Considerations for data acquisition and modeling strategies: Mitosis detection in computational pathology Conference Paper

  


Authors: Ji, Z.; Rosenfield, P.; Eng, C. H.; Bettigole, S. E.; Gibson, D. C.; Masoudi, H.; Hanna, M. G.; Fusi, N.; Severson, K. A.
Title: Considerations for data acquisition and modeling strategies: Mitosis detection in computational pathology
Conference Title: Medical Imaging with Deep Learning (MIDL 2023)
Abstract: Preparing data for machine learning tasks in health and life science applications requires decisions that affect the cost, model properties and performance. In this work, we study the implication of data collection strategies, focusing on a case study of mitosis detection. Specifically, we investigate the use of expert and crowd-sourced labelers, the impact of aggregated vs single labels, and the framing of the problem as either classification or object detection. Our results demonstrate the value of crowd-sourced labels, importance of uncertainty quantification, and utility of negative samples. © 2023 CC-BY 4.0, Z. Ji et al.
Keywords: breast cancer; pathology; health science; data acquisition; mitosis detection; machine-learning; computational pathology; crowdsourcing; object detection; mitosis detections; life-sciences; science applications; acquisition strategies; learning tasks; modelling strategies
Journal Title Proceedings of Machine Learning Research
Volume: 227
Conference Dates: 2023 Jul 10-12
Conference Location: Nashville, TN
ISBN: 2640-3498
Publisher: Journal Machine Learning Research  
Date Published: 2023-01-01
Start Page: 1051
End Page: 1066
Language: English
PROVIDER: scopus
DOI/URL:

https://www.scopus.com/inward/record.uri?eid=2-s2.0-85189305525&partnerID=40&md5=aab9bbad019e75e825b0fca0c5151a85
Notes: Conference paper -- Source: Scopus