Hausdorff Distance Values Research Articles

Automated Organ Segmentation for Radiation Therapy: A Comparative Analysis of AI-Based Tools Versus Manual Contouring in Korean Cancer Patients.

Accurate delineation of tumors and organs at risk (OARs) is crucial for intensity-modulated radiation therapy. This study aimed to evaluate the performance of OncoStudio, an AI-based auto-segmentation tool developed for Korean patients, compared with Protégé AI, a globally developed tool that uses data from Korean cancer patients. A retrospective analysis of 1200 Korean cancer patients treated with radiotherapy was conducted. Auto-contours generated via OncoStudio and Protégé AI were compared with manual contours across the head and neck and thoracic, abdominal, and pelvic organs. Accuracy was assessed using the Dice similarity coefficient (DSC), mean surface distance (MSD), and 95% Hausdorff distance (HD). Feedback was obtained from 10 participants, including radiation oncologists, residents, and radiation therapists, via an online survey with a Turing test component. OncoStudio outperformed Protégé AI in 85% of the evaluated OARs (p < 0.001). For head and neck organs, OncoStudio achieved a similar DSC (0.70 vs. 0.70, p = 0.637) but significantly lower MSD and 95% HD values (p < 0.001). In thoracic organs, OncoStudio performed excellently in 90% of cases, with a significantly greater DSC (male: 0.87 vs. 0.82, p < 0.001; female: 0.95 vs. 0.87, p < 0.001). OncoStudio also demonstrated superior accuracy in abdominal (DSC 0.88 vs. 0.81, p < 0.001) and pelvic organs (male: DSC 0.95 vs. 0.85, p < 0.001; female: DSC 0.82 vs. 0.73, p < 0.001). Clinicians favored OncoStudio in 70% of cases, with 90% endorsing its clinical suitability for Korean patients. OncoStudio, which is tailored for Korean patients, demonstrated superior segmentation accuracy across multiple anatomical regions, suggesting its suitability for radiotherapy planning in this population.

Artificial intelligence-assisted delineation for postoperative radiotherapy in patients with lung cancer: a prospective, multi-center, cohort study

BackgroundPostoperative radiotherapy (PORT) is an important treatment for lung cancer patients with poor prognostic features, but accurate delineation of the clinical target volume (CTV) and organs at risk (OARs) is challenging and time-consuming. Recently, deep learning-based artificial intelligent (AI) algorithms have shown promise in automating this process.ObjectiveTo evaluate the clinical utility of a deep learning-based auto-segmentation model for AI-assisted delineating CTV and OARs in patients undergoing PORT, and to compare its accuracy and efficiency with manual delineation by radiation oncology residents from different levels of medical institutions.MethodsWe previously developed an AI auto-segmentation model in 664 patients and validated its contouring performance in 149 patients. In this multi-center, validation trial, we prospectively involved 55 patients and compared the accuracy and efficiency of 3 contouring methods: (i) unmodified AI auto-segmentation, (ii) fully manual delineation by junior radiation oncology residents from different medical centers, and (iii) manual modifications based on AI segmentation model (AI-assisted delineation). The ground truth of CTV and OARs was delineated by 3 senior radiation oncologists. Contouring accuracy was evaluated by Dice similarity coefficient (DSC), Hausdorff distance (HD), and mean distance of agreement (MDA). Inter-observer consistency was assessed by volume and coefficient of variation (CV).ResultsAI-assisted delineation achieved significantly higher accuracy compared to unmodified AI auto-contouring and fully manual delineation by radiation oncologists, with median HD, MDA, and DCS values of 20.03 vs. 21.55 mm, 2.57 vs. 3.06 mm, 0.745 vs. 0.703 (all P&amp;lt;0.05) for CTV, respectively. The results of OARs contours were similar. CV for OARs was reduced by approximately 50%. In addition to better contouring accuracy, the AI-assisted delineation significantly decreased the consuming time and improved the efficiency.ConclusionAI-assisted CTV and OARs delineation for PORT significantly improves the accuracy and efficiency in the real-world setting, compared with pure AI auto-segmentation or fully manual delineation by junior oncologists. AI-assisted approach has promising clinical potential to enhance the quality of radiotherapy planning and further improve treatment outcomes of patients with lung cancer.

A cascaded FAS-UNet+ framework with iterative optimization strategy for segmentation of organs at risk.

