Nasopharyngeal Carcinoma Segmentation Research Articles

Self-supervised learning on dual-sequence magnetic resonance imaging for automatic segmentation of nasopharyngeal carcinoma

Automating the segmentation of nasopharyngeal carcinoma (NPC) is crucial for therapeutic procedures but presents challenges given the hurdles in amassing extensively annotated datasets. Although previous studies have applied self-supervised learning to capitalize on unlabeled data to improve segmentation performance, these methods often overlooked the benefits of dual-sequence magnetic resonance imaging (MRI). In the present study, we incorporated self-supervised learning with a saliency transformation module using unlabeled dual-sequence MRI for accurate NPC segmentation. 44 labeled and 72 unlabeled patients were collected to develop and evaluate our network. Impressively, our network achieved a mean Dice similarity coefficient (DSC) of 0.77, which is consistent with a previous study that relied on a training set of 4,100 annotated cases. The results further revealed that our approach required minimal adjustments, primarily < 20% tweak in the DSC, to meet clinical standards. By enhancing the automatic segmentation of NPC, our method alleviates the annotation burden on oncologists, curbs subjectivity, and ensures reliable NPC delineation.

Domain knowledge-driven encoder–decoder for nasopharyngeal carcinoma segmentation

Accurate image segmentation is vital for the diagnosis and treatment of nasopharyngeal carcinoma (NPC). Although deep neural networks have shown promising performance in NPC segmentation, they depend on large-scale pixel-level annotation datasets for purely data-driven model training, which poses practical limitations. To address this challenge, this paper explores the domain knowledge, namely the expert knowledge from radiologists, and proposes a domain knowledge-driven encoder–decoder architecture. Specifically, the domain knowledge on image spatial information is formulated as Gaussian mixture distribution and then transformed into an optimal transport-based expert-prior regularization of the encoder, which enhances the model’s ability in capturing discriminative features. To project the encoded features onto pixel space and obtain the segmentation maps, a cross-scale feature refinement module is built in the decoder with theoretical justification, which integrates the domain knowledge that radiologists segment NPC in a gradual refinement process. Experimental results verify the effectiveness of the proposed method for NPC segmentation. Remarkably, despite using only 10% of pixel-level annotation data, the domain knowledge-driven model outperforms recent deep neural networks that use the entire training data.

Domain composition and attention network trained with synthesized unlabeled images for generalizable medical image segmentation

Despite that deep learning models have achieved remarkable performance in medical image segmentation, their performance is often limited on testing images from new centers with a domain shift. To achieve Domain Generalization (DG) for medical image segmentation, we propose a Domain Composition and Attention-based Network (DCA-Net) combined with structure- and style-based data augmentation that generates unlabeled synthetic images for training. First, DCA-Net represents features in one certain domain by a linear combination of a set of basis representations that are learned by parallel domain preceptors with a divergence constraint. The linear combination is used to calibrate the feature maps of an input image, which enables the model to generalize to unseen domains. Second, considering the number of domains and images for training is limited, we employ generative models to synthesize images with a higher structure diversity, and to leverage the unlabeled synthetic images, we introduce a consistency constraint for their predictions under style augmentation based on frequency amplitude mixture. Additionally, a Test-Time Frequency Augmentation (TTFA) is proposed to neutralize the domain shift from the target to source domains. Experimental results on two multi-domain datasets for fundus structure and nasopharyngeal carcinoma segmentation showed that: (1) our method significantly outperformed several existing DG methods, and (2) the model’s generalizability was largely improved by domain composition and attention modules; (3) by leveraging the unlabeled synthetic images and the TTFA, the model could better deal with images from unseen domains.

