MEDIAR: Harmony of Data-Centric and Model-Centric for Multi-Modality Microscopy

Gihun Lee, SangMook Kim, Joonkee Kim, Se-Young. Yun
Proceedings of The Cell Segmentation Challenge in Multi-modality High-Resolution Microscopy Images, PMLR 212:1-16, 2023.

Abstract

Cell segmentation is a fundamental task for computational biology analysis. Identifying the cell instances is often the first step in various downstream biomedical studies. However, many cell segmentation algorithms, including the recently emerging deep learning-based methods, still show limited generality under the multimodality environment. Weakly Supervised Cell Segmentation in Multi-modality High-Resolution Microscopy Images was hosted at NeurIPS 2022 to tackle this problem. We propose MEDIAR, a holistic pipeline for cell instance segmentation under multi-modality in this challenge. MEDIAR harmonizes data-centric and model-centric approaches as the learning and inference strategies, achieving a 0.9067 F1-score at the validation phase while satisfying the time budget. To facilitate subsequent research, we provide the source code and trained model as open-source.

Cite this Paper

BibTeX
@InProceedings{pmlr-v212-lee23a, title = {MEDIAR: Harmony of Data-Centric and Model-Centric for Multi-Modality Microscopy}, author = {Lee, Gihun and Kim, SangMook and Kim, Joonkee and Yun, Se-Young.}, booktitle = {Proceedings of The Cell Segmentation Challenge in Multi-modality High-Resolution Microscopy Images}, pages = {1--16}, year = {2023}, editor = {Ma, Jun and Xie, Ronald and Gupta, Anubha and Guilherme de Almeida, José and Bader, Gary D. and Wang, Bo}, volume = {212}, series = {Proceedings of Machine Learning Research}, month = {28 Nov--09 Dec}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v212/lee23a/lee23a.pdf}, url = {https://proceedings.mlr.press/v212/lee23a.html}, abstract = {Cell segmentation is a fundamental task for computational biology analysis. Identifying the cell instances is often the first step in various downstream biomedical studies. However, many cell segmentation algorithms, including the recently emerging deep learning-based methods, still show limited generality under the multimodality environment. Weakly Supervised Cell Segmentation in Multi-modality High-Resolution Microscopy Images was hosted at NeurIPS 2022 to tackle this problem. We propose MEDIAR, a holistic pipeline for cell instance segmentation under multi-modality in this challenge. MEDIAR harmonizes data-centric and model-centric approaches as the learning and inference strategies, achieving a 0.9067 F1-score at the validation phase while satisfying the time budget. To facilitate subsequent research, we provide the source code and trained model as open-source.} }
Endnote
%0 Conference Paper %T MEDIAR: Harmony of Data-Centric and Model-Centric for Multi-Modality Microscopy %A Gihun Lee %A SangMook Kim %A Joonkee Kim %A Se-Young. Yun %B Proceedings of The Cell Segmentation Challenge in Multi-modality High-Resolution Microscopy Images %C Proceedings of Machine Learning Research %D 2023 %E Jun Ma %E Ronald Xie %E Anubha Gupta %E José Guilherme de Almeida %E Gary D. Bader %E Bo Wang %F pmlr-v212-lee23a %I PMLR %P 1--16 %U https://proceedings.mlr.press/v212/lee23a.html %V 212 %X Cell segmentation is a fundamental task for computational biology analysis. Identifying the cell instances is often the first step in various downstream biomedical studies. However, many cell segmentation algorithms, including the recently emerging deep learning-based methods, still show limited generality under the multimodality environment. Weakly Supervised Cell Segmentation in Multi-modality High-Resolution Microscopy Images was hosted at NeurIPS 2022 to tackle this problem. We propose MEDIAR, a holistic pipeline for cell instance segmentation under multi-modality in this challenge. MEDIAR harmonizes data-centric and model-centric approaches as the learning and inference strategies, achieving a 0.9067 F1-score at the validation phase while satisfying the time budget. To facilitate subsequent research, we provide the source code and trained model as open-source.
APA
Lee, G., Kim, S., Kim, J. & Yun, S.. (2023). MEDIAR: Harmony of Data-Centric and Model-Centric for Multi-Modality Microscopy. Proceedings of The Cell Segmentation Challenge in Multi-modality High-Resolution Microscopy Images, in Proceedings of Machine Learning Research 212:1-16 Available from https://proceedings.mlr.press/v212/lee23a.html.

Related Material