Segmentation of organs at risks (OARs) in the thorax plays a critical role in radiation therapy for lung and esophageal cancer. Although automatic segmentation of OARs has been extensively studied, it remains challenging due to the varying sizes and shapes of organs, as well as the low contrast between the target and background. This paper proposes a cascaded FAS-UNet+ framework, which integrates convolutional neural networks and nonlinear multi-grid theory to solve a modified Mumford-shah model for segmenting OARs. This framework is equipped with an enhanced iteration block, a coarse-to-fine multiscale architecture, an iterative optimization strategy, and a model ensemble technique. The enhanced iteration block aims to extract multiscale features, while the cascade module is used to refine coarse segmentation predictions. The iterative optimization strategy improves the network parameters to avoid unfavorable local minima. An efficient data augmentation method is also developed to train the network, which significantly improves its performance. During the prediction stage, a weighted ensemble technique combines predictions from multiple models to refine the final segmentation. The proposed cascaded FAS-UNet+ framework was evaluated on the SegTHOR dataset, and the results demonstrate significant improvements in Dice score and Hausdorff Distance (HD). The Dice scores were 95.22%, 95.68%, and HD values were 0.1024, and 0.1194 for the segmentations of the aorta and heart in the official unlabeled dataset, respectively. Our code and trained models are available at https://github.com/zhuhui100/C-FASUNet-plus .

FreqSNet: a multiaxial integration of frequency and spatial domains for medical image segmentation

Objective. In recent years, convolutional neural networks, which typically focus on extracting spatial domain features, have shown limitations in learning global contextual information. However, frequency domain can offer a global perspective that spatial domain methods often struggle to capture. To address this limitation, we propose FreqSNet, which leverages both frequency and spatial features for medical image segmentation. Approach. To begin, we propose a frequency-space representation aggregation block (FSRAB) to replace conventional convolutions. FSRAB contains three frequency domain branches to capture global frequency information along different axial combinations, while a convolutional branch is designed to interact information across channels in local spatial features. Secondly, the multiplex expansion attention block extracts long-range dependency information using dilated convolutional blocks, while suppressing irrelevant information via attention mechanisms. Finally, the introduced Feature Integration Block enhances feature representation by integrating semantic features that fuse spatial and channel positional information. Main results. We validated our method on 5 public datasets, including BUSI, CVC-ClinicDB, CVC-ColonDB, ISIC-2018, and Luna16. On these datasets, our method achieved Intersection over Union (IoU) scores of 75.46%, 87.81%, 79.08%, 84.04%, and 96.99%, and Hausdorff distance values of 22.22 mm, 13.20 mm, 13.08 mm, 13.51 mm, and 5.22 mm, respectively. Compared to other state-of-the-art methods, our FreqSNet achieves better segmentation results. Significance. Our method can effectively combine frequency domain information with spatial domain features, enhancing the segmentation performance and generalization capability in medical image segmentation tasks.

Retrospective Comparison of Geometrical Accuracy among Atlas-based Auto-segmentation, Deep Learning Auto-segmentation, and Deformable Image Registration in the Treatment Replanning for Adaptive Radiotherapy of Head-and-Neck Cancer.

This study aimed to evaluate the geometrical accuracy of atlas-based auto-segmentation (ABAS), deformable image registration (DIR), and deep learning auto-segmentation (DLAS) in adaptive radiotherapy (ART) for head-and-neck cancer (HNC). Seventeen patients who underwent replanning for ART were retrospectively studied, and delineated contours on their replanning computed tomography (CT2) images were delineated. For DIR, the planning CT image (CT1) of the evaluated patients was utilized. In contrast, ABAS was performed using an atlas dataset comprising 30 patients who were not part of the evaluated group. DLAS was trained with 143 patients from different patients from the evaluated patients. The ABAS model was improved, and a modified ABAS (mABAS) was created by adding the evaluated patients' own CT1 to the atlas datasets of ABAS (number of patients of the atlas dataset, 31). The geometrical accuracy of DIR, DLAS, ABAS, and mABAS was evaluated. The Dice similarity coefficient in DIR was the highest, at >0.8 at all organs at risk. The mABAS was delineated slightly more accurately than the standard ABAS. There was no significant difference between ABAS and DLAS in delineation accuracy. DIR had the lowest Hausdorff distance (HD) value (within 10 mm). The HD values in ABAS, mABAS, and DLAS were within 16 mm. DIR delineation is the most geometrically accurate ART for HNC.

