Lesion-Aware Post-Training of Latent Diffusion Models for Synthesizing Diffusion MRI from CT Perfusion
- URL: http://arxiv.org/abs/2510.09056v1
- Date: Fri, 10 Oct 2025 06:51:27 GMT
- Title: Lesion-Aware Post-Training of Latent Diffusion Models for Synthesizing Diffusion MRI from CT Perfusion
- Authors: Junhyeok Lee, Hyunwoong Kim, Hyungjin Chung, Heeseong Eom, Joon Jang, Chul-Ho Sohn, Kyu Sung Choi,
- Abstract summary: Latent diffusion models (LDMs) leverage efficient learning in compressed latent space.<n>This efficiency comes with a trade-off, potentially compromising crucial pixel-level detail.<n>We propose a novel post-training framework for LDMs in medical image-to-image translation by incorporating lesion-aware medical pixel space objectives.
- Score: 12.777937532267105
- License: http://creativecommons.org/licenses/by/4.0/
- Abstract: Image-to-Image translation models can help mitigate various challenges inherent to medical image acquisition. Latent diffusion models (LDMs) leverage efficient learning in compressed latent space and constitute the core of state-of-the-art generative image models. However, this efficiency comes with a trade-off, potentially compromising crucial pixel-level detail essential for high-fidelity medical images. This limitation becomes particularly critical when generating clinically significant structures, such as lesions, which often occupy only a small portion of the image. Failure to accurately reconstruct these regions can severely impact diagnostic reliability and clinical decision-making. To overcome this limitation, we propose a novel post-training framework for LDMs in medical image-to-image translation by incorporating lesion-aware medical pixel space objectives. This approach is essential, as it not only enhances overall image quality but also improves the precision of lesion delineation. We evaluate our framework on brain CT-to-MRI translation in acute ischemic stroke patients, where early and accurate diagnosis is critical for optimal treatment selection and improved patient outcomes. While diffusion MRI is the gold standard for stroke diagnosis, its clinical utility is often constrained by high costs and low accessibility. Using a dataset of 817 patients, we demonstrate that our framework improves overall image quality and enhances lesion delineation when synthesizing DWI and ADC images from CT perfusion scans, outperforming existing image-to-image translation models. Furthermore, our post-training strategy is easily adaptable to pre-trained LDMs and exhibits substantial potential for broader applications across diverse medical image translation tasks.
Related papers
- Structure-constrained Language-informed Diffusion Model for Unpaired Low-dose Computed Tomography Angiography Reconstruction [72.80209358480424]
overdose of iodinated contrast media (ICM) can cause kidney damage and life-threatening allergic reactions.<n>Deep learning methods can generate CT images of normal-dose ICM from low-dose ICM, reducing the required dose.<n>We propose a Structure-constrained Language-informed Diffusion Model (SLDM) that integrates structural synergy and spatial intelligence.
arXiv Detail & Related papers (2026-01-28T06:54:06Z) - A Semantically Enhanced Generative Foundation Model Improves Pathological Image Synthesis [82.01597026329158]
We introduce a Correlation-Regulated Alignment Framework for Tissue Synthesis (CRAFTS) for pathology-specific text-to-image synthesis.<n>CRAFTS incorporates a novel alignment mechanism that suppresses semantic drift to ensure biological accuracy.<n>This model generates diverse pathological images spanning 30 cancer types, with quality rigorously validated by objective metrics and pathologist evaluations.
arXiv Detail & Related papers (2025-12-15T10:22:43Z) - Causal Disentanglement for Robust Long-tail Medical Image Generation [80.15257897500578]
We propose a novel medical image generation framework, which generates independent pathological and structural features.<n>We leverage a diffusion model guided by pathological findings to model pathological features, enabling the generation of diverse counterfactual images.
arXiv Detail & Related papers (2025-04-20T01:54:18Z) - MedIL: Implicit Latent Spaces for Generating Heterogeneous Medical Images at Arbitrary Resolutions [2.2427832125073732]
MedIL is a first-of-its-kind autoencoder built for encoding medical images with heterogeneous sizes and resolutions.<n>We show how MedIL compresses and preserves clinically-relevant features over large multi-site, multi-resolution datasets.
arXiv Detail & Related papers (2025-04-12T19:52:56Z) - Latent Drifting in Diffusion Models for Counterfactual Medical Image Synthesis [55.959002385347645]
Latent Drifting enables diffusion models to be conditioned for medical images fitted for the complex task of counterfactual image generation.<n>We evaluate our method on three public longitudinal benchmark datasets of brain MRI and chest X-rays for counterfactual image generation.
arXiv Detail & Related papers (2024-12-30T01:59:34Z) - A Unified Model for Compressed Sensing MRI Across Undersampling Patterns [69.19631302047569]
We propose a unified MRI reconstruction model robust to various measurement undersampling patterns and image resolutions.<n>Our model improves SSIM by 11% and PSNR by 4 dB over a state-of-the-art CNN (End-to-End VarNet) with 600$times$ faster inference than diffusion methods.
arXiv Detail & Related papers (2024-10-05T20:03:57Z) - Multiscale Latent Diffusion Model for Enhanced Feature Extraction from Medical Images [5.395912799904941]
variations in CT scanner models and acquisition protocols introduce significant variability in the extracted radiomic features.<n> LTDiff++ is a multiscale latent diffusion model designed to enhance feature extraction in medical imaging.
arXiv Detail & Related papers (2024-10-05T02:13:57Z) - Cross-Modal Domain Adaptation in Brain Disease Diagnosis: Maximum Mean Discrepancy-based Convolutional Neural Networks [0.0]
Brain disorders are a major challenge to global health, causing millions of deaths each year.
Accurate diagnosis of these diseases relies heavily on advanced medical imaging techniques such as MRI and CT.
The scarcity of annotated data poses a significant challenge in deploying machine learning models for medical diagnosis.
arXiv Detail & Related papers (2024-05-06T07:44:46Z) - A New Multimodal Medical Image Fusion based on Laplacian Autoencoder
with Channel Attention [3.1531360678320897]
Deep learning models have achieved end-to-end image fusion with highly robust and accurate performance.
Most DL-based fusion models perform down-sampling on the input images to minimize the number of learnable parameters and computations.
We propose a new multimodal medical image fusion model is proposed that is based on integrated Laplacian-Gaussian concatenation with attention pooling.
arXiv Detail & Related papers (2023-10-18T11:29:53Z) - On Sensitivity and Robustness of Normalization Schemes to Input
Distribution Shifts in Automatic MR Image Diagnosis [58.634791552376235]
Deep Learning (DL) models have achieved state-of-the-art performance in diagnosing multiple diseases using reconstructed images as input.
DL models are sensitive to varying artifacts as it leads to changes in the input data distribution between the training and testing phases.
We propose to use other normalization techniques, such as Group Normalization and Layer Normalization, to inject robustness into model performance against varying image artifacts.
arXiv Detail & Related papers (2023-06-23T03:09:03Z) - Negligible effect of brain MRI data preprocessing for tumor segmentation [36.89606202543839]
We conduct experiments on three publicly available datasets and evaluate the effect of different preprocessing steps in deep neural networks.
Our results demonstrate that most popular standardization steps add no value to the network performance.
We suggest that image intensity normalization approaches do not contribute to model accuracy because of the reduction of signal variance with image standardization.
arXiv Detail & Related papers (2022-04-11T17:29:36Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.