Over 27,000 people die of cancer every day, representing a staggering one in six deaths in the world.Our World in Data
Treating cancer takes teamwork. Pathologists, in particular, play an increasingly crucial role. Inputs from these behind-the-scenes specialists, who examine body tissues, diagnose and guide patient care, contribute to some 70 per cent of all hospital decisions.
The problem is that pathologists are in short supply around the world. In the US alone, the pathologist population declined by over 17 per cent between 2007 and 2017, while their diagnostic workload rose by over 41 percent.https://meridian.allenpress.com/aplm/article/148/6/735/496206/International-Medical-Graduates-and-the-Shortage Patients can wait up to 60 extra days to begin treatment because of understaffed laboratories.https://proscia.com/infographic-the-state-of-the-pathology-workforce-2022/ The problem is getting even worse after Covid.
Aneesh Sathe completed his PhD in Mechanobiology at the National University of Singapore where he developed computer vision systems to identify cancer cells. Credit: istockphoto
Singapore-based start-up Qritive is trying to ease the burden with the help of artificial intelligence. Qritive uses AI to process large data sets, which leads to faster and more accurate cancer diagnoses. The company works with a network of more than 150 pathologists around the world to train the digital pathology software, which has already analysed almost half a million images.
“In some cases, you need to count millions of cells to establish a ratio and treat the patient accordingly. Most often it’s an estimate, but if a computer can count it, the result is much more accurate by default,” says Aneesh Sathe, co-founder.
"Weeks to minutes"
Qritive’s technology is based on Sathe’s PhD research at the National University of Singapore.
Sathe, with an interest in both biology and computation, was studying mechanobiology – an emerging field of science which involves analysing the world of cells through the lens of physics and engineering.
During the research, he realised how laborious the process of biological image analysis was. Sathe created a programme that used machine learning and AI to assist him.
“It brought the (image analysis) time down from three or four weeks, to 10 minutes,” he says.
Persuading hospitals to take a bet on Qritive’s system was an early obstacle for the company, which was founded in 2017. It took two years for Qritive to publish its first studies.
Seeing clients use its tools a year later in Singapore, then in some of the largest hospitals in India, was a significant step for the start-up.
“It proved the tech was feasible,” says Sathe.
In January 2023, Qritive received USD7.5 million in funding to develop new products and expand its operations.
Qritive has been gaining traction in the US, where the shortage of pathologists is becoming acute.
Currently, the number of practicing pathologists in the US stands at 5.7 per 10,000 people, but this is forecast to fall to 3.7 by the end of decade.https://pubmed.ncbi.nlm.nih.gov/23738764/
Sathe hopes Qritive will make diagnostic expertise more accessible to hospitals and patients in developing regions, such as Africa – where there is just one pathologist for every million people – and even in rural parts of Europe.
“In these smaller towns it takes way too long,” he says.
“The patient has to be sent to the big hospital in the city to get a diagnosis. If we can democratise who gets access to this care, I see that as an area where we can grow.”
This article is an edited version of the original, which can be accessed here.
Investment insights
by Flora Liu, Senior Client Portfolio Manager, thematic equities, Pictet Asset Management
-
As the global number of cancer deaths continues to increase and associated costs are expected to exceed USD24 trillion by 2050, investment in cancer screening, diagnosis and treatment could yield substantial health and economic benefits.
-
Companies in both private and public markets have been actively developing innovative solutions, including the use of artificial intelligence (AI) and machine learning to improve cancer diagnosis and treatment.
-
Diagnostics imaging and digital pathology are areas that are fast growing within healthcare as cancer cases rise and as more screening programmes are being offered by national governments. Medical imaging’s total addressable market is expected to grow to USD2.7 billion by 2032 from less than USD300 million today (Source: Jefferies). AI and machine learning not only enable faster and more accurate diagnostics readings that help improve patient’s outcomes but also help alleviate labour shortage in healthcare.
