Robust, Primitive, and Unsupervised Quality Estimation for Segmentation Ensembles

Frontiers in Neuroscience

Carolin M. Pirkl¹  Benedikt Wiestler³  Malek El Husseini⁴  Bjoern H. Menze⁴  Suprosanna Shit⁴  Johannes C. Paetzold⁴  Claus Zimmer⁵  Sarthak Pati⁷  Spyridon Bakas⁷  Lucas Fidon⁸  Florian Kofler⁹  Ivan Ezhov⁹  Fernando Navarro⁹  Jan Kirschke⁹  Egon Burian⁹ 

[1] Imaging Sciences, King's College London, London, United Kingdom;Center for Biomedical Image Computing and Analytics, University of Pennsylvania, Pennsylvania, PA, United States;Department of Diagnostic and Interventional Neuroradiology, School of Medicine, Klinikum rechts der Isar, Technical University of Munich, Munich, Germany;Department of Informatics, Technical University Munich, Munich, Germany;Department of Pathology and Laboratory Medicine, Perelman School of Medicine, University of Pennsylvania, Pennsylvania, PA, United States;Department of Radio Oncology and Radiation Therapy, School of Medicine, Klinikum rechts der Isar, Technical University of Munich, Munich, Germany;Department of Radiology, Perelman School of Medicine, University of Pennsylvania, Pennsylvania, PA, United States;;School of Biomedical Engineering &TranslaTUM - Central Institute for Translational Cancer Research, Technical University of Munich, Munich, Germany;

关键词: quality estimation;    failure prediction;    anomaly detection;    ensembling;    fusion;    OOD;   

DOI  :  10.3389/fnins.2021.752780

来源: DOAJ

【 摘 要 】

A multitude of image-based machine learning segmentation and classification algorithms has recently been proposed, offering diagnostic decision support for the identification and characterization of glioma, Covid-19 and many other diseases. Even though these algorithms often outperform human experts in segmentation tasks, their limited reliability, and in particular the inability to detect failure cases, has hindered translation into clinical practice. To address this major shortcoming, we propose an unsupervised quality estimation method for segmentation ensembles. Our primitive solution examines discord in binary segmentation maps to automatically flag segmentation results that are particularly error-prone and therefore require special assessment by human readers. We validate our method both on segmentation of brain glioma in multi-modal magnetic resonance - and of lung lesions in computer tomography images. Additionally, our method provides an adaptive prioritization mechanism to maximize efficacy in use of human expert time by enabling radiologists to focus on the most difficult, yet important cases while maintaining full diagnostic autonomy. Our method offers an intuitive and reliable uncertainty estimation from segmentation ensembles and thereby closes an important gap toward successful translation of automatic segmentation into clinical routine.

【 授权许可】

Unknown