Conference proceeding
A deep (learning) dive into visual search behaviour of breast radiologists
Progress in Biomedical Optics and Imaging - Proceedings of SPIE, Vol.10577, pp.1057708-1057708-11
03/07/2018
DOI: 10.1117/12.2293366
Abstract
Visual search, the process of detecting and identifying objects using the eye movements (saccades) and the foveal vision, has been studied for identification of root causes of errors in the interpretation of mammography. The aim of this study is to model visual search behaviour of radiologists and their interpretation of mammograms using deep machine learning approaches. Our model is based on a deep convolutional neural network, a biologically-inspired multilayer perceptron that simulates the visual cortex, and is reinforced with transfer learning techniques. Eye tracking data obtained from 8 radiologists (of varying experience levels in reading mammograms) reviewing 120 two-view digital mammography cases (59 cancers) have been used to train the model, which was pre-trained with the ImageNet dataset for transfer learning. Areas of the mammogram that received direct (foveally fixated), indirect (peripherally fixated) or no (never fixated) visual attention were extracted from radiologists’ visual search maps (obtained by a head mounted eye tracking device). These areas, along with the radiologists’ assessment (including confidence of the assessment) of suspected malignancy were used to model: 1) Radiologists’ decision; 2) Radiologists’ confidence on such decision; and 3) The attentional level (i.e. foveal, peripheral or none) obtained by an area of the mammogram. Our results indicate high accuracy and low misclassification in modelling such behaviours.
Details
- Title: Subtitle
- A deep (learning) dive into visual search behaviour of breast radiologists
- Creators
- Suneeta Mall - The Univ. of Sydney (Australia)Patrick C Brennan - The Univ. of Sydney (Australia)Claudia Mello-Thoms - The Univ. of Sydney (Australia)
- Contributors
- Robert M Nishikawa (Editor) - Univ. of Pittsburgh (United States)Frank W Samuelson (Editor) - U.S. Food and Drug Administration (United States)
- Resource Type
- Conference proceeding
- Publication Details
- Progress in Biomedical Optics and Imaging - Proceedings of SPIE, Vol.10577, pp.1057708-1057708-11
- Publisher
- SPIE
- DOI
- 10.1117/12.2293366
- ISSN
- 1605-7422
- Language
- English
- Date published
- 03/07/2018
- Academic Unit
- Roy J. Carver Department of Biomedical Engineering; Radiology
- Record Identifier
- 9984051887702771
Metrics
12 Record Views