Radiomics as a measure superior to common similarity metrics for tumor segmentation performance evaluation.

To propose radiomics features as a superior measure for evaluating the segmentation ability of physicians and auto-segmentation tools and to compare its performance with the most commonly used metrics: Dice similarity coefficient (DSC), surface Dice similarity coefficient (sDSC), and Hausdorff distance (HD). The data of 10 lung cancer patients' CT images with nine tumor segmentations per tumor were downloaded from the RIDER (Reference Database to Evaluate Response) database. Radiomics features of 90 segmented tumors were extracted using the PyRadiomics program. The intraclass correlation coefficient (ICC) of radiomics features were used to evaluate the segmentation similarity and compare their performance with DSC, sDSC, and HD. We calculated one ICC per radiomics feature and per tumor for nine segmentations and 36 ICCs per radiomics feature for 36 pairs of nine segmentations. Meanwhile, there were 360 DSC, sDSC, and HD values calculated for 36 pairs for 10 tumors. The ICC of radiomics features exhibited greater sensitivity to segmentation changes than DSC and sDSC. The ICCs of the wavelet-LLL first order Maximum, wavelet-LLL glcm MCC, wavelet-LLL glcm Cluster Shade features ranged from 0.130 to 0.997, 0.033 to 0.978, and 0.160 to 0.998, respectively. On the other hand, all DSC and sDSC were larger than 0.778 and 0.700, respectively, while HD varied from 0 to 1.9mm. The results indicated that the radiomics features could capture subtle variations in tumor segmentation characteristics, which could not be easily detected by DSC and sDSC. This study demonstrates the superiority of radiomics features with ICC as a measure for evaluating a physician's tumor segmentation ability and the performance of auto-segmentation tools. Radiomics features offer a more sensitive and comprehensive evaluation, providing valuable insights into tumor characteristics. Therefore, the new metrics can be used to evaluate new auto-segmentation methods and enhance trainees' segmentation skills in medical training and education.

Dummy run for planning of isotoxic dose-escalated radiation therapy for glioblastoma used in the PRIDE trial (NOA-28; ARO-2024-01; AG-NRO-06)

BackgroundThe PRIDE trial (NOA-28; ARO-2024-01; AG-NRO-06; NCT05871021) is designed to determine whether a dose escalation with 75.0 Gy in 30 fractions can enhance the median overall survival (OS) in patients with methylguanine methyltransferase (MGMT) promotor unmethylated glioblastoma compared to historical median OS rates, while being isotoxic to historical cohorts through the addition of concurrent bevacizumab (BEV). To ensure protocol-compliant irradiation planning with all study centers, a dummy run was planned and the plan quality was evaluated. MethodsA suitable patient case was selected and the computed tomography (CT), magnetic resonance imaging (MRI) and O-(2-[18F]fluoroethyl)-L-tyrosine (FET) positron emission tomography (PET) contours were made available. Participants at the various intended study sites performed radiation planning according to the PRIDE clinical trial protocol. The treatment plans and dose grids were uploaded as Digital Imaging and Communications in Medicine (DICOM) files to a cloud-based platform. Plan quality and protocol adherence were analyzed using a standardized checklist, scorecards and indices such as Dice Score (DSC) and Hausdorff Distance (HD). ResultsMedian DSC was 0.89, 0.90, 0.88 for PTV60, PTV60ex (planning target volume receiving 60.0 Gy for the standard and the experimental plan, respectively) and PTV75 (PTV receiving 75.0 Gy in the experimental plan), respectively. Median HD values were 17.0 mm, 13.9 mm and 12.1 mm, respectively. These differences were also evident in the volumes: The PTV60 had a volume range of 219.1–391.3 cc (median: 261.9 cc) for the standard plans, while the PTV75 volumes for the experimental plans ranged from 71.5–142.7 cc (median: 92.3 cc). The structures with the largest deviations in Dice score were the pituitary gland (median 0.37, range 0.00–0.69) and the right lacrimal gland (median 0.59, range 0.42–0.78). ConclusionsThe deviations revealed the necessity of systematic trainings with appropriate feedback before the start of clinical trials in radiation oncology and the constant monitoring of protocol compliance throw-out the study. Trial registrationNCT05871021

A deep learning-based 3D Prompt-nnUnet model for automatic segmentation in brachytherapy of postoperative endometrial carcinoma.