Deep Learning for Nasopharyngeal Carcinoma Segmentation in Magnetic Resonance Imaging: A Systematic Review and Meta-Analysis

Nasopharyngeal carcinoma is a significant health challenge that is particularly prevalent in Southeast Asia and North Africa. MRI is the preferred diagnostic tool for NPC due to its superior soft tissue contrast. The accurate segmentation of NPC in MRI is crucial for effective treatment planning and prognosis. We conducted a search across PubMed, Embase, and Web of Science from inception up to 20 March 2024, adhering to the PRISMA 2020 guidelines. Eligibility criteria focused on studies utilizing DL for NPC segmentation in adults via MRI. Data extraction and meta-analysis were conducted to evaluate the performance of DL models, primarily measured by Dice scores. We assessed methodological quality using the CLAIM and QUADAS-2 tools, and statistical analysis was performed using random effects models. The analysis incorporated 17 studies, demonstrating a pooled Dice score of 78% for DL models (95% confidence interval: 74% to 83%), indicating a moderate to high segmentation accuracy by DL models. Significant heterogeneity and publication bias were observed among the included studies. Our findings reveal that DL models, particularly convolutional neural networks, offer moderately accurate NPC segmentation in MRI. This advancement holds the potential for enhancing NPC management, necessitating further research toward integration into clinical practice.

MFNet: Meta-learning based on frequency-space mix for MRI segmentation in nasopharyngeal carcinoma.

Deep learning techniques have been applied to medical image segmentation and demonstrated expert-level performance. Due to the poor generalization abilities of the models in the deployment in different centres, common solutions, such as transfer learning and domain adaptation techniques, have been proposed to mitigate this issue. However, these solutions necessitate retraining the models with target domain data and annotations, which limits their deployment in clinical settings in unseen domains. We evaluated the performance of domain generalization methods on the task of MRI segmentation of nasopharyngeal carcinoma (NPC) by collecting a new dataset of 321 patients with manually annotated MRIs from two hospitals. We transformed the modalities of MRI, including T1WI, T2WI and CE-T1WI, from the spatial domain to the frequency domain using Fourier transform. To address the bottleneck of domain generalization in MRI segmentation of NPC, we propose a meta-learning approach based on frequency domain feature mixing. We evaluated the performance of MFNet against existing techniques for generalizing NPC segmentation in terms of Dice and MIoU. Our method evidently outperforms the baseline in handling the generalization of NPC segmentation. The MF-Net clearly demonstrates its effectiveness for generalizing NPC MRI segmentation to unseen domains (Dice = 67.59%, MIoU = 75.74% T1W1). MFNet enhances the model's generalization capabilities by incorporating mixed-feature meta-learning. Our approach offers a novel perspective to tackle the domain generalization problem in the field of medical imaging by effectively exploiting the unique characteristics of medical images.

SICNet: Learning selective inter-slice context via Mask-Guided Self-knowledge distillation for NPC segmentation

Accurate segmentation of nasopharyngeal carcinoma (NPC) in magnetic resonance (MR) images is crucial for radiotherapy planning. However, vanilla 2D/3D deep convolutional networks fail to gain satisfying NPC segmentation results, since 3D methods suffer from inter-slice discontinuities and 2D methods lack inter-slice context learning. To address this problem, this paper proposes a 2.5D learning Selective Inter-slice Context Network (SICNet) for NPC segmentation. The proposed SICNet comprises a main encoder that extracts target features from the middle slice and an auxiliary encoder that learns inter-slice context from adjacent slices. The features from two encoders are fused by the Inter-slice Context Fusion (ICF) module and then sent to the decoder for predictions. Furthermore, a knowledge distillation module named Mask-Guided Self-knowledge Distillation (MGSD) is proposed to help the auxiliary encoder learn context selectively. Specifically, with the guidance of mask labels, MGSD tells the auxiliary encoder where to look and how to learn, thus achieving selective inter-slice context learning and boosting segmentation accuracy. The proposed SICNet achieves a DSC of 74.38 ± 11.99, an HD95 of 9.31 ± 2.68, and an ASSD of 1.46 ± 0.76 on an in-house NPC dataset, outperforming other state-of-the-art segmentation methods.

A transformer-based multi-task deep learning model for simultaneous T-stage identification and segmentation of nasopharyngeal carcinoma.

