Bayesian Landmark-Based Shape Analysis of Tumor Pathology Images

Abstract

Medical imaging is a form of technology that has revolutionized the medical field over the past decades. Digital pathology imaging, which captures histological details at the cellular level, is rapidly becoming a routine clinical procedure for cancer diagnosis support and treatment planning. Recent developments in deep-learning methods have facilitated tumor region segmentation from pathology images. The traditional shape descriptors that characterize tumor boundary roughness at the anatomical level are no longer suitable. New statistical approaches to model tumor shapes are in urgent need. In this article, we consider the problem of modeling a tumor boundary as a closed polygonal chain. A Bayesian landmark-based shape analysis model is proposed. The model partitions the polygonal chain into mutually exclusive segments, accounting for boundary roughness. Our Bayesian inference framework provides uncertainty estimations on both the number and locations of landmarks, while outputting metrics that can be used to quantify boundary roughness. The performance of our model is comparable with that of a recently developed landmark detection model for planar elastic curves. In a case study of 143 consecutive patients with stage I to IV lung cancer, we demonstrated the heterogeneity of tumor boundary roughness derived from our model effectively predicted patient prognosis (p-value $<0.001$). Supplementary materials for this article are available online.

KEYWORDS:

• Artificial intelligence-reconstructed images
• Landmark detection
• Markov chain Monte Carlo
• Shape analysis
• Tumor boundary roughness

Supplementary Materials

The supplementary materials for Sections 2–5 are available online, including the derivation of the point-to-line distance, a detailed description of the MCMC algorithms, explicit definitions of evaluation metrics, reports on sensitivity analysis and scalability test, supplementary tables and figures from the lung case study, and an additional U.S. state shape case study. We provide software in the form of R/C++ code on GitHub (https://github.com/bougetsu/BayesLASA). To reproduce the figures presented in the paper, refer to the corresponding R scripts in the ‘manuscript_reproducibility’ directory at the same repository.

Acknowledgments

The authors would like to thank the editor, associate editor, and the three committed reviewers for their careful and constructive review. The authors’ thanks also go to Kevin C. Lutz and Kevin W. Jin for their help proofreading the paper.

Disclosure Statement

No potential conflict of interest was reported by the author(s).

Note

Notes

1 https://dabi.temple.edu/external/shape/MPEG7/dataset.html

Additional information

Funding

This work was supported by NSF DMS 2210912, NSF DMS 2113674, and NIH 1R01GM141519.