Modification of DDIM Encoding for Generating Counterfactual Pathology Images of Malignant Lymphoma

Ryoichi Koga, Mauricio Kugler, Tatsuya Yokota, Kouichi Ohshima, Kouichi Ohshima, Hiroaki Miyoshi, Hiroaki Miyoshi, Miharu Nagaishi, Noriaki Hashimoto, Ichiro Takeuchi, Ichiro Takeuchi, Hidekata Hontani

2024

Abstract

We propose a method that modifies encoding in DDIM (Denoising Diffusion Implicit Model) to improve the quality of counterfactual histopathological images of malignant lymphoma. Counterfactual medical images are widely employed for analyzing the changes in images accompanying disease. For the analysis of pathological images, it is desired to accurately represent the types of individual cells in the tissue. We employ DDIM because it can refer to exogenous variables in causal models and can generate counterfactual images. Here, one problem of DDIM is that it does not always generate accurate images due to approximations in the forward process. In this paper, we propose a method that reduces the errors in the encoded images obtained in the forward process. Since the computation in the backward process of DDIM does not include any approximation, the accurate encoding in the forward process can improve the accuracy of the image generation. Our proposed method improves the accuracy of encoding by explicitly referring to the given original image. Experiments demonstrate that our proposed method accurately reconstructs original images, including microstructures such as cell nuclei, and outperforms the conventional DDIM in several measures of image generation.

Paper Citation

in Harvard Style

Koga R., Kugler M., Yokota T., Ohshima K., Miyoshi H., Nagaishi M., Hashimoto N., Takeuchi I. and Hontani H. (2024). Modification of DDIM Encoding for Generating Counterfactual Pathology Images of Malignant Lymphoma. In Proceedings of the 19th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 2: VISAPP; ISBN 978-989-758-679-8, SciTePress, pages 519-527. DOI: 10.5220/0012366100003660

in Bibtex Style

@conference{visapp24,
author={Ryoichi Koga and Mauricio Kugler and Tatsuya Yokota and Kouichi Ohshima and Hiroaki Miyoshi and Miharu Nagaishi and Noriaki Hashimoto and Ichiro Takeuchi and Hidekata Hontani},
title={Modification of DDIM Encoding for Generating Counterfactual Pathology Images of Malignant Lymphoma},
booktitle={Proceedings of the 19th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 2: VISAPP},
year={2024},
pages={519-527},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012366100003660},
isbn={978-989-758-679-8},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 19th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 2: VISAPP
TI - Modification of DDIM Encoding for Generating Counterfactual Pathology Images of Malignant Lymphoma
SN - 978-989-758-679-8
AU - Koga R.
AU - Kugler M.
AU - Yokota T.
AU - Ohshima K.
AU - Miyoshi H.
AU - Nagaishi M.
AU - Hashimoto N.
AU - Takeuchi I.
AU - Hontani H.
PY - 2024
SP - 519
EP - 527
DO - 10.5220/0012366100003660
PB - SciTePress