-Author name in bold denotes the presenting author
-Asterisk * with author name denotes a Non-ASH member
Clinically Relevant Abstract denotes an abstract that is clinically relevant.

PhD Trainee denotes that this is a recommended PHD Trainee Session.

Ticketed Session denotes that this is a ticketed session.

2841 Deep Learning for Morphology-Based, Bone Marrow Cell Classification

Program: Oral and Poster Abstracts
Session: 613. Acute Myeloid Leukemias: Clinical and Epidemiological: Poster II
Hematology Disease Topics & Pathways:
Research, Translational Research
Sunday, December 10, 2023, 6:00 PM-8:00 PM

Shenghuan Sun1*, Jacob Cleave2*, Linlin Wang3*, Fabienne Lucas4*, Laura Brown5*, Jacob Spector6*, Leonardo Boiocchi2*, Jeeyeon Baik, BA, MPH7*, Menglei Zhu2*, Orly Ardon2*, Chuanyi M. Lu, MD, PhD8, Ahmet Dogan, MD, PhD7, Dmitry Goldgof9*, Iain Carmichael10*, Sonam Prakash5*, Atul Butte5* and Gregory Mark Goldgof11*

1University of California, San Francisco, San Francisco, CA
2Memorial Sloan Kettering Cancer Center, New York, NY
3UCSF, San Francisco
4Harvard Medical School, Boston
5University of California, San Francisco, San Francisco
6Boston Children's Hospital, Boston, MA
7Department of Pathology and Laboratory Medicine, Hematopathology Service, Memorial Sloan Kettering Cancer Center, New York, NY
8United States Department of Veterans Affairs, San Francisco, CA
9University of South Florida, Tampa, FL
10University of California, Berkeley, Berkeley, CA
11Memorial Sloan Kettering Cancer Center, Yew York

The morphological classification of cells in bone marrow aspirate (BMA) is central to the diagnosis of hematologic diseases, including leukemias. Despite being a critical task, its monotonous, time-consuming nature and dependency on highly skilled clinical experts makes it prone to human error. Such errors can lead to delays and misdiagnoses that negatively impact patient care. To counter these challenges, we curated an expansive dataset of more than 40,000 hematopathologist consensus-annotated single-cell images, extracted from BMA whole slide images (WSIs), each annotated into one of 23 distinct morphologic classes.

We then utilized this data to develop DeepHeme, a convolutional neural network classifier designed for bone marrow cell typing tasks. DeepHeme achieves state-of-the-art performance in both the breadth of differentiable classes and accuracy across these classes. By comparing its performance to that of individual hematopathologists from three premier academic medical centers, using our gold standard consensus-labelled images, we found our AI algorithm either matched or surpassed the average performance across all classes. In addition, we integrated DeepHeme with internally developed region classifier and cell detection algorithms, culminating in a comprehensive diagnostic pipeline for whole slide cell differential.

We next tested DeepHeme on slides from an external hospital system at a major cancer center to evaluate the generalizability of our model, a necessary precondition to widespread application. DeepHeme demonstrated a high level of generalizability, evidenced by a decrease of only 4% in the mean F-1 score, from 0.89 to 0.85, across all 23 cell classes. Lastly, to improve access to the DeepHeme algorithm results and encourage further real-world generalizability testing, we developed a web application that allows scientists and clinicians to test the DeepHeme algorithm on either test images from our study or their own user-uploaded aspirates.

Disclosures: Baik: Pauling.AI: Current Employment. Dogan: Seattle Genetics: Consultancy; Physicians' Education Resource: Consultancy, Honoraria; EUSA Pharma: Consultancy; Loxo: Consultancy; Peer View: Honoraria; Incyte: Consultancy; Takeda: Other: Research Funding; Roche: Other: Research Funding.

*signifies non-member of ASH