Accurate tumor target contouring and T staging are vital for precision radiation therapy in nasopharyngeal carcinoma (NPC). Identifying T-stage and contouring the Gross tumor volume (GTV) manually is a laborious and highly time-consuming process. Previous deep learning-based studies have mainly been focused on tumor segmentation, and few studies have specifically addressed the tumor staging of NPC. To bridge this gap, we aim to devise a model that can simultaneously identify T-stage and perform accurate segmentation of GTV in NPC. We have developed a transformer-based multi-task deep learning model that can perform two tasks simultaneously: delineating the tumor contour and identifying T-stage. Our retrospective study involved contrast-enhanced T1-weighted images (CE-T1WI) of 320 NPC patients (T-stage: T1-T4) collected between 2017 and 2020 at our institution, which were randomly allocated into three cohorts for three-fold cross-validations, and conducted the external validation using an independent test set. We evaluated the predictive performance using the area under the receiver operating characteristic curve (ROC-AUC) and accuracy (ACC), with a 95% confidence interval (CI), and the contouring performance using the Dice similarity coefficient (DSC) and average surface distance (ASD). Our multi-task model exhibited sound performance in GTV contouring (median DSC: 0.74; ASD: 0.97 mm) and T staging (AUC: 0.85, 95% CI: 0.82-0.87) across 320 patients. In early T category tumors, the model achieved a median DSC of 0.74 and ASD of 0.98 mm, while in advanced T category tumors, it reached a median DSC of 0.74 and ASD of 0.96 mm. The accuracy of automated T staging was 76% (126 of 166) for early stages (T1-T2) and 64% (99 of 154) for advanced stages (T3-T4). Moreover, experimental results show that our multi-task model outperformed the other single-task models. This study emphasized the potential of multi-task model for simultaneously delineating the tumor contour and identifying T-stage. The multi-task model harnesses the synergy between these interrelated learning tasks, leading to improvements in the performance of both tasks. The performance demonstrates the potential of our work for delineating the tumor contour and identifying T-stage and suggests that it can be a practical tool for supporting clinical precision radiation therapy.

Dual-Reference Source-Free Active Domain Adaptation for Nasopharyngeal Carcinoma Tumor Segmentation across Multiple Hospitals.

Nasopharyngeal carcinoma (NPC) is a prevalent and clinically significant malignancy that predominantly impacts the head and neck area. Precise delineation of the Gross Tumor Volume (GTV) plays a pivotal role in ensuring effective radiotherapy for NPC. Despite recent methods that have achieved promising results on GTV segmentation, they are still limited by lacking carefully-annotated data and hard-to-access data from multiple hospitals in clinical practice. Although some unsupervised domain adaptation (UDA) has been proposed to alleviate this problem, unconditionally mapping the distribution distorts the underlying structural information, leading to inferior performance. To address this challenge, we devise a novel Sourece-Free Active Domain Adaptation framework to facilitate domain adaptation for the GTV segmentation task. Specifically, we design a dual reference strategy to select domain-invariant and domain-specific representative samples from a specific target domain for annotation and model fine-tuning without relying on source-domain data. Our approach not only ensures data privacy but also reduces the workload for oncologists as it just requires annotating a few representative samples from the target domain and does not need to access the source data. We collect a large-scale clinical dataset comprising 1057 NPC patients from five hospitals to validate our approach. Experimental results show that our method outperforms the previous active learning (e.g., AADA and MHPL) and UDA (e.g., Tent and CPR) methods, and achieves comparable results to the fully supervised upper bound, even with few annotations, highlighting the significant medical utility of our approach. In addition, there is no public dataset about multi-center NPC segmentation, we will release code and dataset for future research (Git).

SECP-Net: SE-Connection Pyramid Network for Segmentation of Organs at Risk with Nasopharyngeal Carcinoma.

Nasopharyngeal carcinoma (NPC) is a kind of malignant tumor. The accurate and automatic segmentation of computed tomography (CT) images of organs at risk (OAR) is clinically significant. In recent years, deep learning models represented by U-Net have been widely applied in medical image segmentation tasks, which can help to reduce doctors' workload. In the OAR segmentation of NPC, the sizes of the OAR are variable, and some of their volumes are small. Traditional deep neural networks underperform in segmentation due to the insufficient use of global and multi-size information. Therefore, a new SE-Connection Pyramid Network (SECP-Net) is proposed. For extracting global and multi-size information, the SECP-Net designs an SE-connection module and a pyramid structure for improving the segmentation performance, especially that of small organs. SECP-Net also uses an auto-context cascaded structure to further refine the segmentation results. Comparative experiments are conducted between SECP-Net and other recent methods on a private dataset with CT images of the head and neck and a public liver dataset. Five-fold cross-validation is used to evaluate the performance based on two metrics; i.e., Dice and Jaccard similarity. The experimental results show that SECP-Net can achieve SOTA performance in these two challenging tasks.

