![<?echo $_SERVER['SERVER_NAME'];?>](/template/twentyseventeen/skin/images/header.jpg)
The IBM website released a message on Monday that it is incorporating machine learning (ML) into the diagnostic field, hoping that one day ML technology can help effectively diagnose early-onset Alzheimer's disease.
Technology giant IBM said on Monday that machine learning and artificial intelligence (AI) can be used to replace the invasive and expensive detection of existing Alzheimer's disease.
The paper published in the Science Report by the IBM Australia team reports on the results of the study.
Alzheimer's disease is currently incurable and can only be treated by palliative care. Symptoms of Alzheimer's disease include progressive deterioration of memory, memory confusion, and the inability of patients to successfully complete routine tasks that were once familiar. Early diagnosis of the disease can help patients and their families prepare, and early patients can also participate in relevant medical experiments. Since the beginning of this century, researchers have conducted hundreds of medical experiments on Alzheimer's disease.
However, the current diagnosis of Alzheimer's terminal illness is not only expensive but also highly invasive. Current early diagnostic methods include finding specific biomarkers in the spinal fluid, and requiring lumbar puncture to obtain spinal fluid, which is painful and can cause bleeding. Because there is no cure for Alzheimer's disease, finding a non-invasive test that is conducive to the development of early diagnosis of Alzheimer's disease will greatly promote a new wave that does not depend on brain tissue damage. Clinical trials in patients with advanced disease. According to IBM, ML may help narrow the gap between early detection and clinical trials.
The use of this technique depends on the successful development of a method for testing amyloid-beta, a peptide of the spinal fluid, studies have shown. There is a long period of time between the change of amyloid β in Alzheimer's patients and the loss of memory in patients.
A research article published by IBM describes a method for predicting the concentration of amyloid beta in spinal fluid using machine learning based on the identification of protein pools in the blood.
The article proposes some ML-based models that may one day predict the risk of Alzheimer's disease by simple blood tests. The research team of the article believes that their ML model predicts future risk factors with an accuracy rate of up to 77%.
IBM said, "Although the test is still in the early stages of research, the results may help to improve the choice of individual drug testers: the possibility of Alzheimer's disease in patients with mild cognitive impairment in the spinal fluid with abnormal amyloid concentrations Sex is 2.5 times higher."
Models developed based on this ML application are likely to provide a framework for future new forms of Alzheimer's disease testing, which can replace lumbar puncture and speed up the diagnostic process, and can greatly reduce the cost and invasiveness of surgery.
These models are still in the early stages. There is still a long way to go before machine learning truly enters the field of cognitive disease diagnosis. However, the IBM team said that the ML algorithm they developed, in addition to Alzheimer's disease, is also applicable to other diseases and can be extended to other models and tests based on spinal fluid biomarkers.
Racking Accessories,Custom Mezzanine Systems,Storage Rack,Warehouse Rack
Nanjing Chinylion Metal Products Co., Ltd , https://www.chinylionclrack.com