echemi logo
Product
  • Product
  • Supplier
  • Inquiry
    Home > Biochemistry News > Biotechnology News > Machine learning AI tools: helping oncologists make better treatment decisions

    Machine learning AI tools: helping oncologists make better treatment decisions

    • Last Update: 2021-09-29
    • Source: Internet
    • Author: User
    Search more information of high quality chemicals, good prices and reliable suppliers, visit www.echemi.com

    When treating cancer patients, the goal of oncologists is to predict the patient’s disease progression and make key treatment decisions
    .
    Understanding the unique molecular characteristics of tumors can help guide these decisions and provide clues to distinguish whether the cancer is slow-growing, aggressive and deadly, and whether it will resist treatment

    .
    New molecular profiling techniques have generated a lot of information about tumors, but doctors have been working hard to translate all this data into meaningful prognosis

    .

    Researchers at the Broad Institute of MIT, Harvard University, and the Dana-Farber Cancer Institute have developed a new model that can distinguish the genomic characteristics of deadly prostate cancer from those that are unlikely to cause symptoms or death
    .
    It can also help clinicians predict whether prostate cancer patients’ tumors will spread to other parts of the body, or whether they will become more resistant to treatment over time

    .
    This model is called P-NET and can also identify molecular features, genes, and biological pathways that may be related to disease progression

    .
    P-NET uses machine learning-based algorithms to analyze the known molecular characteristics of tumors, distinguish whether the tumor is an aggressive, potentially fatal type, and whether there are signs of spreading to other parts of the body.
    It can also help cancer researchers understand More knowledge about the biology of drug-resistant diseases, and may be applicable to other cancers

    .
    The model was published in the journal Nature

    .

    Eliezer (Eli) Van Allen, senior author of the study, an associate professor at Broad, an associate professor at the Dana-Farber Cancer Institute and Harvard Medical School, said that P-NET not only provides prognosis for patients, "we have not only improved the prediction" whether cancer will The ability to metastasize and which genes may be associated with this state, and as cancer researchers, we can use the interpretability of this model to understand the biology of these disease states
    .
    "

    Build a better model

    In order to build a model that can distinguish early and advanced prostate cancer tumors, the researchers developed a specialized deep learning model with a customized architecture that is more interpretable than other algorithms
    .
    In the deep learning model, a multi-layer neural network "learns" from a large data set with a "recognition pattern like the human brain"

    .

    Using this method, a team led by Haitham Elmarakeby, a lecturer at the Dana-Farber Cancer Institute, an affiliated researcher at the Broad Institute, and the first author of the study, combined relevant biological information-such as known genes and metabolism or signaling pathways.
    The relationship-added to their model

    .
    Then, they trained P-NET to use the genome sequence and somatic or non-genetic mutation data of more than 1,000 prostate cancer patients to predict whether the tumor is aggressive

    .
    When the research team tested their model with data from other prostate cancer patients, they found that the model correctly distinguished 80% of metastatic tumors from primary, slower progressing tumors

    .
    This shows that the trained model can perform the same function on the new data

    .

    In the process of ranking related genes and signaling pathways based on importance and verifying P-NET by weight, the team also determined that the MDM4 gene may be related to prostate cancer progression and drug resistance
    .
    Previously, scientists believed that the gene was related to other cancers, but not to prostate cancer

    .
    In collaboration with the laboratory of William Hahn, a member of the Broad Institute, the team found that overexpression of MDM4 in prostate tumor cells is associated with drug resistance

    .
    When they turned off the gene using gene editing, cell proliferation decreased, indicating that cancer cells may be more sensitive to treatment

    .
    These results indicate that scientists can reuse drugs that inhibit MDM4 to treat prostate tumors-some of these drugs are currently being studied for the treatment of other cancers

    .

    Researchers say that through modification, P-NET can also help oncologists predict the disease progression and treatment response of other cancers
    .
    "This structure is not limited to prostate cancer," Elmarakeby said

    .
    "Our model has great potential and can be expanded in different ways

    .
    " P-NET will continue to improve, and he and his team are integrating other types of data-including more genetic and imaging data-into In the model

    .
    He said: "This is just the beginning of

    our fusion of cancer biology and machine learning .
    " "We believe that through this fusion, we can truly provide more discoveries for cancer patients

    .
    "


    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.