Quantitative pharmacokinetic parameter Ktrans map assists in regional segmentation of nasopharyngeal carcinoma in dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI)

Accurate segmentation of nasopharyngeal carcinoma (NPC) lesion areas from dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) facilitates subsequent diagnostic and prognostic workups. However, in previous studies, little consideration has been given to incorporating the pharmacokinetic feature Ktrans as auxiliary information for segmenting NPC in DCE-MRI. Therefore, this paper proposes the use of a pharmacokinetic extended Tofts and Kermode (ETK) model to obtain the Ktrans feature from DCE-MRI and combine it with MRI images for nasopharyngeal tumor segmentation. Additionally, this paper proposes a multi-input branch residual U-Net (MBRU-Net) model that effectively fuses DCE-MRI features and Ktrans features. The effectiveness of the multibranch network is validated by comparing MBRU-Net with ResU-Net with DCE-MRI + Ktrans data. Additionally, different models are trained with DCE-MRI and DCE-MRI + Ktrans data separately and compared to validate the effectiveness of multimodal data using the Dice coefficient. Our proposed MBRU-Net achieves the best Dice in this study (67.39 ± 15.79%), higher than ResU-Net's Dice (65.57 ± 17.52) based on DCE-MRI and Ktrans data. U-Net, SegNet, R2U-Net, and ResU-Net achieve better results in terms of segmenting tumor regions with DCE-MRI + Ktrans data than those of the corresponding models with DCE-MRI data alone, where U-Net has the best performance (DCE-MRI + Ktrans: DCE-MRI = 66.31 ± 17.80%: 61.10 ± 24.14%). The results show that it is beneficial to add a pharmacokinetic parametric (Ktrans) map as prior information to the conventional anatomical MRI-based segmentation task, and multibranch network structures perform better than single-branch network structures in terms of NPC segmentation.

Uncertainty-weighted and relation-driven consistency training for semi-supervised head-and-neck tumor segmentation

Accurately segmenting nasopharyngeal carcinoma (NPC), a type of head-and-neck tumor, in computed tomography (CT) images is essential for clinical treatment. However, due to the limited number of annotated samples, large shape variations, and boundary ambiguity, NPC segmentation remains a challenging task. Semi-supervised learning (SSL), which makes use of unlabeled data, has become a popular solution for NPC segmentation. In this paper, we propose a consistency training framework for semi-supervised NPC segmentation, which includes an uncertainty-weighted prediction consistency training (UPCT) strategy and a relation-driven consistency training (RCT) strategy. Our model is an encoder–decoder architecture consisting of a shared encoder, a main decoder, and several auxiliary decoders. We apply various perturbations to the shared encoder’s output to leverage the unlabeled data and enforce consistency between the predictions of the main and auxiliary decoders. To avoid being misled by unreliable outputs during training due to annotation scarcity, we introduce uncertainty estimation into our model. The UPCT strategy emphasizes predictions with high reliability and weakens unreliable predictions. Additionally, we propose the RCT strategy to explore intrinsic relations among samples in each mini-batch, encouraging the model to extract richer semantic information. Our extensive experiments on the in-house NPC dataset and the public 2017 ACDC dataset not only demonstrate the effectiveness of the UPCT and RCT strategies but also show the superiority and generalization of our method compared to the state-of-the-art methods.

