Towards capturing sonographic experience: cognition-inspired ultrasound video saliency prediction
For visual tasks like ultrasound (US) scanning, experts direct their gaze towards regions of task-relevant information. Therefore, learning to predict the gaze of sonographers on US videos captures the spatio-temporal patterns that are important for US scanning. The spatial distribution of gaze poin...
Main Authors: | , , , , , |
---|---|
Format: | Conference item |
Language: | English |
Published: |
Springer Verlag
2020
|
_version_ | 1797085794251309056 |
---|---|
author | Droste, R Cai, Y Sharma, H Chatelain, P Papageorghiou, A Noble, J |
author_facet | Droste, R Cai, Y Sharma, H Chatelain, P Papageorghiou, A Noble, J |
author_sort | Droste, R |
collection | OXFORD |
description | For visual tasks like ultrasound (US) scanning, experts direct their gaze towards regions of task-relevant information. Therefore, learning to predict the gaze of sonographers on US videos captures the spatio-temporal patterns that are important for US scanning. The spatial distribution of gaze points on video frames can be represented through heat maps termed saliency maps. Here, we propose a temporally bidirectional model for video saliency prediction (BDS-Net), drawing inspiration from modern theories of human cognition. The model consists of a convolutional neural network (CNN) encoder followed by a bidirectional gated-recurrent-unit recurrent convolutional network (GRU-RCN) decoder. The temporal bidirectionality mimics human cognition, which simultaneously reacts to past and predicts future sensory inputs. We train the BDS-Net alongside spatial and temporally one-directional comparative models on the task of predicting saliency in videos of US abdominal circumference plane detection. The BDS-Net outperforms the comparative models on four out of five saliency metrics. We present a qualitative analysis on representative examples to explain the model’s superior performance. |
first_indexed | 2024-03-07T02:13:01Z |
format | Conference item |
id | oxford-uuid:a14df633-3dc5-4918-ba90-09dda3f51363 |
institution | University of Oxford |
language | English |
last_indexed | 2024-03-07T02:13:01Z |
publishDate | 2020 |
publisher | Springer Verlag |
record_format | dspace |
spelling | oxford-uuid:a14df633-3dc5-4918-ba90-09dda3f513632022-03-27T02:12:11ZTowards capturing sonographic experience: cognition-inspired ultrasound video saliency predictionConference itemhttp://purl.org/coar/resource_type/c_5794uuid:a14df633-3dc5-4918-ba90-09dda3f51363EnglishSymplectic Elements at OxfordSpringer Verlag2020Droste, RCai, YSharma, HChatelain, PPapageorghiou, ANoble, JFor visual tasks like ultrasound (US) scanning, experts direct their gaze towards regions of task-relevant information. Therefore, learning to predict the gaze of sonographers on US videos captures the spatio-temporal patterns that are important for US scanning. The spatial distribution of gaze points on video frames can be represented through heat maps termed saliency maps. Here, we propose a temporally bidirectional model for video saliency prediction (BDS-Net), drawing inspiration from modern theories of human cognition. The model consists of a convolutional neural network (CNN) encoder followed by a bidirectional gated-recurrent-unit recurrent convolutional network (GRU-RCN) decoder. The temporal bidirectionality mimics human cognition, which simultaneously reacts to past and predicts future sensory inputs. We train the BDS-Net alongside spatial and temporally one-directional comparative models on the task of predicting saliency in videos of US abdominal circumference plane detection. The BDS-Net outperforms the comparative models on four out of five saliency metrics. We present a qualitative analysis on representative examples to explain the model’s superior performance. |
spellingShingle | Droste, R Cai, Y Sharma, H Chatelain, P Papageorghiou, A Noble, J Towards capturing sonographic experience: cognition-inspired ultrasound video saliency prediction |
title | Towards capturing sonographic experience: cognition-inspired ultrasound video saliency prediction |
title_full | Towards capturing sonographic experience: cognition-inspired ultrasound video saliency prediction |
title_fullStr | Towards capturing sonographic experience: cognition-inspired ultrasound video saliency prediction |
title_full_unstemmed | Towards capturing sonographic experience: cognition-inspired ultrasound video saliency prediction |
title_short | Towards capturing sonographic experience: cognition-inspired ultrasound video saliency prediction |
title_sort | towards capturing sonographic experience cognition inspired ultrasound video saliency prediction |
work_keys_str_mv | AT droster towardscapturingsonographicexperiencecognitioninspiredultrasoundvideosaliencyprediction AT caiy towardscapturingsonographicexperiencecognitioninspiredultrasoundvideosaliencyprediction AT sharmah towardscapturingsonographicexperiencecognitioninspiredultrasoundvideosaliencyprediction AT chatelainp towardscapturingsonographicexperiencecognitioninspiredultrasoundvideosaliencyprediction AT papageorghioua towardscapturingsonographicexperiencecognitioninspiredultrasoundvideosaliencyprediction AT noblej towardscapturingsonographicexperiencecognitioninspiredultrasoundvideosaliencyprediction |