Search terms

Home
Philips
Computational pathology

What is computational
pathology?

Contact Us

Defining computational pathology


There are many definitions of Computational pathology however the one cited in the article “Computational Pathology: An Emerging Definition”1 Louis DN et al is a holistic way of looking at this space. The author defines computational pathology as an approach to diagnosis that incorporates multiple sources of data (e.g., pathology, radiology, clinical, molecular and lab operations); uses mathematical models to generate diagnostic inferences; and presents clinically actionable knowledge to customers.  This vision goes beyond an informatics-centric view and leverages the core competency of pathology and the ability to effectively communicate clinically actionable knowledge.

Contact information

* This field is mandatory
*
*
*
*
*
*
*
*
*
Previous
By specifying your reason for contact we will be able to provide you with a better service.
*
*

What does this mean?
Previous

What does computational pathology mean to the pathologists?

“In the future I envisage computational pathology tools enabling us to address the developing complex medical environment for cancer care”


–– Ian Ellis, Professor of Cancer Pathology, University of Nottingham

“Computational pathology solutions will help streamline workflow in the future by screening situations that do not require a pathologist review” 


–– Jeroen van der Laak , Associate Professor at Radboud University Medical Center - ‎Radboudumc

Need for computational pathology


We are moving rapidly into an era where next-generation pathology is becoming a reality with the advent of digital pathology. Paradigm shifts are being witnessed in cancer care with precision medicine and personalized treatments increasing by the day. Pathologists are absolutely central to this dream of personalized medicine. It is at the desk of the pathologist first clinical decisions regarding the patient is made and that will continue to be the case.²

Increasing workload

  • Cancer projections are growing
  • Number of tests applied are increasing
  • 44% of pathologists work overtime weekly.²
  • 24%  having to outsource services weekly.²

Shortage of pathologist

  • 10.4% decrease in active physicians pathology in 2008 – 20133
  • 60.7% of active physicians in pathology are age 55 or older3

What is the relevance today and how does this impact care delivery?


One of the main topics of conversation today in the treatment of cancer is the impact of precision medicine. The speed in which precision medicine will become affordable and as accurate as possible will only will become reality when institutions and departments begin to support the development of computational pathology.

 

Computational pathology was created to help improve the pathology ecosystem and go beyond to help a range of laboratory activities and goals, including:

Improving diagnostic accuracy
Optimizing patient care
Reducing costs by improving laboratory efficiencies 

Is the technology ready for this change?


The evolution of deep learning and the improvement in accuracy for image pattern recognition has been staggering in the last few years. The performance of ImageNet has been slowly improving and for the first time is meaningfully able to challenge humans with less than 5% error rates.4

A practical example is we are seeing trials of driver less cars potentially powered by deep learning technologies. There have been many reasons to think that digitizing images and using machine learning could help pathologists be faster, more accurate and make more accurate diagnoses for patients. This has been a big mission in the field of pathology for more than 30 years. But it’s been only recently that improved scanning, storage, processing and algorithms have made it possible to pursue this mission effectively.5

Related articles that talk about the potential of deep learning today
 

  1. Publication: Computational pathology of pre-treatment biopsies identifies lymphocyte density as a predictor of response to neoadjuvant chemotherapy in breast cancer
  2. Publication: Deep learning algorithm diagnoses skin cancer as well as seasoned dermatologists
  3. Publication: Artificial Intelligence Achieves Near-Human Performance in Diagnosing Breast Cancer
  4. Publication: First FDA Approval For Clinical Cloud-Based Deep Learning In Healthcare

How can computational pathology be used to improve the practice of pathology?

A central role for Pathologists

Pathologists taking more of a leadership position within their institutions is key to the success of computational pathology as it require deep changes within the lab.

Connecting the non-pathology department

To get the entire patient data the pathologist must network with several non-pathology departments to obtain the data required to make clinical decisions.

Sharing best practices
Improving standards and procedures within pathology departments to advance patient care. This has been the goal for many departments across the country, for both clinical or research. One of the best methods to improve is collaboration, this helps accelerate research and discovery.

1 – Louis DN et al, Computational pathology: an emerging definition, Arch Pathol Lab Med, Volume 138, Issue 9, 2014

2 - Sara Bainbridge et al.. (2016). TESTING TIMES TO COME?. Available: http://www.cancerresearchuk.org/sites/default/files/testing_times_to_come_nov_16_cruk.pdf

3 - Cancer Facts & Figures - Association of American Medical Colleges. (2014). Physician Specialty Data Book. Available: https://members.aamc.org/eweb/upload/Physician%20Specialty%20Databook%202014.pdf

4 – Kaiming He et al. (2015). Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification

5 - Louis DN et al, Computational Pathology. A Path Ahead, Arch Pathol Lab Med, Vol 140,  2016

By clicking on the link, you will be leaving the official Royal Philips Healthcare ("Philips") website. Any links to third-party websites that may appear on this site are provided only for your convenience and in no way represent any affiliation or endorsement of the information provided on those linked websites. Philips makes no representations or warranties of any kind with regard to any third-party websites or the information contained therein.

I understand

By clicking on the link, you will be leaving the official Royal Philips Healthcare ("Philips") website. Any links to third-party websites that may appear on this site are provided only for your convenience and in no way represent any affiliation or endorsement of the information provided on those linked websites. Philips makes no representations or warranties of any kind with regard to any third-party websites or the information contained therein.

I understand