Assessing quality in the modern histology laboratory: using image analysis and AI to develop next generation quality assurance programs


Histology and histochemistry are the technical basis for pathologic interpretation of disease in tissue. Our use of these techniques has driven 200 years of medical discovery. Thus, the quality of histology, histochemistry and the resultant product, the microscopic slide is recognized to be critical to quality pathology. Typically, such quality assessment has been the provenance of subjective interpretation by expert histologists and pathologists. For decades similarly subjective quality assurance programs have been used to assess the quality of histology of both routine staining and special stains such as immunohistochemistry (IHC).

With the advent of digital pathology, particularly whole slide imaging techniques, it now possible to assess histologic quality using image analysis and artificial intelligence algorithms. Leveraging machine learning and automation, this can be done at scale. To that end “tissue controls” and “phantoms” are now being developed to allow for assessment of the quality of histology processes within and between laboratories. Such quality assessment is of major importance in AI and image analysis where changes in histology quality and staining can have impact on algorithms. This is even more essential in AI based interpretation of biomarker analysis using classic IHC or C-ISH technology.

Given these advancements, it is becoming clear that subjective quality assessment of histology is insufficient particularly with IHC where very low limits of biologic activity have meaning. This is further complicated by the modern impact of pathologist interpretation. It is no longer simply a matter of providing a specific diagnosis. Rather, pathologic classification can determine therapeutic decisions and are indicative of prognosis. This is a game changer in the use of histochemistry and image analysis and our resultant application of AI.

This interactive, didactic session will address these critical issues through drawing on the speaker’s firsthand experiences in the diagnostic laboratory. Using the lens of quality assurance in anatomic pathology, the focus will be on surgical pathology, and the approaches to quality assurance in the histology production line. Leveraging this historic perspective, we will explore how quality management approaches will need to adapt and be improved upon to allow for effective, reliable whole slide imaging in support of image analysis and machine learning. There are now products on the market that support the routine analysis of histology quality leveraging tissues controls as ground truth. Experience is growing in their use that can be shared. Regulatory agencies are recognizing the significant impact of these approaches leading to a leveling up in quality assurance requirement, including new ISO standards. Finally, as part of the discussion we will draw on participant experiences regarding best practices and practical tips for implementation of digital pathology quality assurance programs.



  1. Describe the current state of histology quality assurance in the histology laboratory.
  2. Understand the impact of the quality of histology and histochemistry on digital pathology, whole slide imaging and artificial intelligence algorithms and the unique requirements when applied to IHC.
  3. Discuss approaches and best practices to applying image analysis and AI to develop next generation quality assurance programs.


Presented by:


J. Mark Tuthill, MD

Division Head of Pathology Informatics

Henry Ford Health


J. Mark Tuthill, MD, completed pathology residency and informatics fellowship training at the University of Vermont College of Medicine-Fletcher Allen Health Care, and created the department’s division of pathology informatics. Dr. Tuthill is Division Head of Pathology Informatics at Henry Ford Health System in Detroit. Areas of interest include digital pathology implementation, Internet applications for laboratory services, laboratory information systems, business analytics, electronic health records and informatics training and education. Active in organized medicine, he is an advisor to the ASCP Annual Meeting Steering Committee; Delegate, Wayne Medical Society; Co-director for the API’s Pathology Informatics Summit; and Delegate for CDC’s CLIAC committee. As a charter member of the Association for Pathology Informatics, Dr. Tuthill has worked for the API from its inception serving as president, chair of the membership committee, education committee, and the organization’s original planning group. Dr. Tuthill is the recent recipient of the API’s distinguished service award.