Diagnostic test incorporates artificial intelligence and could shorten the time clinical laboratories need to determine patients’ risk for antimicrobial resistance
Sepsis continues to be a major killer in hospitals worldwide. Defeating it requires early diagnosis, including antimicrobial susceptibility testing (AST), and timely administration of antibiotics. Now, in a pilot study, scientists at Seoul National University in South Korea have developed a new clinical laboratory test that uses artificial intelligence (AI) to pinpoint the condition sooner, enabling faster treatment of the deadly bacterial infection.
Sepsis, also known as septicemia or blood poisoning, is a serious medical condition that occurs when the body overreacts to an infection or injury. This often takes place in hospitals through blood-line infections and exposure to deadly bacteria. The dangerous reaction causes extensive inflammation throughout the body. If not treated early, sepsis can lead to organ failure, tissue damage, and even death.
Research teams around the world are creating new technologies and approaches to slash time to answer from when blood specimen is collected to a report of whether the patient is or is not positive for sepsis. The Seoul National University scientists’ new approach is yet another sign for microbiologists and clinical laboratory managers of the priority test developers are giving to solving the problem of diagnosing sepsis faster than using blood culture methodology, which requires several days of incubation.
The Seoul scientists published their findings in the journal Nature titled, “Blood Culture-free Ultra-rapid Antimicrobial Susceptibility Testing.”
“Sepsis strikes over 40 million people worldwide each year, with a mortality rate ranging from 20% to 50%,” said Sunghoon Kwon, PhD (above), professor of electrical and computer engineering at Seoul National University and senior author of the study, in an interview with The Times in the UK. “This high mortality rate leads to over 10 million deaths annually. Thus, accurate and prompt antibiotic prescription is essential for treatment,” he added. Clinical laboratories play a critical role in the testing and diagnosis of sepsis. (Photo copyright: Seoul National University.)
Reducing Time to Diagnosis
Seoul National University’s approach begins with drawing a sample of the patient’s blood. The researchers then attach special peptide molecules to magnetic nanoparticles and add those nanoparticles to the blood sample. The particles bind to the harmful pathogens in the blood.
The harmful bacteria are then collected using magnets. Their DNA is extracted, amplified, and analyzed to establish the type of microbes that are present in the sample.
The pathogens are exposed to antibiotics and an AI algorithm evaluates their growth patterns to forecast what treatments would be most beneficial to the patient. This last step is known as antimicrobial susceptibility testing or AST.
“The principle is simple,” said Sunghoon Kwon, PhD, professor of electrical and computer engineering at Seoul National University and senior author of the study, in a Nature podcast. “We have a magnetic nanoparticle. The surface of the magnetic nanoparticle we coat in a peptide that can capture the bacteria.”
Kwon is the CEO of Quantamatrix, the developer of the test.
The complete process can be performed on one machine and results are available in about 12 hours, which reduces typical AST time by 30 to 40 hours when compared to traditional processes.
“Sepsis progresses very quickly, with the survival rate dropping with each passing hour,” Kwon told The Times UK. “Every minute is crucial.”
Preventing Antimicrobial Resistance
The team assessed the performance of their test on 190 hospital patients who had a suspected sepsis infection. The test achieved a 100% match in the identification of a bacterial species. The test also achieved an efficiency of 96.2% for capturing Escherichia coli (E. coli) and 91.5% for capturing Staphylococcus aureus.
“Treatment assessment and patient outcome for sepsis depend predominantly on the timely administration of appropriate antibiotics,” the authors wrote in Nature.
“However,” they added, “the clinical protocols used to stratify and select patient-specific optimal therapy are extremely slow,” due to existing blood culture procedures that may take two or three days to complete.
“The microbial load in patient blood is extremely low, ranging between 1 and 100 colony-forming units (CFU) ml−1 and is vastly outnumbered by blood cells,” the study authors explained. “Due to this disparity, prior steps—including blood culture (BC) to amplify the number of pathogens followed by pure culture to subculture purified colonies of isolates—have been essential for subsequent pathogen species identification (ID) and AST.”
Further research, studies and regulatory approval are needed before this technique becomes available, but the South Korean scientists believe it could be ready for use within two to three years. They also state their test can help prevent antimicrobial resistance (AMR) and bolster the strength of existing antibiotics.
Previous Studies
The Seoul National University study is just the latest effort by scientists to develop faster methods for clinical laboratory testing and diagnosing of sepsis.
In September, Dark Daily reported on a similar test that uses digital imaging and AI to determine sepsis risk for emergency room patients.
That ebrief, titled, “10-Minute Blood Test Uses Digital Images and AI to Determine Sepsis Risk for Emergency Room Patients,” outlined how a tool called IntelliSep, which was created through a partnership between San Francisco-based medical diagnostics company Cytovale and the Louisiana State University Health Sciences Center (LSUHSC) in Baton Rouge, can spot biomarkers for sepsis within 10 minutes.
According to the Centers for Disease Control and Prevention (CDC), at least 1.7 million adults develop sepsis annually in the US, and that at least 350,000 die as a result of the condition. CDC also lists sepsis as one of the main reasons people are readmitted to hospitals.
Microbiologists and clinical laboratory managers should be aware that scientists are prioritizing the creation of new testing methods for faster detection of sepsis. Various research teams around the world are devising technologies and approaches to reduce the time needed to diagnose sepsis to improve patient outcomes and save lives.
—JP Schlingman
Related Information:
Scientists Say They Developed Faster Way to Diagnose, Treat Sepsis
Rapid Sepsis Test Identifies Bacteria That Spark Life-threatening Infection
“Game-changing” Sepsis Test Could Save Thousands of Lives
10-Minute Blood Test Uses Digital Images and AI to Determine Sepsis Risk for Emergency Room Patients
Blood Culture-free Ultra-rapid Antimicrobial Susceptibility Testing