Dataset Information

Semisupervised 3D segmentation of pancreatic tumors in positron emission tomography/computed tomography images using a mutual information minimization and cross-fusion strategy.

ABSTRACT:

Background

Accurate segmentation of pancreatic cancer tumors using positron emission tomography/computed tomography (PET/CT) multimodal images is crucial for clinical diagnosis and prognosis evaluation. However, deep learning methods for automated medical image segmentation require a substantial amount of manually labeled data, making it time-consuming and labor-intensive. Moreover, addition or simple stitching of multimodal images leads to redundant information, failing to fully exploit the complementary information of multimodal images. Therefore, we developed a semisupervised multimodal network that leverages limited labeled samples and introduces a cross-fusion and mutual information minimization (MIM) strategy for PET/CT 3D segmentation of pancreatic tumors.

Methods

Our approach combined a cross multimodal fusion (CMF) module with a cross-attention mechanism. The complementary multimodal features were fused to form a multifeature set to enhance the effectiveness of feature extraction while preserving specific features of each modal image. In addition, we designed an MIM module to mitigate redundant high-level modal information and compute the latent loss of PET and CT. Finally, our method employed the uncertainty-aware mean teacher semi-supervised framework to segment regions of interest from PET/CT images using a small amount of labeled data and a large amount of unlabeled data.

Results

We evaluated our combined MIM and CMF semisupervised segmentation network (MIM-CMFNet) on a private dataset of pancreatic cancer, yielding an average Dice coefficient of 73.14%, an average Jaccard index score of 60.56%, and an average 95% Hausdorff distance (95HD) of 6.30 mm. In addition, to verify the broad applicability of our method, we used a public dataset of head and neck cancer, yielding an average Dice coefficient of 68.71%, an average Jaccard index score of 57.72%, and an average 95HD of 7.88 mm.

Conclusions

The experimental results demonstrate the superiority of our MIM-CMFNet over existing semisupervised techniques. Our approach can achieve a performance similar to that of fully supervised segmentation methods while significantly reducing the data annotation cost by 80%, suggesting it is highly practicable for clinical application.

SUBMITTER: Shao M

PROVIDER: S-EPMC10895119 | biostudies-literature | 2024 Feb

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Semisupervised 3D segmentation of pancreatic tumors in positron emission tomography/computed tomography images using a mutual information minimization and cross-fusion strategy.

Shao Min M Cheng Chao C Hu Chengyuan C Zheng Jian J Zhang Bo B Wang Tao T Jin Gang G Liu Zhaobang Z Zuo Changjing C

Quantitative imaging in medicine and surgery 20240123 2

<h4>Background</h4>Accurate segmentation of pancreatic cancer tumors using positron emission tomography/computed tomography (PET/CT) multimodal images is crucial for clinical diagnosis and prognosis evaluation. However, deep learning methods for automated medical image segmentation require a substantial amount of manually labeled data, making it time-consuming and labor-intensive. Moreover, addition or simple stitching of multimodal images leads to redundant information, failing to fully exploit ...[more]

PMID: 38415108

Similar Datasets

Project description:BackgroundMicrocalcifications in atherosclerotic plaques are destabilizing, predict adverse cardiovascular events, and are associated with increased morbidity and mortality.18F-fluoride positron emission tomography (PET)/computed tomography (CT) imaging has demonstrated promise as a useful clinical diagnostic tool in identifying high-risk plaques; however, there is confusion as to the underlying mechanism of signal amplification seen in PET-positive, CT-negative image regions. This study tested the hypothesis that 18F-fluoride PET/CT can identify early microcalcifications.Methods18F-fluoride signal amplification derived from microcalcifications was validated against near-infrared fluorescence molecular imaging and histology using an in vitro 3-dimensional hydrogel collagen platform, ex vivo human specimens, and a mouse model of atherosclerosis.ResultsMicrocalcification size correlated inversely with collagen concentration. The 18F-fluoride ligand bound to microcalcifications formed by calcifying vascular smooth muscle cell derived extracellular vesicles in the in vitro 3-dimensional collagen system and exhibited an increasing signal with an increase in collagen concentration (0.25 mg/mL collagen -33.8×102±12.4×102 counts per minute; 0.5 mg/mL collagen -67.7×102±37.4×102 counts per minute; P=0.0014), suggesting amplification of the PET signal by smaller microcalcifications. We further incubated human atherosclerotic endarterectomy specimens with clinically relevant concentrations of 18F-fluoride. The 18F-fluoride ligand labeled microcalcifications in PET-positive, CT-negative regions of explanted human specimens as evidenced by 18F-fluoride PET/CT imaging, near-infrared fluorescence, and histological analysis. Additionally, the 18F-fluoride ligand identified micro and macrocalcifications in atherosclerotic aortas obtained from low-density lipoprotein receptor-deficient mice.ConclusionsOur results suggest that 18F-fluoride PET signal in PET-positive, CT-negative regions of human atherosclerotic plaques is the result of developing microcalcifications, and high surface area in regions of small microcalcifications may amplify PET signal.

Project description:BackgroundDeep learning has recently shown great potential in medical image reconstruction tasks. For positron emission tomography (PET) images, the direct reconstruction from raw data to radioactivity images using deep learning without any constraint may lead to the production of nonexistent structures. The aim of this study was to specifically develop and test a flexibly deep learning-based reconstruction network guided by any form of prior knowledge to achieve high quality and high reliability reconstruction.MethodsWe developed a novel prior information-guided reconstruction network (PIGRN) with a dual-channel generator and a 2-scale discriminator based on a conditional generative adversarial network (cGAN). Besides the raw data channel, an additional channel is provided in the generator for prior information (PI) to guide the training phase. The PI can be reconstructed images obtained via conventional methods, nuclear medical images from other modalities, attenuation correction maps from time-of-flight-PET (TOF-PET) data, or any other physical parameters. For this study, the reconstructed images generated by filtered back projection (FBP) were chosen as the input of the additional channel. To improve the image quality, a 2-scale discriminator was adopted which can focus on both the coarse and fine field of the reconstruction images. Experiments were carried out on both a simulation dataset and a real Sprague Dawley (SD) rat dataset.ResultsTwo classic deep learning-based reconstruction networks, including U-Net and Deep-PET, were compared in our study. Compared with these two methods, our method could provide much higher quality PET image reconstruction in the study of the simulation dataset. The peak signal-to-noise ratio (PSNR) value reached 31.8498, and the structure similarity index measure (SSIM) value reached 0.9754. The real study on SD rats indicated that the proposed network also has strong generalization ability.ConclusionsThe flexible PIGRN based on cGAN for PET images combines both raw data and PI. The results of comparison experiments and a generalization experiment based on simulation and SD rat datasets demonstrated that the proposed PIGRN has the ability to improve image quality and has strong generalization ability.

Dataset Information

Semisupervised 3D segmentation of pancreatic tumors in positron emission tomography/computed tomography images using a mutual information minimization and cross-fusion strategy.

Background

Methods

Results

Conclusions

Publications

Semisupervised 3D segmentation of pancreatic tumors in positron emission tomography/computed tomography images using a mutual information minimization and cross-fusion strategy.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets