Molecular Biology and Genetics / Moleküler Biyoloji ve Genetik
Permanent URI for this collectionhttps://hdl.handle.net/11147/9
Browse
2 results
Search Results
Conference Object Detection and Restoration Pipeline for Phase Contrast Microscopy Time Series Images(IEEE, 2022) Iheme, Leonardo O.; Uçar, Mahmut; Önal, Sevgi; Yalçın Özuysal, Özden; Pesen Okvur, Devrim; Töreyin, Behçet U.; Ünay, DevrimWe propose a pre-processing pipeline for the de-tection and restoration of distorted frames in phase-contrast microscopy time-series images. The analysis is based on the average intensity values of the frames within any given time- series image. The extent of the correction of intensity variation in frames is determined by the normalization of the difference between the current frame's average intensity and the median of average intensity of all frames. Our restoration algorithm preserves regional trans-passing pixels, does not cause new distortions, and increases the histogram similarity between the distorted and non-distorted frames. The algorithm was validated on 15,395 time-series image frames from 27 experiments and the results were found to be visually and quantitatively accurate.Conference Object Citation - WoS: 7Citation - Scopus: 11Cell Segmentation of 2d Phase-Contrast Microscopy Images With Deep Learning Method(Institute of Electrical and Electronics Engineers Inc., 2019) Ayanzadeh, Aydın; Yağar, Hüseyin Onur; Yalçın Özuysal, Özden; Pesen Okvur, Devrim; Töreyin, Behçet Uğur; Unay, Devrim; Önal, SevgiThe quantitative and qualitative ascertainment of cell culture is integral to the robust determination of the cell structure analysis. Microscopy cell analysis and the epithet structures of cells in cell cultures are momentous in the fields of the biological research process. In this paper, we addressed the problem of phase-contrast microscopy under cell segmentation application. In our proposed method, we utilized the state-of-the-art deep learning models trained on our proposed dataset. Due to the low number of annotated images, we propose a multi-resolution network which is based on the U-Net architecture. Moreover, we applied multi-combination augmentation to our dataset which has increased the performance of segmentation accuracy significantly. Experimental results suggest that the proposed model provides superior performance in comparison to traditional state-of-the-art segmentation algorithms.