Scientists use supercomputers and AI to find out how good (or lethal) your road medicine are
A team of researchers from the University of Victoria has developed an AI system that can be used to determine the expected chemical composition of drugs. While these are supercomputers and a robust cocktail of cloud-based machine learning technologies, the ultimate goal is to make it easy for just about anyone to tell what’s in their medication.
About 70,000 deaths from overdose are recorded annually in the US alone. While the causes are both innumerable and systemic, a significant number of these tragedies could potentially be avoided if consumers knew what was in their drugs.
The problem affects those who take “legal” prescription drugs under the care and advice of a properly licensed medical practitioner, as well as those who abuse prescription drugs or use so-called “street” drugs.
[Read next: Meet the 4 scale-ups using data to save the planet]
According to a report by Ken Strandberg in Computer Science for Technology Networks, the project addressed concerns about fentanyl levels in opioids and inconsistencies in unregulated prescription drug markets.
When most lay people think about drug testing, they are likely thinking of something like a urine sample or a hair test. However, drug control is used to determine what is in a drug itself. Usually we have to take a drug manufacturer at their word. When Big Pharma tells us what’s in our pills, we pretty much have to believe it.
The same applies to the so-called street drug market. Without a laboratory and some specialized equipment, it is virtually impossible for anyone to determine what is actually in Molly, MDMA, or other drugs people take.
The big idea here is ultimately to develop a system for medical professionals and pharmacists alike to quickly and accurately determine what is actually in the drugs they dispense. The scientists also want their work to be made available to the public.
Strandberg’s play quotes Dennis Hore, a member of the University of Victoria team:
A recently funded project aims to build an interactive kiosk where people can bring their samples for analysis. The computer provides scientifically based guidance, with no gender or racial prejudice or prejudice based on their answers to questions.
Dirty drugs, whether on prescription or on the street, are responsible for countless deaths. Having a system that makes it quick and easy to find out what’s in your medication can be an incredible game changer. However, shrinking a laboratory to the size of a kiosk is not an easy task.
When normal drug analysis involves physical chemistry – with machines and cups etc – the Victoria team’s system relies on a robust cocktail of AI, machine learning, and supercomputers to infer the surface. The reason for this is simple: we can’t put a beaker or a centrifuge on the internet, but we can use cloud computing to put supercomputer-based AI online.
Take quickly: The researchers use brute force here, and not just because the supercomputer they use (called the Arbutus 2) has 1,000 Intel Xeon processors. The AI that powers analysis uses a kitchen sink approach that encompasses different types of AI and ML paradigms. According to the team, this enables the system to determine both known quantities and completely new drug compounds.
Much like laboratory tests help protect pharmacies and their patients from “bad batches” and needle exchange programs help protect addicts from disease, this AI system could serve as a new, powerful form of protection for drug users.
Published on January 11, 2021 – 18:59 UTC