Pathan, one of the largest pathology laboratories in the Netherlands, has started a collaboration with Boston-based Aiforia, a leading software company providing artificial intelligence (AI) solutions for medical image analysis. The aim of the partnership is to build clinically validated AI models to assist pathologists with primary diagnostics, enhance laboratory workflows to make faster and better informed decisions to ultimately improve patient care.
Pathologists around the world are facing an increasing burden as rates of disease continue to rise as a result of a higher life expectancy worldwide. Meanwhile, the methods these healthcare professionals currently rely on are manual, sometimes time-consuming, and subject to a certain degree of bias. Pathan’s pathologists alone handle about 110,000 requests a year. After adopting digital pathology in their lab to digitize their analysis, they are ready to add artificial intelligence to this workflow to optimize their output.
“With this collaboration Pathan’s aim is to better understand the opportunities that AI can offer in the future organization of the diagnostic process, especially given the fact that the number of oncology patients will increase substantially over the next 20 years,” says Arlinke Bokhorst, director of Pathan. “With AI, Pathan expects to support pathologists with a toolbox of routine assessments, measures and recognition options to optimize their diagnostic workflow.”
Artificial intelligence excels at image analysis, with the potential to surpass human capabilities in speed and accuracy. The Aiforia Platform brings deep learning AI straight to the hands of medical experts with its intuitive software. Pathologists are able to create AI models to analyze images from any sample or disease, for example breast cancer, improving the accuracy and speed of analysis while producing quantitative and reproducible results for clinicians.
The Pathan pathologists are working closely with Aiforia’s expert team of scientists to use Aiforia’s software to build clinical grade algorithms, or AI models, aiming to be one of the first to receive CE-IVD marking for these clinical applications.
Pathan starts out with the development and the routine use of relatively simple algorithms, for instance algorithms that aid in screening digital histological slides for focal abnormalities, like helicobacter or acid-fast bacteria. Later on they expect to develop more complex algorithms, for instance ones that aid in tumor grading, quantitative assessment of immunohistochemical stains or screening for angioinvasion.
“This is an exciting step for Aiforia, if not the most exciting!” explains Kaisa Helminen, chief operating officer of Aiforia. “Through this collaboration we are able to bring the latest advances in technology to the daily clinical laboratory practice and ultimately influence patient care by helping pathologists make speedier, more consistent, and accurate decisions.”
For more information, visit Aiforia.
Featured image: Aiforia’s cloud-based AI development platform can be used for the analysis of any image in pathology for research and clinical work. (Courtesy: Aiforia)