Browsing by Author "He, Chongyu"

Now showing 1 - 2 of 2
  • Thumbnail Image
    Deep Learning Approach for Cell Nuclear Pore Detection and Quantification over High Resolution 3D Data
    He, Chongyu (Virginia Tech, 2023-12-21)
    The intricate task of segmenting and quantifying cell nuclear pores in high-resolution 3D microscopy data is critical for cellular biology and disease research. This thesis introduces a deep learning pipeline crafted to automate the segmentation and quantification of nuclear pores from high-resolution 3D cell organelle images. Our aim is to refine computational methods capable of handling the data's complexity and size, thus improving accuracy and reducing manual labor in biological image analysis. The developed pipeline incorporates data preprocessing, augmentation strategies, random block sampling, and a three-stage post-processing algorithm. It utilizes a 3D U-Net with a VGG-16 backbone, optimized through cyclical data augmentation and random block sampling to tackle the challenges posed by limited labeled data and the processing of large-scale 3D images. The pipeline has demonstrated its capability to effectively learn and predict nuclear pore structures, achieving improvements in validation metrics compared to baseline models. Our experiments suggest that cyclical augmentation helps prevent overfitting, and random block sampling contributes to managing data imbalance. The post-processing phase successfully automates the quantification of nuclear pores without the need for manual intervention. The proposed pipeline offers an efficient and scalable approach to segmenting and quantifying nuclear pores in 3D microscopy images. Despite the ongoing challenges of computational intensity and data volume, the techniques developed in this study provide insights into the automation of complex biological image analysis tasks, with potential applications extending beyond the detection of nuclear pores.
  • Thumbnail Image
    TextMining
    He, Chongyu; Wei, Jianchi; Mao, Chenyu (Virginia Tech, 2022-05-10)
    Electronic theses and dissertations (ETDs) contain valuable knowledge that can be useful in a wide range of research areas. Accordingly, we are building electronic infrastructure leveraging advanced work on digital libraries, for discovering and accessing the knowledge buried in ETDs. We focus on our work to incorporate topic modeling into digital libraries for ETDs. We present ETD-Topics, a framework that extracts topics from a large text corpus in an unsupervised way. The representations learnt from topic models can be useful for downstream tasks such as searching and/or browsing documents by topic, document recommendation, topic recommendation, and describing temporal topic trends (e.g., from the perspective of disciplines or universities). The characteristics of different models make the classification distinguished. We provide four modes (LDA, NeuralLDA, ProdLDA, and CTM) to serve user groups with different browsing and searching requirements. Our job was to import the preprocessed database and the trained models (four models with different topic numbers), and to accurately display key information (such as topics, document title, abstract, etc.) on web pages. We chose Python as the main language to implement the back-end, while using Flask as a bridge connecting the back-end and front-end. On the basis of using HTML for displaying data, we were able to use JavaScript and CSS to make the whole set of web pages look more fluent and comfortable by optimizing the UI, to include graphic bars, buttons (like “Submit”, “Show more”, etc.), and tables.