Dergi makalesi Açık Erişim
Ayanzadeh, Aydin; Ozuysal, Ozden Yalcin; Okvur, Devrim Pesen; Onal, Sevgi; Toreyin, Behcet Ugur; Unay, Devrim
{ "conceptrecid": "237149", "created": "2022-10-07T09:50:26.902534+00:00", "doi": "10.3906/elk-2105-244", "files": [ { "bucket": "8482e291-0c87-4e04-b81a-c19a5761dd0b", "checksum": "md5:2cd42689243f763b855cbd9610dc93c2", "key": "bib-f65cee29-3ab4-4039-aa25-77321626f448.txt", "links": { "self": "https://aperta.ulakbim.gov.tr/api/files/8482e291-0c87-4e04-b81a-c19a5761dd0b/bib-f65cee29-3ab4-4039-aa25-77321626f448.txt" }, "size": 253, "type": "txt" } ], "id": 237150, "links": { "badge": "https://aperta.ulakbim.gov.tr/badge/doi/10.3906/elk-2105-244.svg", "bucket": "https://aperta.ulakbim.gov.tr/api/files/8482e291-0c87-4e04-b81a-c19a5761dd0b", "doi": "https://doi.org/10.3906/elk-2105-244", "html": "https://aperta.ulakbim.gov.tr/record/237150", "latest": "https://aperta.ulakbim.gov.tr/api/records/237150", "latest_html": "https://aperta.ulakbim.gov.tr/record/237150" }, "metadata": { "access_right": "open", "access_right_category": "success", "communities": [ { "id": "tubitak-destekli-proje-yayinlari" } ], "creators": [ { "affiliation": "Istanbul Tech Univ, Informat Inst, Istanbul, Turkey", "name": "Ayanzadeh, Aydin" }, { "affiliation": "Izmir Inst Technol, Dept Mol Biol & Genet, Izmir, Turkey", "name": "Ozuysal, Ozden Yalcin" }, { "affiliation": "Izmir Inst Technol, Dept Mol Biol & Genet, Izmir, Turkey", "name": "Okvur, Devrim Pesen" }, { "affiliation": "Izmir Inst Technol, Biotechnol & Bioengn Grad Program, Izmir, Turkey", "name": "Onal, Sevgi" }, { "affiliation": "Istanbul Tech Univ, Informat Inst, Istanbul, Turkey", "name": "Toreyin, Behcet Ugur" }, { "affiliation": "Izmir Democracy Univ, Dept Elect & Elect Engn, Izmir, Turkey", "name": "Unay, Devrim" } ], "description": "The recently popular deep neural networks (DNNs) have a significant effect on the improvement of segmentation accuracy from various perspectives, including robustness and completeness in comparison to conventional methods. We determined that the naive U-Net has some lacks in specific perspectives and there is high potential for further enhancements on the model. Therefore, we employed some modifications in different folds of the U-Net to overcome this problem. Based on the probable opportunity for improvement, we develop a novel architecture by using an alternative feature extractor in the encoder of U-Net and replacing the plain blocks with residual blocks in the decoder. This alteration makes the model superconvergent yielding improved performance results on two challenging optical microscopy image series: a phase-contrast dataset of our own (MDA-MB-231) and a brightfield dataset from a well-known challenge (DSB2018). We utilized the U-Net with pretrained ResNet-18 as the encoder for the segmentation task. Hence, following the modifications, we redesign a novel skip-connection to reduce the semantic gap between the encoder and the decoder. The proposed skip-connection increases the accuracy of the model on both datasets. The proposed segmentation approach results in Jaccard Index values of 85.0% and 89.2% on the DSB2018 and MDA-MB-231 datasets, respectively. The results reveal that our method achieves competitive results compared to the state-of-the-art approaches and surpasses the performance of baseline approaches.", "doi": "10.3906/elk-2105-244", "has_grant": false, "journal": { "pages": "2855-2868", "title": "TURKISH JOURNAL OF ELECTRICAL ENGINEERING AND COMPUTER SCIENCES", "volume": "29" }, "license": { "id": "cc-by" }, "publication_date": "2021-01-01", "relations": { "version": [ { "count": 1, "index": 0, "is_last": true, "last_child": { "pid_type": "recid", "pid_value": "237150" }, "parent": { "pid_type": "recid", "pid_value": "237149" } } ] }, "resource_type": { "subtype": "article", "title": "Dergi makalesi", "type": "publication" }, "science_branches": [ "Di\u011fer" ], "title": "Improved cell segmentation using deep learning in label-free optical microscopy images" }, "owners": [ 1 ], "revision": 1, "stats": { "downloads": 5.0, "unique_downloads": 5.0, "unique_views": 20.0, "version_downloads": 5.0, "version_unique_downloads": 5.0, "version_unique_views": 20.0, "version_views": 21.0, "version_volume": 1265.0, "views": 21.0, "volume": 1265.0 }, "updated": "2022-10-07T09:50:26.951614+00:00" }
Görüntülenme | 21 |
İndirme | 5 |
Veri hacmi | 1.3 kB |
Tekil görüntülenme | 20 |
Tekil indirme | 5 |