PathEdit: Diagnostic-Preserving Counterfactual Medical Image Editing through State-Factorized Generative Intervention
- 1Zhengzhou University
Abstract
Counterfactual medical images are useful only when a requested clinical change is introduced without silently altering patient characteristics that should remain fixed. Existing diffusion editors can produce convincing images, but realism or target-classifier flips do not establish that an edit is localized, diagnosis-specific, anatomically faithful, or useful on real clinical data. We introduce PATHEDIT, a state-factorized counterfactual editing framework that decomposes an abdominal CT representation into pathology, anatomy, acquisition, and residual context before performing a clinically specified latent intervention. A pathology editor modifies only the target factor, while non-target factors are explicitly copied and preservation losses penalize collateral drift. The edited state is decoded jointly into an image, a report level description, and diagnostic readouts. Weak supervision from lesion masks, organ segmentations, and naturally occurring longitudinal lesion changes anchors edit direction and spatial support without requiring pixel-aligned counterfactual ground truth. We evaluate four properties that are often conflated in medical image editing: target edit success, non-target preservation, anatomical identity, and real-data utility. In our evaluation protocol, PATHEDIT achieves 94.2% target success while reducing mean off-target diagnostic change to 3.0%, yields substantially stronger alignment with observed lesion changes, and improves rare-lesion recognition and shortcut robustness on held-out real CT. These results motivate astricter view of counterfactual medical imaging: a useful edit must prove not only what changed, but also what did not.
Related
- Community benchmarking of generative models for privacy-preserving synthetic bulk RNA-seq data generation
- A method for correcting background non-uniformity while preserving specimen-level information in Fourier ptychographic microscopy images
- Branching Suppresses Self-Aggregation while Preserving Membrane Association in SET-M33: An Atomistic Molecular Dynamics Study
- A Simple Temperature‐Controlled Method for Preserving Viability in Human Pluripotent Stem Cell‐Derived Retinal Organoids During Shipment
- GenAI-Net: A generative AI framework for automated biomolecular network design
BibTeX
@misc{chen2026pathedit,
title = {{PathEdit: Diagnostic-Preserving Counterfactual Medical Image Editing through State-Factorized Generative Intervention}},
author = {J. Chen and W. Chen},
howpublished = {bioRxiv},
year = {2026},
month = sep,
doi = {10.64898/2026.09.21.753387},
url = {https://www.biorxiv.org/content/10.64898/2026.09.21.753387v1},
note = {CC BY-NC},
}
RIS
TY - UNPB
TI - PathEdit: Diagnostic-Preserving Counterfactual Medical Image Editing through State-Factorized Generative Intervention
AU - J. Chen
AU - W. Chen
PY - 2026
DA - 2026/09/21
PB - bioRxiv
DO - 10.64898/2026.09.21.753387
UR - https://www.biorxiv.org/content/10.64898/2026.09.21.753387v1
AB - Counterfactual medical images are useful only when a requested clinical change is introduced without silently altering patient characteristics that should remain fixed. Existing diffusion editors can produce convincing images, but realism or target-classifier flips do not establish that an edit is localized, diagnosis-specific, anatomically faithful, or useful on real clinical data. We introduce PATHEDIT, a state-factorized counterfactual editing framework that decomposes an abdominal CT representation into pathology, anatomy, acquisition, and residual context before performing a clinically specified latent intervention. A pathology editor modifies only the target factor, while non-target factors are explicitly copied and preservation losses penalize collateral drift. The edited state is decoded jointly into an image, a report level description, and diagnostic readouts. Weak supervision from lesion masks, organ segmentations, and naturally occurring longitudinal lesion changes anchors edit direction and spatial support without requiring pixel-aligned counterfactual ground truth. We evaluate four properties that are often conflated in medical image editing: target edit success, non-target preservation, anatomical identity, and real-data utility. In our evaluation protocol, PATHEDIT achieves 94.2% target success while reducing mean off-target diagnostic change to 3.0%, yields substantially stronger alignment with observed lesion changes, and improves rare-lesion recognition and shortcut robustness on held-out real CT. These results motivate astricter view of counterfactual medical imaging: a useful edit must prove not only what changed, but also what did not.
ER -