MSU-Net: Multi-scale Sensitive U-Net based on pixel-edge-region level collaborative loss for nasopharyngeal MRI segmentation

Radiotherapy is the traditional treatment of early nasopharyngeal carcinoma (NPC). Automatic accurate segmentation of risky lesions in the nasopharynx is crucial in radiotherapy. U-Net has been proved its effective medical image segmentation ability. However, the great difference in the structure and size of nasopharynx among different patients requires a network that pays more attention to multi-scale information. In this paper, we propose a multi-scale sensitive U-Net (MSU-Net) based on pixel-edge-region level collaborative loss (LCo−PER) for NPC segmentation task. A series of novel feature fusion modules based on spatial continuity and multi-scale semantic are proposed for extracting multi-level features while efficiently searching for all size lesions. A spatial continuity information extraction module (SCIEM) is proposed for effectively using the spatial continuity information of context slices to search small lesions. And a multi-scale semantic feature extraction module (MSFEM) is proposed for extracting features of different receptive fields. LCo−PER is proposed for the network training which makes network model could take into account the size of different lesions. The global Dice, Precision, Recall and IOU of the testing set are 84.50%, 97.48%, 84.33% and 82.41%, respectively. The results show that our method is better than the other state-of-the-art methods for NPC segmentation which obtain higher accuracy and effective segmentation performance.

BSMM-Net: Multi-modal neural network based on bilateral symmetry for nasopharyngeal carcinoma segmentation.

Automatically and accurately delineating the primary nasopharyngeal carcinoma (NPC) tumors in head magnetic resonance imaging (MRI) images is crucial for patient staging and radiotherapy. Inspired by the bilateral symmetry of head and complementary information of different modalities, a multi-modal neural network named BSMM-Net is proposed for NPC segmentation. First, a bilaterally symmetrical patch block (BSP) is used to crop the image and the bilaterally flipped image into patches. BSP can improve the precision of locating NPC lesions and is a simulation of radiologist locating the tumors with the bilateral difference of head in clinical practice. Second, modality-specific and multi-modal fusion features (MSMFFs) are extracted by the proposed MSMFF encoder to fully utilize the complementary information of T1- and T2-weighted MRI. The MSMFFs are then fed into the base decoder to aggregate representative features and precisely delineate the NPC. MSMFF is the output of MSMFF encoder blocks, which consist of six modality-specific networks and one multi-modal fusion network. Except T1 and T2, the other four modalities are generated from T1 and T2 by the BSP and DT modal generate block. Third, the MSMFF decoder with similar structure to the MSMFF encoder is deployed to supervise the encoder during training and assure the validity of the MSMFF from the encoder. Finally, experiments are conducted on the dataset of 7633 samples collected from 745 patients. The global DICE, precision, recall and IoU of the testing set are 0.82, 0.82, 0.86, and 0.72, respectively. The results show that the proposed model is better than the other state-of-the-art methods for NPC segmentation. In clinical diagnosis, the BSMM-Net can give precise delineation of NPC, which can be used to schedule the radiotherapy.

Lightweight Compound Scaling Network for Nasopharyngeal Carcinoma Segmentation from MR Images.

Nasopharyngeal carcinoma (NPC) is a category of tumours with a high incidence in head-and-neck. To treat nasopharyngeal cancer, doctors invariably need to perform focal segmentation. However, manual segmentation is time consuming and laborious for doctors and the existing automatic segmentation methods require large computing resources, which makes some small and medium-sized hospitals unaffordable. To enable small and medium-sized hospitals with limited computational resources to run the model smoothly and improve the accuracy of structure, we propose a new LW-UNet network. The network utilises lightweight modules to form the Compound Scaling Encoder and combines the benefits of UNet to make the model both lightweight and accurate. Our model achieves a high accuracy with a Dice coefficient value of 0.813 with 3.55 M parameters and 7.51 G of FLOPs within 0.1 s (testing time in GPU), which is the best result compared with four other state-of-the-art models.

