S LABORATORY INVESTIGATION THE BASIC AND TRANSLATIONAL PATHOLOGY RESEARCH JOURNAL LI VIRTUAL and INTERACTIVE MARCH 13 -18, 2021 NEVER STOP LEARNING 2021 USCAP 110TH ANNUAL MEETING VOLUME 101 | SUPPLEMENT 1 | MARCH 2021 (238-270) DERMATOPATHOLOGY 2021 ABSTRACTS | PLATFORM & POSTER PRESENTATIONS To cite abstracts in this publication, please use the following format: Author A, Author B, Author C, et al. Abstract title (abs#). In “File Title.” Laboratory Investigation 2021; 101 (suppl 1): page# EDUCATION COMMITTEE ABSTRACT REVIEW BOARD Jason L. Hornick ChairREVIEW BOARD Jason L. Hornick Chair Rhonda K. Yantiss, Chair Abstract Review Board and Assignment Committee Kristin C. Jensen Chair, CME Subcommittee Laura C. Collins Interactive Microscopy Subcommittee Raja R. Seethala Short Course Coordinator Ilan Weinreb Subcommittee for Unique Live Course Offerings Benjamin Adam Rouba Ali-Fehmi Daniela Allende Ghassan Allo Isabel Alvarado-Cabrero Catalina Amador Tatjana Antic Roberto Barrios Rohit Bhargava Luiz Blanco Jennifer Boland Alain Borczuk Elena Brachtel Marilyn Bui Eric Burks Shelley Caltharp Wenqing (Wendy) Cao Barbara Centeno Joanna Chan Jennifer Chapman Yunn-Yi Chen Hui Chen Wei Chen Sarah Chiang Nicole Cipriani Beth Clark Alejandro Contreras Claudiu Cotta Jennifer Cotter Sonika Dahiya Farbod Darvishian Jessica Davis Heather Dawson Elizabeth Demicco Katie Dennis Anand Dighe Suzanne Dintzis Michelle Downes Charles Eberhart Andrew Evans Julie Fanburg-Smith Michael Feely Dennis Firchau Gregory Fishbein Andrew Folpe Larissa Furtado Billie Fyfe-Kirschner Giovanna Giannico Christopher Giffith Anthony Gill Paula Ginter Tamar Giorgadze Purva Gopal Abha Goyal Rondell Graham Alejandro Gru Nilesh Gupta Mamta Gupta Gillian Hale Suntrea Hammer Malini Harigopal Douglas Hartman Kammi Henriksen John Higgins Mai Hoang Aaron Huber Doina Ivan Wei Jiang Vickie Jo Dan Jones Kirk Jones Neerja Kambham Dipti Karamchandani Nora Katabi Darcy Kerr Francesca Khani Joseph Khoury Rebecca King Veronica Klepeis Christian Kunder Steven Lagana Keith Lai Michael Lee Cheng-Han Lee Madelyn Lew Faqian Li Ying Li Haiyan Liu Xiuli Liu Lesley Lomo Tamara Lotan Sebastian Lucas Anthony Magliocco Kruti Maniar Brock Martin Emily Mason David McClintock Anne Mills Richard Mitchell Neda Moatamed Sara Monaco Atis Muehlenbachs Bita Naini Dianna Ng Tony Ng Michiya Nishino Scott Owens Jacqueline Parai Avani Pendse Peter Pytel Stephen Raab Stanley Radio Emad Rakha Robyn Reed Michelle Reid Natasha Rekhtman Jordan Reynolds Andres Roma Lisa Rooper Avi Rosenberg Esther (Diana) Rossi Souzan Sanati Gabriel Sica Alexa Siddon Deepika Sirohi Kalliopi Siziopikou Maxwell Smith Adrian Suarez Sara Szabo Julie Teruya-Feldstein Khin Thway Rashmi Tondon Jose Torrealba Gary Tozbikian Andrew Turk Evi Vakiani Christopher VandenBussche Paul VanderLaan Hannah Wen Sara Wobker Kristy Wolniak Shaofeng Yan Huihui Ye Yunshin Yeh Anjana Yeldandi Gloria Young Lei Zhao Minghao Zhong Yaolin Zhou Hongfa Zhu David B. Kaminsky (Ex-Officio)

Although melanoma occurs more rarely than several other skin cancers, patients' long term survival rate is extremely low if the diagnosis is missed. Diagnosis is complicated by a high discordance rate among pathologists when distinguishing between melanoma and benign melanocytic lesions. A tool that allows pathology labs to sort and prioritize melanoma cases in their workflow could improve turnaround time by prioritizing challenging cases and routing them directly to the appropriate sub-specialist. We present a pathology deep learning system (PDLS) that performs hierarchical classification of digitized whole slide image (WSI) specimens into six classes defined by their morphological characteristics, including classification of "Melanocytic Suspect" specimens likely representing melanoma or severe dysplastic nevi. We trained the system on 7,685 images from a single lab (the reference lab), including the the largest set of triple-concordant melanocytic specimens compiled to date, and tested the system on 5,099 images from two distinct validation labs. We achieved Area Underneath the ROC Curve (AUC) values of 0.93 classifying Melanocytic Suspect specimens on the reference lab, 0.95 on the first validation lab, and 0.82 on the second validation lab. We demonstrate that the PDLS is capable of automatically sorting and triaging skin specimens with high sensitivity to Melanocytic Suspect cases and that a pathologist would only need between 30% and 60% of the caseload to address all melanoma specimens.

Standard of care diagnostic procedure for suspected skin cancer is microscopic examination of hematoxylin & eosin stained tissue by a pathologist. Areas of high inter-pathologist discordance and rising biopsy rates necessitate higher efficiency and diagnostic reproducibility. We present and validate a deep learning system which classifies digitized dermatopathology slides into 4 categories. The system is developed using 5,070 images from a single lab, and tested on an uncurated set of 13,537 images from 3 test labs, using whole slide scanners manufactured by 3 different vendors. The system’s use of deep-learning-based confidence scoring as a criterion to consider the result as accurate yields an accuracy of up to 98%, and makes it adoptable in a real-world setting. Without confidence scoring, the system achieved an accuracy of 78%. We anticipate that our deep learning system will serve as a foundation enabling faster diagnosis of skin cancer, identification of cases for specialist review, and targeted diagnostic classifications.

Tomosynthesis, i.e. reconstruction of 3D volumes using projections from a limited perspective is a classical inverse, ill-posed or under constrained problem. Data insufficiency leads to reconstruction artifacts that vary in severity depending on the particular problem, the reconstruction method and also on the object being imaged. Machine learning has been used successfully in tomographic problems where data is insufficient, but the challenge with machine learning is that it introduces bias from the learning dataset. A novel framework to improve the quality of the tomosynthesis reconstruction that limits the learning dataset bias by maintaining consistency with the observed data is proposed. Convolutional Neural Networks (CNN) are embedded as regularizers in the reconstruction process to introduce the expected features and characterstics of the likely imaged object. The minimization of the objective function keeps the solution consistent with the observations and limits the bias introduced by the machine learning regularizers, improving the quality of the reconstruction. The proposed method has been developed and studied in the specific problem of Cone Beam Tomosynthesis Flouroscopy (CBT-fluoroscopy)1 but it is a general framework that can be applied to any image reconstruction problem that is limited by data insufficiency.