Interpretability-Driven Sample Selection Using Self Supervised Learning For Disease Classification And Segmentation.

Mahapatra, Dwarikanath; Poellinger, Alexander; Shao, Ling; Reyes, Mauricio (2021). Interpretability-Driven Sample Selection Using Self Supervised Learning For Disease Classification And Segmentation. (In Press). IEEE transactions on medical imaging, PP Institute of Electrical and Electronics Engineers IEEE 10.1109/TMI.2021.3061724

[img] Text
P_llinger_Interpretability-driven.pdf - Published Version
Restricted to registered users only
Available under License Publisher holds Copyright.

Download (12MB) | Request a copy

In supervised learning for medical image analysis, sample selection methodologies are fundamental to attain optimum system performance promptly and with minimal expert interactions (e.g. label querying in an active learning setup). In this paper we propose a novel sample selection methodology based on deep features leveraging information contained in interpretability saliency maps. In the absence of ground truth labels for informative samples, we use a novel self supervised learning based approach for training a classifier that learns to identify the most informative sample in a given batch of images. We demonstrate the benefits of the proposed approach, termed Interpretability-Driven Sample Selection (IDEAL), in an active learning setup aimed at lung disease classification and histopathology image segmentation. We analyze three different approaches to determine sample informativeness from interpretability saliency maps: (i) an observational model stemming from findings on previous uncertainty-based sample selection approaches, (ii) a radiomics-based model, and (iii) a novel data-driven self-supervised approach. We compare IDEAL to other baselines using the publicly available NIH chest X-ray dataset for lung disease classification, and a public histopathology segmentation dataset (GLaS), demonstrating the potential of using interpretability information for sample selection in active learning systems. Results show our proposed self supervised approach outperforms other approaches in selecting informative samples leading to state of the art performance with fewer samples.

Item Type:

Journal Article (Original Article)

Division/Institute:

10 Strategic Research Centers > ARTORG Center for Biomedical Engineering Research
04 Faculty of Medicine > Department of Radiology, Neuroradiology and Nuclear Medicine (DRNN) > Institute of Diagnostic, Interventional and Paediatric Radiology

UniBE Contributor:

Pöllinger, Alexander and Reyes, Mauricio

Subjects:

500 Science > 570 Life sciences; biology
600 Technology > 610 Medicine & health

ISSN:

0278-0062

Publisher:

Institute of Electrical and Electronics Engineers IEEE

Language:

English

Submitter:

Maria de Fatima Henriques Bernardo

Date Deposited:

27 Apr 2021 16:00

Last Modified:

28 Apr 2021 01:34

Publisher DOI:

10.1109/TMI.2021.3061724

PubMed ID:

33625979

BORIS DOI:

10.48350/154465

URI:

https://boris.unibe.ch/id/eprint/154465

Actions (login required)

Edit item Edit item
Provide Feedback