( 您好!臺灣時間:2021/03/07 17:39
字體大小: 字級放大   字級縮小   預設字形  


研究生(外文):LIN, GUAN-YU
論文名稱(外文):2D+Z Fully-Convolution-based Neuron Cell Body Segmentation Method for Confocal Fluorescence Image Volumes
外文關鍵詞:confocal fluorescence microscopy imageneuroblastsomaimage segmentationconvolutional operations
  • 被引用被引用:0
  • 點閱點閱:36
  • 評分評分:系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔
  • 下載下載:0
  • 收藏至我的研究室書目清單書目收藏:0
隨著高齡化社會的來臨,腦科學的研究日趨受到重視,但人類大腦複雜、不易研究,使得果蠅腦在生物領域被廣為應用。生物學家希望透過相異表現型的果蠅腦神經細胞分割與計數來了解腦神經細胞的發育、運作與型態改變之情形,但在共軛焦顯微螢光影像內辨識細胞並不容易,尤其是在缺少標記的分割標準答案下,即使有很多基於U-net基礎的分割卷積神經網路(convolutional neural networks, CNNs),我們仍無法利用現有的深度神經網路方法對共軛焦顯微螢光影像內的特定細胞進行有效的影像辨識與影像分割。
With the advent of an aging society, the research of brain science has been paid more and more attention. However, the human brain is complex and difficult to study, making the Drosophila (fruit fly) brain widely used in the biological field. Biologists want to explore the mechanism of the development, the operation, and the morphological changes of brain nervous system by segmenting and counting neural cells differentiated from neuroblasts in different Drosophila phenotypes’ brain. However, it is difficult to identify cells in confocal fluorescence microscopy images. Although there are several benchmark segmentation convolutional neural networks (CNNs) based on U-net, they cannot function successfully in the absence of labeled segmentation ground truth. Hence, we cannot use the existing deep neural network methods to effectively identify and segment specific cells in the confocal fluorescence microscopy image.
Conventionally, the ImageJ software is the mostly used tool for manual preprocessing and manual labeling, but it is time-consuming, laborious, and sensitive to user-specified configurations. Therefore, the semi-manual labeling result suggested by ImageJ is still not accurate enough. To solve the problems and fix errors caused by manual labeling, we propose in this paper a convolution-based algorithm for neural cell recognizing and counting. The experimental results show that our method is more stable and more efficient than the software-assisted manual labeling result, especially in the absence of segmentation ground truth.
第壹章 緒論 1
第一節 研究背景 1
第二節 研究動機與目的 2
第三節 研究流程與架構 3
第貳章 文獻探討 4
第一節 神經細胞追蹤 4
第二節 細胞影像分割 5
第參章 研究方法 8
第一節 影像處理 8
第二節 方法結構 8
(一)模板比對 9
(二)卷積 10
(三)區域最大值 11
(四)去雜訊 (比較Pixel Values) 12
(五)影像去光暈 13
(六)影像膨脹 14
第肆章 實證研究 16
第一節 影像來源與開發環境 16
第二節 人工標記之果蠅腦神經細胞 16
第三節 全卷積神經細胞切割之果蠅腦神經細胞 27
第四節 全卷積神經細胞切割與人工標記結果之比較 38
第伍章 結論與未來展望 52
第一節 結論 52
第二節 未來展望 52
參考文獻 54
Çiçek, Ö., Abdulkadir, A., Lienkamp, S. S., Brox, T., & Ronneberger, O. (2016). 3D U-Net: learning dense volumetric segmentation from sparse annotation. International conference on medical image computing and computer-assisted intervention. Springer, Cham. 424-432.
Cheng, J., & Rajapakse, J. C. (2008). Segmentation of clustered nuclei with shape markers and marking function. IEEE Transactions on Biomedical Engineering, 56(3), 741-748.
Dzyubachyk, O., Van Cappellen, W. A., Essers, J., Niessen, W. J., & Meijering, E. (2010). Advanced level-set-based cell tracking in time-lapse fluorescence microscopy. IEEE transactions on medical imaging, 29(3), 852-867.
Huang, T., Yang, G. J. T. G. Y., & Tang, G. (1979). A fast two-dimensional median filtering algorithm. IEEE transactions on acoustics, speech, and signal processing, 27(1), 13-18.
Kavassery Rajalingam, V. (2016). Cell segmentation in cancer histopathology images using convolutional neural networks. Doctoral dissertation, University of Texas at Arlington, Arlington, Texas.
Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2012). Imagenet classification with deep convolutional neural networks. Advances in neural information processing systems. 1097-1105.
Kleesiek, J., Urban, G., Hubert, A., Schwarz, D., Maier-Hein, K., Bendszus, M., et al. (2016). Deep MRI brain extraction: A 3D convolutional neural network for skull stripping. NeuroImage, 129, 460-469.
Long, J., Shelhamer, E., & Darrell, T. (2015). Fully convolutional networks for semantic segmentation. Proceedings of the IEEE conference on computer vision and pattern recognition, 3431-3440.
Meijering, E. (2012). Cell segmentation: 50 years down the road [life sciences]. IEEE Signal Processing Magazine, 29(5), 140-145.
Motchenbacher, C. D. & Connelly, J. A. (1993). Low-noise electronic system design. Wiley Interscience.
Oberti, D., Kirschmann, M. A., & Hahnloser, R. (2010). Correlative microscopy of densely labeled projection neurons using neural tracers. Frontiers in neuroanatomy, 4, 24.
Ronneberger, O., Fischer, P., & Brox, T. (2015). U-net: Convolutional networks for biomedical image segmentation. International Conference on Medical image computing and computer-assisted intervention. Springer, Cham. 234-241.
Widrow, B., McCool, J., Larimore, M. G., & Johnson, C. R. (1977). Stationary and nonstationary learning characteristics of the LMS adaptive filter. Aspects of signal processing. Springer, Dordrecht. 355-393.
Zou, P., Chan, P., & Rockett, P. (2008). A model-based consecutive scanline tracking method for extracting vascular networks from 2-D digital subtraction angiograms. IEEE Transactions on Medical Imaging, 28(2), 241-249.
Zhou, Z., Siddiquee, M. M. R., Tajbakhsh, N., & Liang, J. (2018). Unet++: A nested u-net architecture for medical image segmentation. Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support. Springer, Cham. 3-11.

電子全文 電子全文(網際網路公開日期:20250811)
第一頁 上一頁 下一頁 最後一頁 top
系統版面圖檔 系統版面圖檔