To create and evaluate a three-dimensional (3D) Prompt-nnUnet module that utilizes the prompts-based model combined with 3D nnUnet for producing the rapid and consistent autosegmentation of high-risk clinical target volume (HR CTV) and organ at risk (OAR) in high-dose-rate brachytherapy (HDR BT) for patients with postoperative endometrial carcinoma (EC). On two experimental batches, a total of 321 computed tomography (CT) scans were obtained for HR CTV segmentation from 321 patients with EC, and 125 CT scans for OARs segmentation from 125 patients. The numbers of training/validation/test were 257/32/32 and 87/13/25 for HR CTV and OARs respectively. A novel comparison of the deep learning neural network 3D Prompt-nnUnet and 3D nnUnet was applied for HR CTV and OARs segmentation. Three-fold cross validation and several quantitative metrics were employed, including Dice similarity coefficient (DSC), Hausdorff distance (HD), 95th percentile of Hausdorff distance (HD95%), and intersection over union (IoU). The Prompt-nnUnet included two forms of parameters Predict-Prompt (PP) and Label-Prompt (LP), with the LP performing most similarly to the experienced radiation oncologist and outperforming the less experienced ones. During the testing phase, the mean DSC values for the LP were 0.96±0.02, 0.91±0.02, and 0.83±0.07 for HR CTV, rectum and urethra, respectively. The mean HD values (mm) were 2.73±0.95, 8.18±4.84, and 2.11±0.50, respectively. The mean HD95% values (mm) were 1.66±1.11, 3.07±0.94, and 1.35±0.55, respectively. The mean IoUs were 0.92±0.04, 0.84±0.03, and 0.71±0.09, respectively. A delineation time<2.35s per structure in the new model was observed, which was available to save clinician time. The Prompt-nnUnet architecture, particularly the LP, was highly consistent with ground truth (GT) in HR CTV or OAR autosegmentation, reducing interobserver variability and shortening treatment time.

A multi-modal vision-language pipeline strategy for contour quality assurance and adaptive optimization

Objective. Accurate delineation of organs-at-risk (OARs) is a critical step in radiotherapy. The deep learning generated segmentations usually need to be reviewed and corrected by oncologists manually, which is time-consuming and operator-dependent. Therefore, an automated quality assurance (QA) and adaptive optimization correction strategy was proposed to identify and optimize ‘incorrect’ auto-segmentations. Approach. A total of 586 CT images and labels from nine institutions were used. The OARs included the brainstem, parotid, and mandible. The deep learning generated contours were compared with the manual ground truth delineations. In this study, we proposed a novel contour quality assurance and adaptive optimization (CQA-AO) strategy, which consists of the following three main components: (1) the contour QA module classified the deep learning generated contours as either accepted or unaccepted; (2) the unacceptable contour categories analysis module provided the potential error reasons (five unacceptable category) and locations (attention heatmaps); (3) the adaptive correction of unacceptable contours module integrate vision-language representations and utilize convex optimization algorithms to achieve adaptive correction of ‘incorrect’ contours. Main results. In the contour QA tasks, the sensitivity (accuracy, precision) of CQA-AO strategy reached 0.940 (0.945, 0.948), 0.962 (0.937, 0.913), and 0.967 (0.962, 0.957) for brainstem, parotid and mandible, respectively. The unacceptable contour category analysis, the (FI,AccI,Fmicro,Fmacro) of CQA-AO strategy reached (0.901, 0.763, 0.862, 0.822), (0.855, 0.737, 0.837, 0.784), and (0.907, 0.762, 0.858, 0.821) for brainstem, parotid and mandible, respectively. After adaptive optimization correction, the DSC values of brainstem, parotid and mandible have been improved by 9.4%, 25.9%, and 13.5%, and Hausdorff distance values decreased by 62%, 70.6%, and 81.6%, respectively. Significance. The proposed CQA-AO strategy, which combines QA of contour and adaptive optimization correction for OARs contouring, demonstrated superior performance compare to conventional methods. This method can be implemented in the clinical contouring procedures and improve the efficiency of delineating and reviewing workflow.

Need for Objective Task-Based Evaluation of Image Segmentation Algorithms for Quantitative PET: A Study with ACRIN 6668/RTOG 0235 Multicenter Clinical Trial Data.