TG-Net: Combining transformer and GAN for nasopharyngeal carcinoma tumor segmentation based on total-body uEXPLORER PET/CT scanner

Nasopharyngeal carcinoma (NPC) is a malignant tumor, and the main treatment is radiotherapy. Accurate delineation of the target tumor is essential for radiotherapy of NPC. NPC tumors are small in size and vary widely in shape and structure, making it a time-consuming and laborious task for even experienced radiologists to manually outline tumors. However, the segmentation performance of current deep learning models is not satisfactory, mainly manifested by poor segmentation boundaries. To solve this problem, this paper proposes a segmentation method for nasopharyngeal carcinoma based on dynamic PET-CT image data, whose input data include CT, PET, and parametric images (Ki images). This method uses a generative adversarial network with a modified UNet integrated with a Transformer as the generator (TG-Net) to achieve automatic segmentation of NPC on combined CT-PET-Ki images. In the coding stage, TG-Net uses moving windows to replace traditional pooling operations to obtain patches of different sizes, which can reduce information loss in the coding process. Moreover, the introduction of Transformer can make the network learn more representative features and improve the discriminant ability of the model, especially for tumor boundaries. Finally, the results of fivefold cross validation with an average Dice similarity coefficient score of 0.9135 show that our method has good segmentation performance. Comparative experiments also show that our network structure is superior to the most advanced methods in the segmentation of NPC. In addition, this work is the first to use Ki images to assist tumor segmentation. We also demonstrated the usefulness of adding Ki images to aid in tumor segmentation.

CAFS: An Attention-Based Co-Segmentation Semi-Supervised Method for Nasopharyngeal Carcinoma Segmentation.

Accurate segmentation of nasopharyngeal carcinoma is essential to its treatment effect. However, there are several challenges in existing deep learning-based segmentation methods. First, the acquisition of labeled data are challenging. Second, the nasopharyngeal carcinoma is similar to the surrounding tissues. Third, the shape of nasopharyngeal carcinoma is complex. These challenges make the segmentation of nasopharyngeal carcinoma difficult. This paper proposes a novel semi-supervised method named CAFS for automatic segmentation of nasopharyngeal carcinoma. CAFS addresses the above challenges through three mechanisms: the teacher–student cooperative segmentation mechanism, the attention mechanism, and the feedback mechanism. CAFS can use only a small amount of labeled nasopharyngeal carcinoma data to segment the cancer region accurately. The average DSC value of CAFS is 0.8723 on the nasopharyngeal carcinoma segmentation task. Moreover, CAFS has outperformed the state-of-the-art nasopharyngeal carcinoma segmentation methods in the comparison experiment. Among the compared state-of-the-art methods, CAFS achieved the highest values of DSC, Jaccard, and precision. In particular, the DSC value of CAFS is 7.42% higher than the highest DSC value in the state-of-the-art methods.

SeqSeg: A sequential method to achieve nasopharyngeal carcinoma segmentation free from background dominance.

Reliable nasopharyngeal carcinoma (NPC) segmentation plays an important role in radiotherapy planning. However, recent deep learning methods fail to achieve satisfactory NPC segmentation in magnetic resonance (MR) images, since NPC is infiltrative and typically has a small or even tiny volume with indistinguishable border, making it indiscernible from tightly connected surrounding tissues from immense and complex backgrounds. To address such background dominance problems, this paper proposes a sequential method (SeqSeg) to achieve accurate NPC segmentation. Specifically, the proposed SeqSeg is devoted to solving the problem at two scales: the instance level and feature level. At the instance level, SeqSeg is forced to focus attention on the tumor and its surrounding tissue through the deep Q-learning (DQL)-based NPC detection model by prelocating the tumor and reducing the scale of the segmentation background. Next, at the feature level, SeqSeg uses high-level semantic features in deeper layers to guide feature learning in shallower layers, thus directing the channel-wise and region-wise attention to mine tumor-related features to perform accurate segmentation. The performance of our proposed method is evaluated by extensive experiments on the large NPC dataset containing 1101 patients. The experimental results demonstrated that the proposed SeqSeg not only outperforms several state-of-the-art methods but also achieves better performance in multi-device and multi-center datasets.

