Machine Learning Tools for COVID-19 Patient Screening and Improved Lab Test Management to Be Discussed at the 2021 AACC Annual Scientific Meeting

ATLANTA, Sept. 27, 2021 /PRNewswire/ — Scientists have created a brand new machine studying instrument that would assist healthcare employees to rapidly display screen and direct the circulation of COVID-19 sufferers arriving at hospitals. Results from an analysis of this algorithm, together with a synthetic intelligence technique that improves take a look at utilization and reimbursement, had been offered at present at the 2021 AACC Annual Scientific Meeting & Clinical Lab Expo.
Streamlining Hospital Admission of COVID-19 Patients
It is necessary for clinicians to rapidly diagnose COVID-19 sufferers once they arrive at hospitals, each to triage them and to separate them from different weak sufferers who could also be immunocompromised or have pre-existing medical situations. This might be tough, nonetheless, as a result of COVID-19 shares many signs with different viral infections, and the most correct PCR-based exams for COVID-19 can take a number of days to yield outcomes.
A group of researchers led by Rana Zeeshan Haider, PhD, and Tahir Sultan Shamsi, FRCP, of the National Institute of Blood Disease in Karachi, Pakistan, has due to this fact created a machine studying algorithm to assist healthcare employees effectively display screen incoming COVID-19 sufferers. The scientists extracted routine diagnostic and demographic information from the information of 21,672 sufferers presenting at hospitals and utilized a number of statistical methods to develop this algorithm, which is a predictive mannequin that differentiates between COVID-19 and non-COVID-19 sufferers. During validation experiments, the mannequin carried out with an accuracy of up to 92.5% when examined with an unbiased dataset and confirmed a destructive predictive worth of up to 96.9%. The latter implies that the mannequin is especially dependable when figuring out sufferers who haven’t got COVID-19.
“The true destructive labeling effectivity of our analysis advocates its utility as a screening take a look at for fast expulsion of SARS-CoV-2 from emergency departments, aiding immediate care choices, directing patient-case circulation, and fulfilling the function of a ‘pre-test’ regarding orderly RT-PCR testing the place it’s not helpful,” stated Haider. “We suggest this take a look at to settle for the problem of essential diagnostic wants in useful resource constrained settings the place molecular testing isn’t underneath the flag of routine testing panels.”
Optimizing Lab Test Selection and Reimbursement
Of the 5 billion lab orders submitted every year, at least 20% are thought-about inappropriate. These inappropriate exams can lead to slower or incorrect diagnoses for sufferers. Such exams may not be coated by Medicare in the event that they weren’t meant to be used for explicit medical situations or in the event that they had been ordered with the incorrect ICD-10 diagnostic codes, which in flip raises well being prices.
Rojeet Shrestha, PhD, of Patients Choice Laboratories in Indianapolis, set out together with colleagues to decide if an automatic take a look at administration system generally known as the Laboratory Decision System (LDS) may assist enhance take a look at ordering. The LDS scores potential exams primarily based on medical necessity and testing indication, serving to suppliers decrease take a look at misutilization and choose the greatest exams for a given medical situation.
Using LDS, the researchers re-evaluated a complete of 374,423 take a look at orders from a reference laboratory, 48,049 of which had not met the standards for protection underneath Medicare. For 96.4% of the first 10,000 take a look at claims, the LDS rating system beneficial various exams that higher matched the medical necessity or had a extra acceptable ICD-10 code. Of these suggestions, 80.5% would additionally meet Medicare insurance policies. All of this means that the LDS may assist right mistaken or inappropriate lab orders.
“Our examine implies that use of the automated take a look at ordering system LDS could be extraordinarily useful for suppliers, laboratories, and payers,” stated Shrestha. “Use of this algorithm-based testing choice and ordering database, which charges and scores potential exams for any given illness primarily based on scientific relevance, medical necessity, and testing indication, would ultimately assist suppliers to choose and order the proper take a look at and cut back over- and under-utilization of exams.”
Abstract InformationAACC Annual Scientific Meeting registration is free for members of the media. Reporters can register on-line right here:
Abstract A-226: Machine-learning primarily based decipherment of cell inhabitants information; a promising hospital front-door screening instrument for COVID-19 might be offered throughout:
Student Poster PresentationMonday, September 279 a.m. – 5 p.m. Scientific Poster Session
Tuesday, September 289:30 a.m. – 5 p.m. (presenting creator in attendance from 1:30 – 2:30 p.m.)Abstract B-011: Use of synthetic intelligence for efficient take a look at utilization and to improve reimbursement might be offered throughout:
Scientific Poster SessionWednesday, September 299:30 a.m. – 5 p.m. (presenting creator in attendance from 1:30 – 2:30 p.m.)
All periods will happen in the Poster Hall, which is positioned in Exhibit Hall C of the Georgia World Congress Center in Atlanta.
About the 2021 AACC Annual Scientific Meeting & Clinical Lab ExpoThe AACC Annual Scientific Meeting gives 5 days full of alternatives to study thrilling science from September 26-30. Plenary periods discover COVID-19 vaccines and virus evolution, analysis classes realized from the pandemic, synthetic intelligence in the clinic, miniaturization of diagnostic platforms, and enhancements to therapies for cystic fibrosis.
At the AACC Clinical Lab Expo, greater than 400 exhibitors will fill the present ground of the Georgia World Congress Center in Atlanta with shows of the newest diagnostic know-how, together with however not restricted to COVID-19 testing, synthetic intelligence, cell well being, molecular diagnostics, mass spectrometry, point-of-care, and automation.
About AACCDedicated to reaching higher well being by way of laboratory medication, AACC brings collectively greater than 50,000 scientific laboratory professionals, physicians, analysis scientists, and enterprise leaders from round the world targeted on scientific chemistry, molecular diagnostics, mass spectrometry, translational medication, lab administration, and different areas of progressing laboratory science. Since 1948, AACC has labored to advance the widespread pursuits of the subject, offering applications that advance scientific collaboration, information, experience, and innovation. For extra info, go to
Christine DeLongAACCSenior Manager, Communications & PR(p) 202.835.8722[email protected]
Molly PolenAACCSenior Director, Communications & PR(p) 202.420.7612(c) 703.598.0472[email protected] 
Related Links

Recommended For You