Daily News / Central Commissioner Zou Zhizhong reported that the mass spectrometry signal generated through the clinical bacterial identification process can predict the risk of drug-resistant bacteria, which can not only effectively reduce the clinical medical examiner’s inspection process, but also assist the clinician to use antibiotics accurately and reduce the occurrence of severe cases. At the same time, reduce the problems caused by the abuse of antibiotics, and help to accelerate the digital transformation of Taiwan’s medical care; the technical team of Professor Chen Chaorong of the Institute of Integrated Chinese and Western Medicine, China Medical University, published the internationally renowned journal “Microbiol Spectr” in April 2022 and won the United States The patent has won the “2022 Future Technology Award” of the Ministry of Science and Technology in 2022, which is well deserved.
It is understood that, in the current clinical medical inspection process, after bacterial species identification and antimicrobial susceptibility testing for the source of infection, it takes about three to five days to provide the test information to clinicians for antibiotic treatment evaluation; The team developed a “Rapid prediction system for intelligent antibiotics-resistant bacteria” (Rapid prediction system for intelligent antibiotics-resistant bacteria) combined with clinical mass spectrometer signal (MALDI-TOF MS) and machine learning model to predict the phenotypes of different drug resistance of the same bacteria, Therefore, 24-48 hours can be greatly shortened to provide adjunctive medication instructions.
The establishment of the “Intelligent Antibiotic-Resistant Bacteria Rapid Prediction System” technology is aided by the cooperation of the China Medical University academic team, including the intelligent antibacterial team led by Deyang Zhou, Dean of the Affiliated Hospital of China Medical University, and the machine learning model developed by You Jiaxin, Director of the Smart Medical Technology Innovation Center Technical support with Tian Ni, Deputy Director of Laboratory Medicine Department.
The “Intelligent Drug Resistant Bacteria Rapid Prediction System” technology uses MALDI-TOF to perform protein mass spectrometry analysis to identify bacterial species, and joins the team to develop a successful machine learning classification model that can directly classify drug-resistant bacteria and non-drug-resistant bacteria. The technical team uses the existing clinical big data to calculate the special performance map of drug-resistant microorganisms. When the existing bacterial species identification process is completed, it can also immediately know whether it is a known drug-resistant bacteria, which will help clinical antibiotics in advance. Shorten the patient’s hospital stay and reduce antibiotic abuse.
The “Intelligent Drug Resistant Bacteria Rapid Prediction System” technology includes the use of automated mass spectrometry data pre-processing units, machine cognitive models and a large number of bacterial drug resistance mass spectrometry databases. According to the above drug-resistant bacteria mass spectrometry analysis system, the mass spectrometry data preprocessing unit pre-processes and standardizes the mass spectrometry data as a machine cognitive model for subsequent drug resistance detection.
The current model is resistant to methicillin-resistant Staphylococcus aureus, carbapenem-resistant Klebsiella pneumoniae, carbapenem-resistant Acinetobacter baumannii, vancomycin-resistant Enterococcus faecium, and carbapenem-resistant The classification and prediction of eight clinical monitoring bacteria including Escherichia coli, Enterobacter cloacae, Morganella, and Pseudomonas aeruginosa, which are anti-carbapenem drugs, have an accuracy rate of 0.75 – 0.93. . In the future, 21 other bacteria will be added to the clinical routine.
In the “Intelligent Drug Resistant Bacteria Rapid Prediction System”, the protein identification technology was developed through Prof. Chen’s Ministry of Science and Technology project: “To develop peptide/phosphopeptide online alkaline fractionation chromatography-mass spectrometry, antibody/protein flat plate technology and metabolite platform to be applied to biomarker search, rapid detection and screening of Chinese herbal medicine inhibitors;” completed with the support. Combined with the machine learning model developed by Director You Jiaxin of the Smart Medical Technology Innovation Center, we successfully identified methicillin-sensitive staphylococcus aureus (MSSA) and methicillin-resistant Staphylococcus aureus (MSSA). resistant staphylococcus aureus, MRSA), the team of Prof. Chaorong Chen successfully identified the protein sequence by liquid chromatography mass spectrometry. Using computer simulation, find out the reason for the decrease in the binding force of the peptide to the antibiotic due to the change of the peptide sequence, which leads to the increase of the drug resistance. In the future, the protein sequences that can distinguish different drug-resistant strains will also be identified one by one, in order to be applied to the development of new antibiotics in the future.
It is reported that the technical team of the “Intelligent Drug-Resistant Bacteria Rapid Prediction System” has cooperated with four hospitals in Taiwan, and its bacterial library database should be the largest database currently known. This system can be combined with rapid bacterial sample preparation method, MALDI-TOF and machine learning to get bacterial resistance prediction system faster. This research can further find novel drug-resistant bacteria biomarkers through mass spectrometry identification, and try to find out the downstream drug resistance mechanisms involved in the biomarkers, so as to further develop test methods and a new generation of antibiotics. In the industry, it can be directly applied to the clinical strain testing system of various hospitals in Taiwan, as well as early drug administration and subsequent medical cost savings. In addition, the protein biomarkers developed and identified from this platform can be applied to the research and development of rapid chip detection and antibody antibiotics.
The post Accelerating the Digital Transformation of Medicine to Predict Antibiotic-Resistant Bacteria Prof. Chaorong Chen won the Future Technology Award Appeared first on Mega News.