The contrast-enhanced MRI can be substituted by unenhanced MRI in identifying and automatically segmenting primary nasopharyngeal carcinoma with the aid of deep learning models: An exploratory study in large-scale population of endemic area

Background and objectivesAdministration of contrast is not desirable for all cases in clinical setting, and no consensus in sequence selection for deep learning model development has been achieved, thus we aim to explore whether contrast-enhanced magnetic resonance imaging (ceMRI) can be substituted in the identification and segmentation of nasopharyngeal carcinoma (NPC) with the aid of deep learning models in a large-scale cohort. MethodsA total of 4478 eligible individuals were randomly split into training, validation and test sets, and self-constrained 3D DenseNet and V-Net models were developed using axial T1-weighted imaging (T1WI), T2WI or enhanced T1WI (T1WIC) images separately. The differential diagnostic performance between NPC and benign hyperplasia were compared among models using chi-square test. Segmentation evaluation metrics, including dice similarity coefficient (DSC) and average surface distance (ASD), were compared using paired student's t-test between T1WIC and T1WI or T2WI models or M_T1/T2, a merged output of malignant region derived from T1WI and T2WI models. ResultsAll models exhibited similar satisfactory diagnostic performance in discriminating NPC from benign hyperplasia, all attaining overall accuracy over 99.00% in all T stages of NPC. And T1WIC model exhibited similar average DSC and ASD with those of M_T1/T2 (DSC, 0.768±0.070 vs 0.764±0.070; ASD, 1.573±10.954 mm vs 1.626±10.975 mm 1.626±0.975 mm vs 1.573±0.954 mm, all p > 0.0167) in primary NPC using DenseNet, but yielded a significantly higher DSC and lower ASD than either T1WI model or T2WI model (DSC, 0.759±0.065 or 0.755±0.071; ASD, 1.661±0.898 mm or 1.722±1.133 mm, respectively, all p < 0.01) in the entire test set of NPC cohort. Moreover, the average DSCs and ASDs were not statistically significant between T1WIC model and M_T1/T2 in both.

AttR2U-Net: A Fully Automated Model for MRI Nasopharyngeal Carcinoma Segmentation Based on Spatial Attention and Residual Recurrent Convolution.

Radiotherapy is an essential method for treating nasopharyngeal carcinoma (NPC), and the segmentation of NPC is a crucial process affecting the treatment. However, manual segmentation of NPC is inefficient. Besides, the segmentation results of different doctors might vary considerably. To improve the efficiency and the consistency of NPC segmentation, we propose a novel AttR2U-Net model which automatically and accurately segments nasopharyngeal carcinoma from MRI images. This model is based on the classic U-Net and incorporates advanced mechanisms such as spatial attention, residual connection, recurrent convolution, and normalization to improve the segmentation performance. Our model features recurrent convolution and residual connections in each layer to improve its ability to extract details. Moreover, spatial attention is fused into the network by skip connections to pinpoint cancer areas more accurately. Our model achieves a DSC value of 0.816 on the NPC segmentation task and obtains the best performance compared with six other state-of-the-art image segmentation models.

MRI Image Segmentation of Nasopharyngeal Carcinoma Using Multi-Scale Cascaded Fully Convolutional Network

Nasopharyngeal carcinoma (NPC) is one of the most common malignant tumors of the head and neck, and its incidence is the highest all around the world. Intensive radiotherapy using computer-aided diagnosis is the best technique for the treatment of NPC. The key step of radiotherapy is the delineation of the target areas and organs at risk, that is, tumor images segmentation. We proposed the segmentation method of NPC image based on multi-scale cascaded fully convolutional network. It used cascaded network and multi-scale feature for a coarse-to-fine segmentation to improve the segmentation effect. In coarse segmentation, image blocks and data augmentation were used to compensate for the shortage of training samples. In fine segmentation, Atrous Spatial Pyramid Pooling (ASPP) was used to increase the receptive field and image feature transfer, which was added in the Dense block of DenseNet. In the process of up-sampling, the features of multiple views were fused to reduce false positive samples. Additionally, in order to improve the class imbalance problem, Focal Loss was used to weight the loss function of tumor voxel distance because it could reduce the weight of background category samples. The cascaded network can alleviate the problem of gradient disappearance and obtain a smoother boundary. The experimental results were quantitatively analyzed by DSC, ASSD and F1_score values, and the results showed that the proposed method was effective for nasopharyngeal carcinoma segmentation compared with other methods in this paper.