused by pathogens cause 1.5 million deaths each year, posing a serious threat to human health. How to quickly and accurately identify fungal infections, especially fatal ones, is a difficult problem that needs to be solved in the world of medicine.

Academician Liao Wanqing delivered a keynote speech

Academician Liao Wanqing revealed that there is a severe shortage of talents in medical fungus identification in my country. "It takes many years to train a physician who can identify fungi. In addition, the misdiagnosis rate of fungi is high. In the past, doctors relied on experience when making diagnoses, and it was difficult to achieve precision medicine."

Liao Wanqing said that currently, among the 16,000 medical institutions below the second-level hospitals, there is basically no professional fungus identification instrument. "What is more worrying is that the level of primary medical care in China is relatively low, and it is difficult to identify the invisible killers caused by deep fungi. Many patients lose their lives because they miss the best time for treatment."

Artificial intelligence has the ability to identify fungi in 30% less time

How to enable doctors in primary hospitals to have a pair of “eagle eyes” to identify medical fungi and accurately diagnose and treat the invisible killers caused by deep-seated fungi.

Professor Zhang Qunhua, CEO of Yikuyun

Professor Zhang Qunhua, CEO of Yikuyun, said that Academician Liao Wanqing has been a doctor for more than 50 years and has rich clinical experience. He is also a world leader in the field of medical fungi research. "The medical fungi library he has built over the years has more than 400 strains of medical fungi, which is the largest medical fungi library in Asia. Under the guidance of Academician Liao Wanqing's team, Yikuyun AI engineers have conducted AI research and development on dozens of medical fungi, especially the identification of deadly fungi, with an accuracy rate of 95%."

Simply put, it takes AI and big data as the core, uses computer vision, natural language processing, and fungal knowledge base to provide the optimal algorithm for medical fungal image recognition, so that AI identification can be performed before the colony grows into a typical form, reducing the identification time by one third.

Zhang Qunhua said, "AI will play an important role in colony identification, fungal mass spectrometry database construction, and electronic medical records of deep fungal infections." He revealed that when grassroots hospitals are connected to the system in the future, it will be equivalent to introducing a senior infectious disease doctor and tester. Patients do not have to go to crowded hospitals in big cities for medical treatment. They can accurately diagnose deep fungal infections and receive effective treatment in county-level hospitals at their doorsteps. This will largely solve the problem of difficulty in seeing a doctor, and patients will become the biggest beneficiaries.

Fungal disease diagnosis wherever there is a signal

Academician Liao Wanqing also proposed the exploration of integrative medicine to prevent and control fungal infections along the Maritime Silk Road at the Integrative Medicine Conference. Comparing the distribution of super fungi with the countries along the strategic “Maritime Silk Road” in my country, he believes that there is a high degree of overlap.

Dr. Liao Wanqing believes that in addition to the integration of multidisciplinary clinical forces of Chinese and Western medicine, the integration of artificial intelligence technology and medical mycology to achieve a digital Silk Road for the prevention and control of fungal diseases are all important contributions to the development of human medicine.

Academician Liao Wanqing emphasized that medicine has no borders, and the promotion and application of medical fungal detection products developed by medical artificial intelligence in the prevention and control of medical fungal infection diagnosis and treatment along the Silk Road is expected to reduce the high risk caused by medical fungal infection along the Silk Road. "As long as there is a mobile phone or network signal, the function of fungal disease diagnosis can be realized. It can provide Chinese experience and wisdom for the prevention and control of medical fungi along the Belt and Road."

It is reported that Yikuyun has nearly a thousand top cross-border doctors and AI experts from China and the United States. They embrace the Internet + AI together and strive to create novel, practical and inclusive AI products. Zhang Qunhua introduced that Yikuyun has natural language understanding, medical knowledge graph, medical imaging, Bayesian network, medical thrust engine, clinical assistance and decision-making diagnosis engine. In addition to discussing the research and development plan of Parkinson's disease AI with the Department of Neurology of Huashan Hospital, it has also established a national liver cancer AI research and development platform with West China Hospital. Together with Cai Tongdetang Pharmaceutical, it has developed the world's first composite AI Chinese medicine identification product. Professor Zhang Qunhua emphasized that the formation of Yikuyun's unique medical AI research and development innovation ecosystem led by cross-border doctors, AI engineers as the main body, and patients as the center is the future development direction of medical AI.