Reliable performance of PET segmentation algorithms on clinically relevant tasks is required for their clinical translation. However, these algorithms are typically evaluated using figures of merit (FoMs) that are not explicitly designed to correlate with clinical task performance. Such FoMs include the Dice similarity coefficient (DSC), the Jaccard similarity coefficient (JSC), and the Hausdorff distance (HD). The objective of this study was to investigate whether evaluating PET segmentation algorithms using these task-agnostic FoMs yields interpretations consistent with evaluation on clinically relevant quantitative tasks. Methods: We conducted a retrospective study to assess the concordance in the evaluation of segmentation algorithms using the DSC, JSC, and HD and on the tasks of estimating the metabolic tumor volume (MTV) and total lesion glycolysis (TLG) of primary tumors from PET images of patients with non-small cell lung cancer. The PET images were collected from the American College of Radiology Imaging Network 6668/Radiation Therapy Oncology Group 0235 multicenter clinical trial data. The study was conducted in 2 contexts: (1) evaluating conventional segmentation algorithms, namely those based on thresholding (SUVmax40% and SUVmax50%), boundary detection (Snakes), and stochastic modeling (Markov random field-Gaussian mixture model); (2) evaluating the impact of network depth and loss function on the performance of a state-of-the-art U-net-based segmentation algorithm. Results: Evaluation of conventional segmentation algorithms based on the DSC, JSC, and HD showed that SUVmax40% significantly outperformed SUVmax50%. However, SUVmax40% yielded lower accuracy on the tasks of estimating MTV and TLG, with a 51% and 54% increase, respectively, in the ensemble normalized bias. Similarly, the Markov random field-Gaussian mixture model significantly outperformed Snakes on the basis of the task-agnostic FoMs but yielded a 24% increased bias in estimated MTV. For the U-net-based algorithm, our evaluation showed that although the network depth did not significantly alter the DSC, JSC, and HD values, a deeper network yielded substantially higher accuracy in the estimated MTV and TLG, with a decreased bias of 91% and 87%, respectively. Additionally, whereas there was no significant difference in the DSC, JSC, and HD values for different loss functions, up to a 73% and 58% difference in the bias of the estimated MTV and TLG, respectively, existed. Conclusion: Evaluation of PET segmentation algorithms using task-agnostic FoMs could yield findings discordant with evaluation on clinically relevant quantitative tasks. This study emphasizes the need for objective task-based evaluation of image segmentation algorithms for quantitative PET.

NnUNet for Automatic Kidney and Cyst Segmentation in Autosomal Dominant Polycystic Kidney Disease.

Autosomal Dominant Polycystic Kidney Disease (ADPKD) is a genetic disorder that causes uncontrolled kidney cyst growth, leading to kidney volume enlargement and renal function loss over time. Total kidney volume (TKV) and cyst burdens have been used as prognostic imaging biomarkers for ADPKD. This study aimed to evaluate nnUNet for automatic kidney and cyst segmentation in T2-weighted (T2W) MRI images of ADPKD patients. 756 kidney images were retrieved from 95 patients in the Consortium for Radiologic Imaging Studies of Polycystic Kidney Disease (CRISP) cohort (95 patients × 2 kidneys × 4 follow-up scans). The nnUNet model was trained, validated, and tested on 604, 76, and 76 images, respectively. In contrast, all images of each patient were exclusively assigned to either the training, validation, or test sets to minimize evaluation bias. The kidney and cyst regions defined using a semi-automatic method were employed as ground truth. The model performance was assessed using the Dice Similarity Coefficient (DSC), the intersection over union (IoU) score, and the Hausdorff distance (HD). The test DSC values were 0.96±0.01 (mean±SD) and 0.90±0.05 for kidney and cysts, respectively. Similarly, the IoU scores were 0.91± 0.09 and 0.81±0.06, and the HD values were 12.49±8.71 mm and 12.04±10.41 mm, respectively, for kidney and cyst segmentation. The nnUNet model is a reliable tool to automatically determine kidney and cyst volumes in T2W MRI images for ADPKD prognosis and therapy monitoring.

