Main Menu

Science Writing Prize 2020 - Pushing the frontiers in onco-pathology: from microscopes to digital diagnostics

24
Feb
2021

Nivedita Ravindran examines cutting edge digital pathology in this shortlisted entry for the 2020 Mel Greaves Science Writing Prize. Nivedita is a Higher Scientific Officer in the Division of Breast Cancer Research at the Institute of Cancer Research.

Posted on 24 February, 2021 by Nivedita Ravindran

A close up of pathology slides as used in the lab

Not many people are aware of what a pathologist really does. Pathology is not all about autopsies and solving crimes, as popular media might lead us to believe. The vast majority of pathologists are doctors who spend hours looking down microscopes at thinly stained sections of a patient’s biopsy diagnosing diseases including a variety of cancers.

In fact, it is only the pathologist who can make a definite cancer diagnosis and identify its subtype – information which is crucial in guiding the treatment plan. The field of pathology has come a long way from the gross examination of tissues to using microscopes to look at cells, to advances in tissue staining and use of biomarkers.

Over the centuries, pathology has seamlessly adapted to new discoveries to enhance patient diagnosis. In this digital era, as in every other aspect of life, pathologists are faced with incorporating and adapting to the digital technologies available at hand. 

Moving away from microscopes – but what is the need?

The incidence of cancer increases with age and we humans are living longer than ever before. This is leading to an increasing caseload for pathologists. There is a growing demand for pathologists all over the world which is not met by recruitment.

Over the next few decades, progress in areas of cancer research such as biomarker discovery, precision medicine and molecular pathology will only add further to the profession’s workload.

Whole slide imaging systems or slide scanners can now scan numerous glass slides at a time and present them as digital images where the entire slide can be viewed as a whole. Pathologists can access these images for primary diagnosis on a computer system.

If executed with the correct approach, a digital workflow holds the capability to ease the pathologist’s work as opposed to the currently mainstream usage of microscopes to diagnose samples from patients. It is not just the availability of technology but also the benefits of its implementation for pathologists and ultimately the patients they serve that is pushing this digital revolution.

Digital pathology – scope and benefits

Around the world, many pathology labs are going digital, including a handful of NHS pathology departments in the UK. Quality digital images allow for ease of measurement of tumour size, digital annotation and parallel viewing of images with multiple biomarkers.

This will increase the scope for image analysis and help to make the process more efficient and less tedious for the pathologist. In addition, digital pathology allows for remote working especially during unforeseen circumstances such as the current Covid-19 pandemic. The Royal College of Pathologists has recognised these benefits and issued guidance to pathologists on use of digital pathology for clinical diagnosis. Images can be referred or shared with other pathologists for expert opinions without the hassle of shipping glass slides.

Digital pathology can aid student training with easier access to images than conventional multi-head microscopes. It also allows for training resources to be accessed remotely when required. In a wider perspective, there is now real scope to achieve a comprehensive view of a patient’s cancer by having virtual access to radiology, pathology and molecular data at the pathologist’s fingertips to aid diagnosis.

Digital pathology - the challenges

The challenge of implementing a digital workflow when there are decades-old set workplace practices in place cannot be underestimated. Radiologists faced a similar dilemma when radiology scans went digital at the beginning of the 21st century. 

The difficulties of accessing, analysing and archiving large volumes of digital data were resolved by integrating with IT infrastructure. Learning from the past, providers of imaging IT solutions could work together with pathology departments at all stages to achieve a smooth digital workflow.

High costs are associated with incorporating slide scanners and IT infrastructure to securely manage and store data. These costs could be recovered through the increased efficiency of the pathology department over the years. Pathologists will need to train themselves on the software and practise diagnosing digitally by validating with glass slides.

Once digital workflow has been implemented, the use of digital images alongside microscope referral will increase until pathologists are completely confident in using the technology.

Future of digital pathology - artificial intelligence (AI) systems

The implementation of digital pathology will allow images to be fed into deep learning neural networks to train them to recognise patterns and present a diagnosis to the pathologist. Any missed or mistaken diagnosis can then be fed back to the network for the AI to learn and correct itself.

AI could in future, with regulatory approval, help with prioritisation of cases, discovery of combinations of features which cannot be visually determined, and design and execution of cancer clinical trials. There is undeniable potential here and there are many research groups in the UK and around the world working on this. In the US, the Food and Drug Administration has approved a few AI-based algorithms in radiology for medical diagnosis. Recently, the company Paige.AI was granted FDA breakthrough device status for computational pathology-based clinical diagnosis starting with prostate cancer.

Although it might be inevitable further in the future, we are still in the first digital wave in pathology. Technology is here to stay but the need of the hour is to implement it wisely and efficiently in order to benefit the gold standard of cancer diagnosis – the field of pathology.

Tags

Mel Greaves breast cancer Science Writing Competition pathology
comments powered by Disqus