Monday, May 22, 2017 | 5:00-7:00 PM
Wyndham Grand Pittsburgh Hotel | King’s Garden 2/3
Moderator: Anil Parwani, MD, PhD, MBA
Quantitative Image Analysis for HER2 Immunohistochemistry for Breast Cancer: Guideline Update and - Review of Draft Recommendations by Dr. Marilyn Bui
Advancements in genomics, computing and imaging technology have spurred new opportunities to use quantitative image analysis for diagnostic testing. Quantitative image analysis has been shown to improve consistency and accuracy of interpretation than manual scoring by pathologists, but has not gained widespread acceptance. One practical gap is the lack of guidelines in how to perform QIA. The College of American Pathologists (CAP) convened a panel of pathologists and histotechnologists with expertise in digital pathology, immunohistochemistry, and quality management to develop an evidence-based guideline that provides recommendations for improving reproducibility, precision, and accuracy in the interpretation of HER2 immunohistochemical (IHC) where quantitative image analysis (QIA) is employed. This presentation will provide an update of the guideline including the 11 draft recommendations and the feedback received during the open comment period. The guideline aims to ensure that diagnoses of HER2 immunohistochemistry for breast cancer are accurate and consistent. Such accuracy is essential to patient diagnosis, prognosis and treatment planning. The CAP Pathology and Laboratory Quality Center (The Center) develops evidence-based guidelines and consensus statements related to the practice of pathology and laboratory medicine. Through this work, the CAP and its members continually improve the quality of diagnostic medicine and patient outcomes.
Primary Diagnosis and Regulatory/DICOM Landscape and How it will Impact the use of Digital Images in Pathology by Dr. Liron Pantanowitz
Several pathology laboratories around the world have started going fully digital. However, for many labs the regulatory, financial and cultural barriers to WSI adoption for primary diagnosis remain a challenge. In the USA, the regulatory field for digital pathology has advanced considerably. The Digital Pathology Association (DPA) had a major role in this accomplishment. This talk will address the past and emerging regulatory environment of WSI for clinical use in the United States. As digital pathology becomes a reality, interoperability standards such as DICOM are being emphasized in the industry. However, should pathology imaging vendors manufacturing proprietary systems offer their clients DICOM-compliant images? This talk will accordingly also address the emerging era of enterprise imaging in healthcare and its impact on digital pathology vendors.
Detailed Critique of a Whole Slide Imaging Solution, Hardware and Software used in a Network Environment by Dr. Mark Tuthill
This session will be a dive into operation, management and utilization of the deployed vendor solutions at Henry Ford Health System. Addressed will be both the hardware scanning operation as well as the software used for managing users and case viewing. Special attention will be paid to training, day to day operations and system deployment as well as perceived strengths and weaknesses of the system as deployed.
Advancing computational Pathology: the CAMELYON challenges by Dr. Jeroen van der Laak
Deep learning (DL) based pattern recognition has shown enormous potential in many different fields. An important requirement for developing these algorithms is the availability of a sufficient amount of labeled training data. Until recently, many research groups and companies excelling in deep learning did not address applications in histopathology . One reason is the unfamiliarity with this field and its most urgent needs. Also, acquiring sufficient high-quality data is an important limitation that withholds development of DL for histopathology. Organizing the CAMELYON16 challenge, we aimed to tackle these shortcomings by posing a clinically relevant problem and providing sufficient, high quality data to train meaningful algorithms. The results of the challenge were remarkable: many researchers and companies all over the world spent significant time in developing algorithms for a diagnostically relevant task in pathology: identifying metastases in sentinel lymph node tissue sections of breast cancer patients. The top performing algorithms were capable of performing this task at the level of an expert pathologist. As a follow-up, at this time CAMELYON17 is being organized, offering almost 3TBytes of image data to interested groups and companies all around the world. CAMELYON17 will bring computer aided diagnosis one step closer to clinical implementation, by requiring participants to combine information from multiple sections into a single outcome.
Marilyn M. Bui, MD, PhD
Dr. Bui is certified in Anatomic/Clinical Pathology and Cytopathology by American Board of Pathology. She is the Scientific Director of Analytic Microscopy Core at Moffitt Cancer Center (MCC) and the Director of Cytopathology Fellowship Program at the University of South Florida Morsani College of Medicine. She is a senior member and a professor practicing at MCC, a premier cancer center and the only NCI/NIH designated comprehensive cancer center and research institute in Florida. She has published over 160 peer-reviewed articles, 15 book chapters and 2 books. She has earned national and institutional awards in cancer research, education and service. Dr. Bui’s expertise in digital pathology (DP) is evidenced by her publication and presentation addressed its clinical, teaching and image analysis utility. She is an executive board member of Digital Pathology Association, a member of the College of American Pathologists (CAP) Digital Pathology Committee, and an editorial board member of Journal of Pathology Informatics. She chairs the expert panel of the CAP HER2 Immunohistochemistry Quantitative Image Analysis Guidelines Committee.
Liron Pantanowitz, MD
Dr. Pantanowitz is the Director of Cytopathology at UPMC Shadyside. He is also the Director of the Pathology Informatics Fellowship Program and the Director of the Pathology Informatics Division.
Mark Tuthill, MD
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 departments division of pathology informatics. Dr. Tuthill is Division Head of Pathology Informatics at Henry Ford Health System in Detroit. Areas of interest include digital imaging, Internet applications for laboratory services, laboratory information systems, laboratory outreach technology solutions, electronic health records and informatics training and education. Active in organized medicine, he is an advisor to the ASCP Annual Meeting Steering Committee, Wayne County District Director for the MSMS, President-Elect of the Sunquest User Group, and Director for the annual Pathology Informatics Summit. As a charter member of the Association for Pathology Informatics, Dr. Tuthill has worked for the API from its inception serving as president, chairman of the membership committee, education committee member, and the organizations original planning group; currently Chair of the Program Committee.
Jeroen van der Laak, PhD
Jeroen van der Laak is senior researcher and group leader at the department of Pathology of the Radboud University Medical Center. He leads a research group in digital pathology, which is part of the diagnostic image analysis group (DIAG) of the Radboud UMC Radiology department. Research focusses on several aspects of digital pathology: efficiency improvement of current diagnostic practice by using computer aided diagnosis algorithms; objective quantification of immunohistochemical markers; and computational pathology where new imaging biomarkers are researched. The group specializes in development and validation of deep learning algorithms and is internationally among the leading research groups in this exiting field. Jeroen has an MSc in computer science and acquired his PhD from the Radboud University in Nijmegen. He co-authored over 80 peer-reviewed publications, is member of the editorial board of the Journal of Pathology Informatics and organizer of sessions at the European Congress of Pathology and Pathology Visions. In 2016, Jeroen coordinated the highly successful Camelyon16 grand challenge. Jeroen acquired research grants from the European Union and the Dutch Cancer Society, among others.