{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/attention-gated-networks-for-improving","title":"Attention-Gated Networks for Improving Ultrasound Scan Plane Detection","arxiv_id":"1804.05338","date":"2018-04-15","proceeding":null,"authors":["Jo Schlemper","Ozan Oktay","Liang Chen","Jacqueline Matthew","Caroline Knight","Bernhard Kainz","Ben Glocker","Daniel Rueckert"],"abstract":"In this work, we apply an attention-gated network to real-time automated scan\nplane detection for fetal ultrasound screening. Scan plane detection in fetal\nultrasound is a challenging problem due the poor image quality resulting in low\ninterpretability for both clinicians and automated algorithms. To solve this,\nwe propose incorporating self-gated soft-attention mechanisms. A soft-attention\nmechanism generates a gating signal that is end-to-end trainable, which allows\nthe network to contextualise local information useful for prediction. The\nproposed attention mechanism is generic and it can be easily incorporated into\nany existing classification architectures, while only requiring a few\nadditional parameters. We show that, when the base network has a high capacity,\nthe incorporated attention mechanism can provide efficient object localisation\nwhile improving the overall performance. When the base network has a low\ncapacity, the method greatly outperforms the baseline approach and\nsignificantly reduces false positives. Lastly, the generated attention maps\nallow us to understand the model's reasoning process, which can also be used\nfor weakly supervised object localisation.","url_abs":"http://arxiv.org/abs/1804.05338v1","url_pdf":"http://arxiv.org/pdf/1804.05338v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"attention-gated-networks-for-improving","repo_url":"https://github.com/ozan-oktay/Attention-Gated-Networks","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"attention-gated-networks-for-improving","repo_url":"https://github.com/RobbieHolland/CrohnsDisease","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"attention-gated-networks-for-improving","repo_url":"https://github.com/SaoYan/LearnToPayAttention","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"GPL-3.0"}},{"paper_slug":"attention-gated-networks-for-improving","repo_url":"https://github.com/iversonicter/Learn-to-pay-attention","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"attention-gated-networks-for-improving","repo_url":"https://github.com/srb-cv/AttentionClassification","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"attention-gated-networks-for-improving","repo_url":"https://github.com/wangyongjie-ntu/Learn-to-pay-attention","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1804.05338","atlas_url":"https://app.syntology.ai/?focus=1804.05338","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}