PhD defence Silas Nyboe Ørting
Assessing emphysema in CT scans of the lungs using machine learning, crowdsourcing and visual similarity.
Date: 26 April 2019, at 13.15
Place: Auditorium 04 – HCØ, Universitetsparken 5, 2100 Kbh. Ø
Emphysema is a pathology in chronic obstructive pulmonary disease (COPD) characterized by destruction of lung tissue. The extent and appearance of emphysema can be assessed in CT scans of the lungs. Current recommendations for assessing emphysema in CT scans is to use a combination of densitometry and visual assessment.
Current densitometry methods are vulnerable to variation in scanners, scan protocols and software implementations, and cannot characterize emphysema patterns. Visual assessment requires expert knowledge, is time-consuming and commonly only provides semi-quantitative estimates of emphysema extent.
Machine learning methods that learn from visual assessment could combine the benefits of densitometry and visual assessment, to provide fully automated assessment of emphysema extent and patterns. One of the main issues when applying supervised machine learning in medical image analysis is obtaining labels.
This thesis investigates three approaches to reducing the need for labels when training machine learning methods to assess emphysema: weakly supervised learning, crowdsourcing and learning from visual similarity. Weakly supervised machine learning aims at learning from global labels instead of local labels, for example learning from image labels instead of pixel labels. The thesis investigates emphysema detection and quantification within two weakly supervised learning settings, multiple instance learning, where labels are binary, and learning with label proportions, where labels are proportions. Crowdsourcing aims at reducing labeling costs by replacing expert annotators with non-experts.
The thesis provides a survey of how crowdsourcing is used in medical imaging, as well as an investigation into how emphysema assessment can be framed as a task that can be solved by non-expert crowd workers. Learning from visual similarity aims at learning representations from relative comparisons of images. The thesis investigates how visual similarity assessments can be obtained and used for training convolutional neural networks to learn representations of chest CT scans.
Associate Professor, Erik Dam,Department of Computer Science, University of Copenhagen (Head of committee)
Associate Professor, Elsa Angelini, Imperial College London
Associate Professor, Georg Langs, Medical University of Vienna
Professor Marleen de Bruijne, Department of Computer Science, University of Copenhagen
Assistant Professor, Jens Petersen, Department of Computer Science, University of Copenhagen
For an electronic copy of the thesis, please contact email@example.com