Managing Expectations with Digital Pathology
Posted 19th June 2017 by Jane Williams
Anyone reading this would be familiar that in April the FDA determined after an extensive clinical trial that a whole slide imaging system presents no greater risk to patient care than an optical microscope. In a purely scientific study design, there is a null hypothesis that states “there is no significant difference between…” and then investigators try to prove or disprove that hypothesis.
One example of this scenario might be that “Drug A has the same ability to lower blood pressure as Drug B.” If you can reject the null hypothesis that Drug A has a greater ability to lower blood pressure than Drug B AND there are no increased side effects, contraindications, etc…it is likely a manufacturer will bring this product to market with appropriate regulatory controls.
My seventh-grade science project looked at what evaporates faster – milk or water. It was the best I could do after putting the project off until the night before the assignment was due. I didn’t have time to grow plants in red light and blue light and white light and see if there was a difference in plant height over a month. My hypothesis was that water and milk, left in a dish, at room temperature in the same room, would evaporate at the same rate. It turned out the drop of water evaporated faster during a St. Elsewhere marathon on TV than the milk did, and I argued in my conclusions that proteins in the milk caused a delay in the rate of evaporation of the milk. The teacher didn’t buy the design, results or conclusion. I received a grade of “D” and was not invited to the Science Fair. However, watching St. Elsewhere for those many hours on TV convinced me I should go to medical school. I just needed to learn how to not procrastinate and tighten up my scientific approaches.
The FDA clearance affirmed what many have thought for many years. That is, digital pathology is equivalent to the light microscope. Give a well-meaning, well-intentioned pathologist a microscope, we are seldom wrong and never in doubt. Give a well-meaning, well-intentioned pathologist a digital image, we are seldom wrong and never in doubt.
Rather than showing a statistical difference between the light microscope and digital pathology, what was demonstrated in the study was that there was no difference when viewing the same case by the same pathologist.
But is digital pathology better than the light microscope? Better than providing a diagnostically suitable image to the back of our retinas and occipital lobes for pattern recognition than conventional optical microscopes?
No study has really ever shown that we should all sell our microscopes.
Yet, many recent headlines suggest we will increase cancer detection, eliminate errors, work better, faster, cheaper and more accurately while glued to computer monitors rather than chained to microscopes, and the health of the public will be improved dramatically.
If this were the case, public hearings on the riskiness of the light microscope would have already been held on Capitol Hill as a threat to the patient public.
What has been affirmed by regulators is that there is no difference in these technologies, not that one is necessarily superior to the other. Other large, US and UK-based studies have shown this to be the case for several organ systems/tissue types/diagnoses.
The real value of digital pathology is in the added value – eliminating geographic constraints where you can review cases from a lab at a distance, or consult with colleagues in real-time without moving the slide(s), or archive and later review material more efficiently than with traditional slide storage and courier models.
While we now have a system for primary diagnosis, and others will soon follow to provide their solutions to the marketplace, we are not going to significantly alter the professional practice of pathology because the modality in which we look at slides has changed. Both technologies will always have their limitations, and as practitioners and vendors it is necessary that we manage those expectations.
Digital pathology will not replace the light microscope. It is not “The End of the Microscope” as some recent headlines have questioned. Nor is it the end of pathologists.
Time is running out to book your place at the Digital Pathology Congress: USA. Register now.
Leave a Reply