Segmentation of electron microscopy data

• Partner: Institute of Cell Biology
• Date: Ongoing

In our work on segmentation of electron microscopy data, we have developed advanced algorithms for both semantic and instance segmentation of four cellular structures - fusiform vesicles, mitochondria, endolysosomes, and the Golgi apparatus. These algorithms, developed for segmentation of volumetric focused ion beam scanning electron microscopy (FIB-SEM) data of urothelial cells, allow for precise identification and characterization of these compartments. Additionally, we have developed specialized algorithms for the shape analysis of mitochondria, providing further insights into cellular structure and function.

One of the key contributions of this project is Urocell, an open-source dataset available at Urocell GitHub. This resource supports segmentation tasks and fosters collaboration across the research community.

Current Research Initiatives

• Expanding Segmentation Methods: Broaden segmentation techniques to cover additional cellular compartments, such as membranes, to support more comprehensive cellular analyses.
• Refining Instance Segmentation Algorithms: Enhance segmentation algorithms to improve accuracy and efficiency, facilitating more reliable analysis of cellular structures.
• Develop a Comprehensive Analysis Pipeline: Build an integrated analysis pipeline tailored to cellular biologists' workflows, enabling high-throughput, reproducible studies.
• Collaborating with the Institute for Biology, University of Ljubljana: Partner with experts to explore advanced segmentation techniques for a diverse range of cellular structures, contributing to cutting-edge biological research.

This project represents a significant advancement in computational tools for cellular biology, offering invaluable resources and innovations to the life sciences community.

If you're interested in contributing to this exciting research—whether as part of your diploma or master's thesis or simply as a hobby—please reach out to us! We would love to discuss how you could get involved. Contact Ciril or Manca.

Bibliography

• M. Ž. Mekuč, C. Bohak, E. Boneš, S. Hudoklin, R. Romih, and M. Marolt, "Automatic segmentation and reconstruction of intracellular compartments in volumetric electron microscopy data," Computer methods and programs in biomedicine, iss. 223, pp. 1-14, 2022.
  [Bibtex]

  @article{112962819,
  author={Manca Žerovnik Mekuč and Ciril Bohak and Eva Boneš and Samo Hudoklin and Rok Romih and Matija Marolt},
  year={2022},
  pages={1-14},
  title={Automatic segmentation and reconstruction of intracellular compartments in volumetric electron microscopy data},
  journal={Computer methods and programs in biomedicine},
  number={223},
  }

• E. Boneš and M. Marolt, "Automatic segmentation of the Golgi apparatus in volumetric data with approximate labels," in TELSIKS 2021 : proceedings of papers : 2021 15th International Conference on Advanced Technologies, Systems and Services in Telecommunications (TELSIKS), Serbia, Niš, October 20 - 22, 2021, 2021, pp. 393-397.
  [Bibtex]

  @conference{87048451,
  author={Eva Boneš and Matija Marolt},
  year={2021},
  pages={393-397},
  title={Automatic segmentation of the Golgi apparatus in volumetric data with approximate labels},
  booktitle={TELSIKS 2021 : proceedings of papers : 2021 15th International Conference on Advanced Technologies, Systems and Services in Telecommunications (TELSIKS), Serbia, Niš, October 20 - 22, 2021},
  }

• M. Ž. Mekuč, C. Bohak, S. Hudoklin, B. H. Kim, R. Romih, M. Y. Kim, and M. Marolt, "Automatic segmentation of mitochondria and endolysosomes in volumetric electron microscopy data," Computers in Biology and Medicine, iss. 119, pp. 1-11, 2020.
  [Bibtex]

  @article{1538555331,
  author={Manca Žerovnik Mekuč and Ciril Bohak and Samo Hudoklin and Byeong Hak Kim and Rok Romih and Min Young Kim and Matija Marolt},
  year={2020},
  pages={1-11},
  title={Automatic segmentation of mitochondria and endolysosomes in volumetric electron microscopy data},
  journal={Computers in Biology and Medicine},
  number={119},
  }

• E. Boneš, "Avtomatska segmentacija Golgijevih aparatov v volumetričnih podatkih z grobo označenim zlatim standardom," in Zbornik devetindvajsete mednarodne Elektrotehniške in računalniške konference ERK 2020 = Proceedings of the Twenty-ninth International Electrotechnical and Computer Science Conference ERK 2020, 2020, pp. 448-451.
  [Bibtex]

  @conference{29391363,
  author={Eva Boneš},
  year={2020},
  pages={448-451},
  title={Avtomatska segmentacija Golgijevih aparatov v volumetričnih podatkih z grobo označenim zlatim standardom},
  booktitle={Zbornik devetindvajsete mednarodne Elektrotehniške in računalniške konference ERK 2020 = Proceedings of the Twenty-ninth International Electrotechnical and Computer Science Conference ERK 2020},
  }

• M. Ž. Mekuč, C. Bohak, R. Romih, S. Hudoklin, and M. Marolt, "Avtomatska segmentacija celičnih predelkov v volumetričnih podatkih, pridobljenih z elektronskim mikroskopom," Uporabna informatika, vol. 27, iss. 4, pp. 153-156, 2019.
  [Bibtex]

  @article{1538498243,
  author={Manca Žerovnik Mekuč and Ciril Bohak and Rok Romih and Samo Hudoklin and Matija Marolt},
  year={2019},
  pages={153-156},
  volume={27},
  title={Avtomatska segmentacija celičnih predelkov v volumetričnih podatkih, pridobljenih z elektronskim mikroskopom},
  journal={Uporabna informatika},
  number={4},
  }