A modified U-Net convolutional neural network for segmenting periprostatic adipose tissue based on contour feature learning

ObjectiveThis study trains a U-shaped fully convolutional neural network (U-Net) model based on peripheral contour measures to achieve rapid, accurate, automated identification and segmentation of periprostatic adipose tissue (PPAT). MethodsCurrently, no studies are using deep learning methods to discriminate and segment periprostatic adipose tissue. This paper proposes a novel and modified, U-shaped convolutional neural network contour control points on a small number of datasets of MRI T2W images of PPAT combined with its gradient images as a feature learning method to reduce feature ambiguity caused by the differences in PPAT contours of different patients. This paper adopts a supervised learning method on the labeled dataset, combining the probability and spatial distribution of control points, and proposes a weighted loss function to optimize the neural network's convergence speed and detection performance. Based on high-precision detection of control points, this paper uses a convex curve fitting to obtain the final PPAT contour. The imaging segmentation results were compared with those of a fully convolutional network (FCN), U-Net, and semantic segmentation convolutional network (SegNet) on three evaluation metrics: Dice similarity coefficient (DSC), Hausdorff distance (HD), and intersection over union ratio (IoU). ResultsCropped images with a 270 × 270-pixel matrix had DSC, HD, and IoU values of 70.1%, 27 mm, and 56.1%, respectively; downscaled images with a 256 × 256-pixel matrix had 68.7%, 26.7 mm, and 54.1%. A U-Net network based on peripheral contour characteristics predicted the complete periprostatic adipose tissue contours on T2W images at different levels. FCN, U-Net, and SegNet could not completely predict them. ConclusionThis U-Net convolutional neural network based on peripheral contour features can identify and segment periprostatic adipose tissue quite well. Cropped images with a 270 × 270-pixel matrix are more appropriate for use with the U-Net convolutional neural network based on contour features; reducing the resolution of the original image will lower the accuracy of the U-Net convolutional neural network. FCN and SegNet are not appropriate for identifying PPAT on T2 sequence MR images. Our method can automatically segment PPAT rapidly and accurately, laying a foundation for PPAT image analysis.

Multi-Center Fetal Brain Tissue Annotation (FeTA) Challenge 2022 Results.

Segmentation is a critical step in analyzing the developing human fetal brain. There have been vast improvements in automatic segmentation methods in the past several years, and the Fetal Brain Tissue Annotation (FeTA) Challenge 2021 helped to establish an excellent standard of fetal brain segmentation. However, FeTA 2021 was a single center study, limiting real-world clinical applicability and acceptance. The multi-center FeTA Challenge 2022 focused on advancing the generalizability of fetal brain segmentation algorithms for magnetic resonance imaging (MRI). In FeTA 2022, the training dataset contained images and corresponding manually annotated multi-class labels from two imaging centers, and the testing data contained images from these two centers as well as two additional unseen centers. The multi-center data included different MR scanners, imaging parameters, and fetal brain super-resolution algorithms applied. 16 teams participated and 17 algorithms were evaluated. Here, the challenge results are presented, focusing on the generalizability of the submissions. Both in- and out-of-domain, the white matter and ventricles were segmented with the highest accuracy (Top Dice scores: 0.89, 0.87 respectively), while the most challenging structure remains the grey matter (Top Dice score: 0.75) due to anatomical complexity. The top 5 average Dices scores ranged from 0.81-0.82, the top 5 average 95th percentile Hausdorff distance values ranged from 2.3-2.5mm, and the top 5 volumetric similarity scores ranged from 0.90-0.92. The FeTA Challenge 2022 was able to successfully evaluate and advance generalizability of multi-class fetal brain tissue segmentation algorithms for MRI and it continues to benchmark new algorithms.

Accurate and robust auto-segmentation of head and neck organ-at-risks based on a novel CNN fine-tuning workflow.

