Digital pathology images create immense quantities of data sourced from databases brimming with healthcare and research findings. This volume of data can be overwhelming and time-consuming for pathologists and scientists to analyse.
Artificial Intelligence is fast becoming a solution to this problem, serving as a supplementary analysis or validation tool in imaging analytics, helping pathologists and scientists to process more tests in less time. AI is capable of handling the gigantic quantity of data created throughout the patient care lifecycle to improve pathologic and scientific diagnosis, classification, prediction, and prognostication of many diseases and infections.
AI can learn what is normal in a tissue sample, detecting subtle patterns and then flagging those biopsies that require further investigation with the pathologist or scientists. The pathologist or scientist can then look at a whole slide or tests at once and zoom in on areas of interest. By contrast, a conventional microscope does not allow for looking at an entire tissue sample, which can increase the likelihood of pathologists or scientist missing serious abnormalities. Thanks to AI’s ability to analyse data rapidly and pick up on areas of concern, pathologists and scientists can spend less time going through what are completely normal biopsies and other tests and devote more time to those patients’ samples that are problematic.
David Dimond, Chief Innovation Officer Global Healthcare & Life Sciences at Dell Technologies, believes AI in healthcare and life sciences organisations is allowing scientists and clinicians to unlock medical breakthroughs at a radical pace. “Determining how to integrate AI technology into workflows is a first step to changing how pathologists work on a day-to-day basis – and many organisations are implementing the traditional and digital workflows in parallel in order to optimise the benefits of modernising their pathology departments,” he told HIT Consultant.
For the patient – whether human or animal – AI in digital pathology can lead to more personalised, patient-focused treatments, by helping health professionals to analyse data quickly and make more informed conclusions about patients. This means that treatment can begin sooner, benefiting the patient’s recovery.
As nature.come notes, AI can also reduce errors in diagnosis and classification, again greatly improving patient outcomes. Worldwide machine learning-based program, The Camelyon Grand Challenge 2016, evaluated new algorithms for the automated detection of cancer in hematoxylin and eosin (H&E)-stained whole-slide imaging (WSI), and showed a 92.4% sensitivity in tumour detection rate, versus a 73.2% sensitivity achieved by a pathologist. “Computational pathology aims to improve diagnostic accuracy, optimise patient care, and reduce costs by bringing global collaboration,” the publication states.
Learn more about the AI in OvaCyte™ here or contact us.