Building robust pathology image analyses with uncertainty quantification

Jeremias Gomes; Jun Kong; Tahsin Kurc; Alba C M A Melo; Renato Ferreira; Joel H Saltz; George Teodoro

doi:10.1016/j.cmpb.2021.106291

Building robust pathology image analyses with uncertainty quantification

Comput Methods Programs Biomed. 2021 Sep:208:106291. doi: 10.1016/j.cmpb.2021.106291. Epub 2021 Jul 24.

Authors

Jeremias Gomes¹, Jun Kong², Tahsin Kurc³, Alba C M A Melo¹, Renato Ferreira⁴, Joel H Saltz⁵, George Teodoro⁶

Affiliations

¹ Department of Computer Science, University of Brasília, Brasília, Brazil.
² Biomedical Informatics Department, Emory University, Atlanta, USA; Department of Biomedical Engineering, Emory-Georgia Institute of Technology, Atlanta, USA; Department of Mathematics and Statistics, Georgia State University, Atlanta, USA.
³ Biomedical Informatics Department, Stony Brook University, Stony Brook, USA; Scientific Data Group, Oak Ridge National Laboratory, Oak Ridge, USA.
⁴ Department of Computer Science, Universidade Federal de Minas Gerais, Belo Horizonte, Brazil.
⁵ Biomedical Informatics Department, Stony Brook University, Stony Brook, USA.
⁶ Department of Computer Science, University of Brasília, Brasília, Brazil; Biomedical Informatics Department, Stony Brook University, Stony Brook, USA; Department of Computer Science, Universidade Federal de Minas Gerais, Belo Horizonte, Brazil. Electronic address: teodoro@dcc.ufmg.br.

PMID: 34333205
DOI: 10.1016/j.cmpb.2021.106291

Abstract

Background and objective: Computerized pathology image analysis is an important tool in research and clinical settings, which enables quantitative tissue characterization and can assist a pathologist's evaluation. The aim of our study is to systematically quantify and minimize uncertainty in output of computer based pathology image analysis.

Methods: Uncertainty quantification (UQ) and sensitivity analysis (SA) methods, such as Variance-Based Decomposition (VBD) and Morris One-At-a-Time (MOAT), are employed to track and quantify uncertainty in a real-world application with large Whole Slide Imaging datasets - 943 Breast Invasive Carcinoma (BRCA) and 381 Lung Squamous Cell Carcinoma (LUSC) patients. Because these studies are compute intensive, high-performance computing systems and efficient UQ/SA methods were combined to provide efficient execution. UQ/SA has been able to highlight parameters of the application that impact the results, as well as nuclear features that carry most of the uncertainty. Using this information, we built a method for selecting stable features that minimize application output uncertainty.

Results: The results show that input parameter variations significantly impact all stages (segmentation, feature computation, and survival analysis) of the use case application. We then identified and classified features according to their robustness to parameter variation, and using the proposed features selection strategy, for instance, patient grouping stability in survival analysis has been improved from in 17% and 34% for BRCA and LUSC, respectively.

Conclusions: This strategy created more robust analyses, demonstrating that SA and UQ are important methods that may increase confidence digital pathology.

Keywords: Microscopy; Sensitivity analysis; Survival analysis; Uncertainty quantification; Whole slide image analysis.

MeSH terms

Humans
Image Processing, Computer-Assisted*
Uncertainty