Abstract:PurposeDiffuse large B‐cell lymphoma (DLBCL) is an aggressive type of lymphoma with high mortality and poor prognosis that especially has a high incidence in Asia. Accurate segmentation of DLBCL lesions is crucial for clinical radiation therapy. However, manual delineation of DLBCL lesions is tedious and time‐consuming. Automatic segmentation provides an alternative solution but is difficult for diffuse lesions without the sufficient utilization of multimodality information. Our work is the first study focusing on positron emission tomography and computed tomography (PET‐CT) feature fusion for the DLBCL segmentation issue. We aim to improve the fusion performance of complementary information contained in PET‐CT imaging with a hybrid learning module in the supervised convolutional neural network.MethodsFirst, two encoder branches extract single‐modality features, respectively. Next, the hybrid learning component utilizes them to generate spatial fusion maps which can quantify the contribution of complementary information. Such feature fusion maps are then concatenated with specific‐modality (i.e., PET and CT) feature maps to obtain a representation of the final fused feature maps in different scales. Finally, the reconstruction part of our network creates a prediction map of DLBCL lesions by integrating and up‐sampling the final fused feature maps from encoder blocks in different scales.ResultsThe ability of our method was evaluated to detect foreground and segment lesions in three independent body regions (nasopharynx, chest, and abdomen) of a set of 45 PET‐CT scans. Extensive ablation experiments compared our method to four baseline techniques for multimodality fusion (input‐level (IL) fusion, multichannel (MC) strategy, multibranch (MB) strategy, and quantitative weighting (QW) fusion). The results showed that our method achieved a high detection accuracy (99.63% in the nasopharynx, 99.51% in the chest, and 99.21% in the abdomen) and had the superiority in segmentation performance with the mean dice similarity coefficient (DSC) of 73.03% and the modified Hausdorff distance (MHD) of 4.39 mm, when compared with the baselines (DSC: IL: 53.08%, MC: 63.59%, MB: 69.98%, and QW: 72.19%; MHD: IL: 12.16 mm, MC: 6.46 mm, MB: 4.83 mm, and QW: 4.89 mm).ConclusionsA promising segmentation method has been proposed for the challenging DLBCL lesions in PET‐CT images, which improves the understanding of complementary information by feature fusion and may guide clinical radiotherapy. The statistically significant analysis based on P‐value calculation has indicated a degree of significant difference between our proposed method and other baselines (almost metrics: P < 0.05). This is a preliminary research using a small sample size, and we will collect data continually to achieve the larger verification study.

Dempster-Shafer Theory Empowered Deep Learning for Enhanced 3D PET-CT Medical Image Segmentation: A First Test on Automated Lesion Segmentation in Whole-body Dataset

Lymphoma segmentation from 3D PET-CT images using a deep evidential network

Deep PET/CT fusion with Dempster-Shafer theory for lymphoma segmentation

Whole-Body Lesion Segmentation in 18F-FDG PET/CT

Generalized Dice Focal Loss trained 3D Residual UNet for Automated Lesion Segmentation in Whole-Body FDG PET/CT Images

AutoPET Challenge III: Testing the Robustness of Generalized Dice Focal Loss trained 3D Residual UNet for FDG and PSMA Lesion Segmentation from Whole-Body PET/CT Images

Evidential segmentation of 3D PET/CT images

Exploring Vanilla U-Net for Lesion Segmentation from Whole-body FDG-PET/CT Scans

Enhancing Lesion Segmentation in PET/CT Imaging with Deep Learning and Advanced Data Preprocessing Techniques

Comprehensive framework for evaluation of deep neural networks in detection and quantification of lymphoma from PET/CT images: clinical insights, pitfalls, and observer agreement analyses

Whole-body tumor segmentation of 18F -FDG PET/CT using a cascaded and ensembled convolutional neural networks

PSR-Nets: Deep neural networks with prior shift regularization for PET/CT based automatic, accurate, and calibrated whole-body lymphoma segmentation

Multimodal Data-Driven Segmentation of Bone Metastasis Lesions in SPECT Bone Scans Using Deep Learning

DECTNet: Dual Encoder Network combined convolution and Transformer architecture for medical image segmentation

Lesion segmentation on 18F-fluciclovine PET/CT images using deep learning

Head and neck tumor segmentation convolutional neural network robust to missing PET/CT modalities using channel dropout

3D Lymphoma Segmentation on PET/CT Images Via Multi-Scale Information Fusion with Cross-Attention

Diffuse Large B-cell Lymphoma Segmentation in PET-CT Images Via Hybrid Learning for Feature Fusion

A cascaded deep network for automated tumor detection and segmentation in clinical PET imaging of diffuse large B-cell lymphoma

Edge of discovery: Enhancing breast tumor MRI analysis with boundary-driven deep learning

Automatic detection and segmentation of lesions in 18F-FDG PET/CT imaging of patients with Hodgkin lymphoma using 3D dense U-Net