• Breakthrough in diagnostic testing enables ‘facial recognition for germs’
• Collaboration with John Radcliffe Hospital used AI to identify respiratory viruses and different strains with >97% accuracy in less than five minutes
• Founders seek further investment to help revolutionize respiratory infection diagnostics with better control of viral outbreaks, easing pressure on NHS

February 6, 2023 – Oxford, UK – Nicola Shiaelis AND Dr. Nicole RobbOxford University scientists and co-founders of a health technology company Organic painthave developed a world-first diagnostic test powered by artificial intelligence, capable of identifying known respiratory viruses such as influenza and COVID-19 within five minutes of just a nose or throat swab.

Current tests are lab-based and either time consuming or fast and less accurate. They are also limited, for example, a lateral flow test only checks for an infection. This means that diseases spread while infected people wait for results or because they are unaware they are infected. These findings, published in a peer-reviewed scientific journal ACS Nano, demonstrate how machine learning can significantly improve the efficiency, accuracy and time spent not only identifying different types of viruses, but also differentiating strains. This could help better control the spread of respiratory infections and relieve pressure on the NHS and healthcare staff, while also reducing healthcare waste.

Shiaelis and Robb teamed up with John Radcliffe Hospital to evaluate a new method that uses AI software to identify viruses. The innovative testing technology combines molecular labeling, computer vision and machine learning to create a universal diagnostic imaging platform that directly examines a patient sample and can identify which pathogen is present in seconds. Just like facial recognition software, but for germs.

Preliminary research has shown that this test could identify the COVID-19 virus in patient samples. But to develop it further, they wanted to determine if the test could be used to diagnose more respiratory infections.

In the study, the researchers began by labeling viruses in more than 200 clinical samples from John Radcliffe Hospital. Images of labeled samples were captured and processed by custom machine learning software trained to recognize specific viruses by analyzing their fluorescent labels, which appear differently for each virus because surface size, shape and chemistry vary. The results show that the technology can rapidly identify different types and strains of respiratory viruses, including influenza and COVID-19, within five minutes and with >97% accuracy.

Scientists have trained Organic paint to further develop this technology and are now seeking further investment to accelerate development and bring it to the forefront of healthcare.

Dr. Nicole Robb, scientist and co-founder of Pictura Bio, comments: “Cases of respiratory infections in winter 2022/23 reached record levels, increasing the number of people seeking medical assistance. This, combined with the COVID-19 backlog, staff shortages, tighter budgets and an aging population, places immense and unsustainable pressure on the NHS and its workforce.

“Pictura Bio’s simplified method of diagnostic testing is faster and more cost-effective, accurate and future-proof than any other test currently available. If we want to detect a new virus, all we have to do is retrain the software to recognize it, rather than developing an entirely new test. Our results demonstrate the potential of this method to revolutionize viral diagnostics and our ability to control the spread of respiratory diseases. »

The publication’s lead author, co-founder and DPhil postgraduate researcher at the University of Oxford, Nicolas Shiaelis, added: “It is inevitable that more COVID-like viruses will emerge. This reinforces the need for more advanced diagnostic test technology so that we can reduce the impact of new viruses on public health and the NHS. »

The technology is now licensed from Pictura Bio, which aims to turn the method into a diagnostic test by creating a dedicated imager and disposable cartridge for use in point-of-care testing, with limited user input. The team will also expand the number of viruses the models are trained on and eventually begin looking at other pathogens, such as bacteria and fungi, in respiratory, blood and urine samples.

About Pictura Bio
Pictura Bio is a health technology company that developed the world’s first one-minute pathogen recognition platform. Software, like facial recognition for pathogens, is a universal testing platform, powered by artificial intelligence, to provide accurate, digital image-based identification of infectious diseases. The innovation includes two patented innovations: PIC-ID and IRIS. PIC-ID coats the lipid membrane while IRIS, a deep learning neural network, simultaneously analyzes and classifies pathogens. It’s a fundamentally different way of identifying ALL infectious agents, and the company’s vision is to have a Lab in a Box for one-minute testing on desktops in hospitals, doctors’ offices and pharmacies around the world.

Pictura Bio’s solutions drive faster, simpler and more cost-effective testing than resource-consuming molecular diagnostics, changing the way we treat disease. Technology will enable healthcare decision makers to revolutionize healthcare as we know it by strengthening the fight against antimicrobial resistance, reducing healthcare waste and limiting exposure to viruses.

PIC-ID and IRIS – the technology
The Pictura Bio team is focusing on packaging its PIC-ID technology into a simple desktop « laboratory in a box », called IRIS (Instant Recognition Identification System) about the size of a household microwave oven, comprising a high-resolution fluorescent microscope disposable power and image acquisition and processing technology. This technology has reached proof of concept and is protected by patents and the team is now developing the technology to scale this product for wider availability and adoption in the medical community. Today, PIC-ID is trained to detect enveloped viruses from respiratory samples, but the future has unlimited potential for recognizing pathogens in samples.

The neural network is the database that IRIS checks to identify a pathogen. This process ensures that as the network learns new tests and pathogens, it can be continuously delivered via a software update, rather than requiring new panels and the infrastructure to follow along.

For more information visit https://pictura.bio/

PR contacts:
Transatlantic entertainment
picturebio@transatlanticent.com
+44 (0)7968 949746