Please use this identifier to cite or link to this item:
https://hdl.handle.net/11147/11832
Title: | Improved Cell Segmentation Using Deep Learning in Label-Free Optical Microscopy Images |
Authors: | Ayanzadeh, Aydın Yalçın Özuysal, Özden Pesen Okvur, Devrim Önal, Sevgi Töreyin, Behçet Uğur Ünay, Devrim |
Keywords: | Segmentation Breast cancer Convolutional neural networks Optical microscopy Phase-contrast microscopy Brightfield |
Publisher: | TÜBİTAK - Türkiye Bilimsel ve Teknolojik Araştırma Kurumu |
Abstract: | 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. |
URI: | https://doi.org/10.3906/elk-2105-244 https://hdl.handle.net/11147/11832 https://search.trdizin.gov.tr/yayin/detay/526977 |
ISSN: | 1300-0632 1303-6203 |
Appears in Collections: | Molecular Biology and Genetics / Moleküler Biyoloji ve Genetik Scopus İndeksli Yayınlar Koleksiyonu / Scopus Indexed Publications Collection TR Dizin İndeksli Yayınlar / TR Dizin Indexed Publications Collection WoS İndeksli Yayınlar Koleksiyonu / WoS Indexed Publications Collection |
Files in This Item:
File | Size | Format | |
---|---|---|---|
elk-29-si-1-18-2105-244.pdf | 3.19 MB | Adobe PDF | View/Open |
CORE Recommender
Items in GCRIS Repository are protected by copyright, with all rights reserved, unless otherwise indicated.