Publication: The effect of class-weighted penalization in deep neural networks for multi-class cell segmentation
dc.contributor.coauthor | Aydin, Musa | |
dc.contributor.coauthor | Kus, Zeki | |
dc.contributor.coauthor | Kiraz, Berna | |
dc.contributor.coauthor | Hosavci, Reyhan | |
dc.contributor.department | Department of Physics | |
dc.contributor.kuauthor | Kiraz, Alper | |
dc.contributor.schoolcollegeinstitute | College of Sciences | |
dc.date.accessioned | 2025-03-06T20:57:15Z | |
dc.date.issued | 2024 | |
dc.description.abstract | Deep learning networks give successful results in many areas, but their complexity leads to problems such as overfitting. Many approaches have been proposed to solve this problem, and class-based penalization has been one of the methods that have yielded successful results. With class-based penalization, it has become possible to increase the prediction performance and improve the model's generalization capability, especially in cases with class imbalance. This study investigates the effect of class-based penalization on the multiclass cell segmentation problem. Two deep neural network models (Resnet18, EfficientNet) are tested with 6 different configurations created for class-based penalization, and the results are compared. The experimental studies show the relationship between class-based loss penalties and multiclass segmentation/classification performance. The results show that class-based penalization improves the total performance of EfficientNet and Resnet18 networks by 11.82(C2+C3) and 12.79(C2+C3) points, respectively. It is shown that the proposed method can improve the prediction performance without increasing the model complexity. | |
dc.description.indexedby | WOS | |
dc.description.indexedby | Scopus | |
dc.description.publisherscope | International | |
dc.description.sponsoredbyTubitakEu | N/A | |
dc.identifier.doi | 10.1109/SIU61531.2024.10601040 | |
dc.identifier.isbn | 9798350388978 | |
dc.identifier.isbn | 9798350388961 | |
dc.identifier.issn | 2165-0608 | |
dc.identifier.quartile | N/A | |
dc.identifier.scopus | 2-s2.0-85200860403 | |
dc.identifier.uri | https://doi.org/10.1109/SIU61531.2024.10601040 | |
dc.identifier.uri | https://hdl.handle.net/20.500.14288/27170 | |
dc.identifier.wos | 1297894700253 | |
dc.keywords | Cell segmentation | |
dc.keywords | Classification | |
dc.keywords | Class weighted penalization | |
dc.language.iso | tur | |
dc.publisher | IEEE | |
dc.relation.ispartof | 32ND IEEE SIGNAL PROCESSING AND COMMUNICATIONS APPLICATIONS CONFERENCE, SIU 2024 | |
dc.subject | Computer science | |
dc.subject | Electrical and electronic | |
dc.subject | Telecommunications | |
dc.title | The effect of class-weighted penalization in deep neural networks for multi-class cell segmentation | |
dc.title.alternative | Çok sınıflı hücre bölütleme için derin sinir ağlarında sınıf ağırlıklı cezalandırmanın etkisi | |
dc.type | Conference Proceeding | |
dspace.entity.type | Publication | |
local.contributor.kuauthor | Kiraz, Alper | |
local.publication.orgunit1 | College of Sciences | |
local.publication.orgunit2 | Department of Physics | |
relation.isOrgUnitOfPublication | c43d21f0-ae67-4f18-a338-bcaedd4b72a4 | |
relation.isOrgUnitOfPublication.latestForDiscovery | c43d21f0-ae67-4f18-a338-bcaedd4b72a4 | |
relation.isParentOrgUnitOfPublication | af0395b0-7219-4165-a909-7016fa30932d | |
relation.isParentOrgUnitOfPublication.latestForDiscovery | af0395b0-7219-4165-a909-7016fa30932d |