echemi logo
Product
  • Product
  • Supplier
  • Inquiry
    Home > Biochemistry News > Biotechnology News > 【Research News】Professor Ning Kang's team at the School of Life Sciences, Huazhong University of Science and Technology uses transfer learning to overcome regional effects and achieve ...

    【Research News】Professor Ning Kang's team at the School of Life Sciences, Huazhong University of Science and Technology uses transfer learning to overcome regional effects and achieve ...

    • Last Update: 2022-11-15
    • Source: Internet
    • Author: User
    Search more information of high quality chemicals, good prices and reliable suppliers, visit www.echemi.com
      

    On October 28, 2022, Professor Ning Kang's team from the Department of Systems Biology and Bioinformatics, School of Life Sciences, Huazhong University of Science and Technology, published a title entitled "Overcoming regional limitations: Transfer learning for cross-regional" in GUT, the only international top journal of gastroenterology and hepatology microbial-based diagnosis of diseases", proposing the use of transfer learning to overcome regional effects and achieve cross-regional disease diagnosis
    based on microbial characteristics.
    Wang Nan, a doctoral student at Huazhong University of Science and Technology, is the first author of the paper, Cheng Mingyue is the second author, and Professor Ning Kang is the corresponding author
    of the paper.

    Due to geographical location, diet, race, disease and other factors, especially the influence of geographical factors, the intestinal microorganisms of people in different regions have very obvious heterogeneity
    .
    At present, machine learning methods based on microbial characteristics have been used in the diagnosis of different types of diseases, but due to
    the inability to overcome regional effects, the accuracy of these methods in cross-regional disease diagnosis will be greatly reduced, and they cannot meet the needs
    of cross-regional disease diagnosis based on microorganisms.

     

    Figure 1.
    Microbiome big data analysis framework
    combining transfer learning and neural networks.

     

    In view of the above problems, researchers propose a framework that integrates transfer learning and neural networks, which can "borrow" the mature disease diagnosis knowledge of one region for disease diagnosis in another region, so as to overcome regional effects and realize cross-regional disease diagnosis
    based on microorganisms.

    In this work, the researchers applied this framework to 6,998 fecal microbiome samples from the Guangdong Gut Microbiome Project (GGMP), which were divided into 14 different prefecture-level cities (districts)
    according to their source.
    The results show that the transfer learning model has the most advantages in the accuracy of cross-prefecture-level city disease diagnosis compared with the ab novo training model, and the transfer learning model still has superior accuracy
    when applied to the disease diagnosis across intercontinental cohorts.
    In addition, the researchers have discovered strains that are more affected by regional factors, such as
    Clostridium, through transfer learning, which may potentially contribute
    to the effectiveness of transfer learning models in cross-regional disease diagnosis.

     

    Figure 2.
    This method has shown outstanding diagnostic capabilities
    on a number of different diseases.

     

    This study shows that the transfer learning model can use cross-regional microbial feature knowledge to achieve microbiome-based cross-regional disease diagnosis with high accuracy and
    robustness.
    In addition, this study provides a
    new feasible way to use artificial intelligence technology to break through regional limitations and realize cross-regional disease diagnosis based on microbial characteristics in clinical trials.

    It should be pointed out that China is a vast country, and the level of doctors varies from place to place
    .
    The artificial intelligence method based on transfer learning proposed in this project is essentially a universal diagnosis and treatment strategy
    based on big data.
    The application and promotion of this strategy can effectively overcome the diagnosis and treatment quality problems caused by regional differences, greatly improve the accuracy and speed of diagnosis and treatment, and help the "barefoot doctors in the new era" to provide higher quality diagnosis and treatment services
    for the people.

    The research was supported by the National Key Research and Development Program of the Ministry of Science and Technology (No.
    2018YFC0910502), National Natural Science Foundation of China (Nos.
    32071465, 31871334, 31671374), etc
    .
    This work has also been strongly supported
    by the main person in charge of the GGMP project, Professor Zhou Hongwei of Southern Medical University, etc.

    In recent years, the team of Professor Ning Kang of the School of Life Sciences of Huazhong University of Science and Technology has been continuously exploring the interdisciplinary field of bioinformatics, and developed a series of artificial intelligence mining methods for human microbiome big data, and successfully applied to the early diagnosis and recurrence monitoring of intestinal diseases, rheumatoid arthritis, non-infectious chronic diseases, human cancer and other diseases, related papers have been published in PNAS, Gut (3), Annals of the Rheumatic Diseases, Genome Biology, Genome Medicine, Gut Microbes and other international top journals
    in the fields of medicine, biology and bioinformatics.
    Relevant methods and models have been clinically tested
    in partner medical institutions.

    References:

    NanWang, MingyueCheng, KangNing.
    Overcoming regional limitations: transfer learning for cross-regional microbial-based diagnosis of diseases.
    GutDOI:
    10.
    1136/gutjnl-2022-328216
    .

     

    Links to papers: https://gut.
    bmj.
    com/content/early/2022/10/28/gutjnl-2022-328216

     

    This article is an English version of an article which is originally in the Chinese language on echemi.com and is provided for information purposes only. This website makes no representation or warranty of any kind, either expressed or implied, as to the accuracy, completeness ownership or reliability of the article or any translations thereof. If you have any concerns or complaints relating to the article, please send an email, providing a detailed description of the concern or complaint, to service@echemi.com. A staff member will contact you within 5 working days. Once verified, infringing content will be removed immediately.

    Contact Us

    The source of this page with content of products and services is from Internet, which doesn't represent ECHEMI's opinion. If you have any queries, please write to service@echemi.com. It will be replied within 5 days.

    Moreover, if you find any instances of plagiarism from the page, please send email to service@echemi.com with relevant evidence.