AttentionBoost: learning what to attend for gland segmentation in histopathological images by boosting fully convolutional networks

Simple item page
buir.contributor.authorGüneşli, Gözde Nur
buir.contributor.authorGündüz-Demir, Çiğdem
dc.citation.epage4273en_US
dc.citation.issueNumber12en_US
dc.citation.spage4262en_US
dc.citation.volumeNumber39en_US
dc.contributor.authorGüneşli, Gözde Nuren_US
dc.contributor.authorSökmensüer, C.en_US
dc.contributor.authorGündüz-Demir, Çiğdemen_US
dc.date.accessioned2021-02-18T10:46:30Z
dc.date.available2021-02-18T10:46:30Z
dc.date.issued2020
dc.departmentDepartment of Computer Engineeringen_US
dc.departmentInterdisciplinary Program in Neuroscience (NEUROSCIENCE)en_US
dc.departmentNational Magnetic Resonance Research Center (UMRAM)en_US
dc.description.abstractFully convolutional networks (FCNs) are widely used for instance segmentation. One important challenge is to sufficiently train these networks to yield good generalizations for hard-to-learn pixels, correct prediction of which may greatly affect the success. A typical group of such hard-to-learn pixels are boundaries between instances. Many studies have developed strategies to pay more attention to learning these boundary pixels. They include designing multi-task networks with an additional task of boundary prediction and increasing the weights of boundary pixels' predictions in the loss function. Such strategies require defining what to attend beforehand and incorporating this defined attention to the learning model. However, there may exist other groups of hard-to-learn pixels and manually defining and incorporating the appropriate attention for each group may not be feasible. In order to provide an adaptable solution to learn different groups of hard-to-learn pixels, this article proposes AttentionBoost, which is a new multi-attention learning model based on adaptive boosting, for the task of gland instance segmentation in histopathological images. AttentionBoost designs a multi-stage network and introduces a new loss adjustment mechanism for an FCN to adaptively learn what to attend at each stage directly on image data without necessitating any prior definition. This mechanism modulates the attention of each stage to correct the mistakes of previous stages, by adjusting the loss weight of each pixel prediction separately with respect to how accurate the previous stages are on this pixel. Working on histopathological images of colon tissues, our experiments demonstrate that the proposed AttentionBoost model improves the results of gland segmentation compared to its counterparts.en_US
dc.description.sponsorshipThis work was supported by the Scientific and Technological Research Council of Turkey under Project TÜBİTAK 116E075.en_US
dc.identifier.doi10.1109/TMI.2020.3015198en_US
dc.identifier.issn0278-0062
dc.identifier.urihttp://hdl.handle.net/11693/75447
dc.language.isoEnglishen_US
dc.publisherIEEEen_US
dc.relation.isversionofhttps://dx.doi.org/10.1109/TMI.2020.3015198en_US
dc.source.titleIEEE Transactions on Medical Imagingen_US
dc.subjectDeep learningen_US
dc.subjectAttention learningen_US
dc.subjectAdaptive boostingen_US
dc.subjectGland instance segmentationen_US
dc.subjectInstance segmentationen_US
dc.titleAttentionBoost: learning what to attend for gland segmentation in histopathological images by boosting fully convolutional networksen_US
dc.typeArticleen_US

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
AttentionBoost_Learning_What_to_Attend_for_Gland_Segmentation_in_Histopathological_Images_by_Boosting_Fully_Convolutional_Networks.pdf
Size:
2.92 MB
Format:
Adobe Portable Document Format
Description:
Download
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
1.71 KB
Format:
Item-specific license agreed upon to submission
Description:
Download

Collections

Department of Computer Engineering
Aysel Sabuncu Brain Research Center (BAM)
National Magnetic Resonance Research Center (UMRAM)