Pseudo-healthy image synthesis via location-guided diffusion models for focal cortical dysplasia lesion localization.
Authors
Affiliations (8)
Affiliations (8)
- Department of Pathology, Honghui Hospital, Xi'an Jiaotong University, Xi'an, China.
- Key Laboratory of Micro-Nano Sensing and IoT of Wenzhou, Wenzhou Institute of Hangzhou Dianzi University, Wenzhou, 325038, China.
- Hangzhou Dianzi University, Hangzhou, China.
- Shenzhen University of Advanced Technology, Shenzhen, China.
- Hangzhou Third people's Hospital, Hangzhou, China.
- Shenzhen Research Institute of Big Data, Shenzhen, China. [email protected].
- Key Laboratory of Micro-Nano Sensing and IoT of Wenzhou, Wenzhou Institute of Hangzhou Dianzi University, Wenzhou, 325038, China. [email protected].
- Hangzhou Dianzi University, Hangzhou, China. [email protected].
Abstract
Focal cortical dysplasia (FCD) is a developmental disorder frequently linked to drug-resistant focal epilepsy, where surgical intervention is often the most promising treatment. Unfortunately, conventional neuroradiological methods often struggle to detect subtle FCD cases, which may result in missed surgical opportunities for patients. To address this challenge, we have developed an innovative approach to FCD lesions detection using conditional diffusion models guided by complementary location-based modal factors. Our method involves employing a classifier to identify epileptic sites while directing the diffusion model to generate pseudo-healthy images. Using the unique features of FCD present in both T1 and FLAIR images, T1 images conditions the diffusion model to produce the corresponding FLAIR image, effectively removing abnormal tissue. Recognizing the difficulty in detecting epileptic lesions due to their subtle presentation, we have incorporated histogram matching techniques to address the color distortion issues commonly associated with diffusion models. This adjustment ensures that chromatic aberration does not hinder the identification of lesions. The effectiveness of our method has been validated using the UHB FCD MRI dataset, achieving an image-level recall metric of 0.952 and a pixel-level dice metric of 0.245. These results surpass those obtained from four other comparative methods, underscoring the superior performance of our approach. Our code is available at https://github.com/CodePYJ/FCD-Detection.