Obvious inconsistencies in auto-segmentations exist among various AI software. In this study, we have developed a novel convolutional neural network (CNN) fine-tuning workflow to achieve precise and robust localized segmentation. The datasets include Hubei Cancer Hospital dataset, Cetuximab Head and Neck Public Dataset, and Québec Public Dataset. Seven organs-at-risks (OARs), including brain stem, left parotid gland, esophagus, left optic nerve, optic chiasm, mandible, and pharyngeal constrictor, were selected. The auto-segmentation results from four commercial AI software were first compared with the manual delineations. Then a new multi-scale lightweight residual CNN model with an attention module (named as HN-Net) was trained and tested on 40 samples and 10 samples from Hubei Cancer Hospital, respectively. To enhance the network's accuracy and generalization ability, the fine-tuning workflow utilized an uncertainty estimation method for automatic selection of candidate samples of worthiness from Cetuximab Head and Neck Public Dataset for further training. The segmentation performances were evaluated on the Hubei Cancer Hospital dataset and/or the entire Québec Public Dataset. A maximum difference of 0.13 and 0.7mm in average Dice value and Hausdorff distance value for the seven OARs were observed by four AI software. The proposed HN-Net achieved an average Dice value of 0.14 higher than that of the AI software, and it also outperformed other popular CNN models (HN-Net: 0.79, U-Net: 0.78, U-Net++: 0.78, U-Net-Multi-scale: 0.77, AI software: 0.65). Additionally, the HN-Net fine-tuning workflow by using the local datasets and external public datasets further improved the automatic segmentation with the average Dice value by 0.02. The delineations of commercial AI software need to be carefully reviewed, and localized further training is necessary for clinical practice. The proposed fine-tuning workflow could be feasibly adopted to implement an accurate and robust auto-segmentation model by using local datasets and external public datasets.

Deep Learning Based on EfficientNet for Multiorgan Segmentation of Thoracic Structures on a 0.35 T MR-Linac Radiation Therapy System

The advent of the 0.35 T MR-Linac (MRIdian, ViewRay) system in radiation therapy allows precise tumor targeting for moving lesions. However, the lack of an automatic volume segmentation function in the MR-Linac’s treatment planning system poses a challenge. In this paper, we propose a deep-learning-based multiorgan segmentation approach for the thoracic region, using EfficientNet as the backbone for the network architecture. The objectives of this approach include accurate segmentation of critical organs, such as the left and right lungs, the heart, the spinal cord, and the esophagus, essential for minimizing radiation toxicity during external radiation therapy. Our proposed approach, when evaluated on an internal dataset comprising 81 patients, demonstrated superior performance compared to other state-of-the-art methods. Specifically, the results for our approach with a 2.5D strategy were as follows: a dice similarity coefficient (DSC) of 0.820 ± 0.041, an intersection over union (IoU) of 0.725 ± 0.052, and a 3D Hausdorff distance (HD) of 10.353 ± 4.974 mm. Notably, the 2.5D strategy surpassed the 2D strategy in all three metrics, exhibiting higher DSC and IoU values, as well as lower HD values. This improvement strongly suggests that our proposed approach with the 2.5D strategy may hold promise in achieving more precise and accurate segmentations when compared to the conventional 2D strategy. Our work has practical implications in the improvement of treatment planning precision, aligning with the evolution of medical imaging and innovative strategies for multiorgan segmentation tasks.

Study of multistep Dense U-Net-based automatic segmentation for head MRI scans.

Despite extensive efforts to obtain accurate segmentation of magnetic resonance imaging (MRI) scans of a head, it remains challenging primarily due to variations in intensity distribution, which depend on the equipment and parameters used. The goal of this study is to evaluate the effectiveness of an automatic segmentation method for head MRI scans using a multistep Dense U-Net (MDU-Net) architecture. The MDU-Net-based method comprises two steps. The first step is to segment the scalp, skull, and whole brain from head MRI scans using a convolutional neural network (CNN). In the first step, a hybrid network is used to combine 2.5D Dense U-Net and 3D Dense U-Net structure. This hybrid network acquires logits in three orthogonal planes (axial, coronal, and sagittal) using 2.5D Dense U-Nets and fuses them by averaging. The resultant fused probability map with head MRI scans then serves as the input to a 3D Dense U-Net. In this process, different ratios of active contour loss and focal loss are applied. The second step is to segment the cerebrospinal fluid (CSF), white matter, and gray matter from extracted brain MRI scans using CNNs. In the second step, the histogram of the extracted brain MRI scans is standardized and then a 2.5D Dense U-Net is used to further segment the brain's specific tissues using the focal loss. A dataset of 100 head MRI scans from an OASIS-3 dataset was used for training, internal validation, and testing, with ratios of 80%, 10%, and 10%, respectively. Using the proposed approach, we segmented the head MRI scans into five areas (scalp, skull, CSF, white matter, and gray matter) and evaluated the segmentation results using the Dice similarity coefficient (DSC) score, Hausdorff distance (HD), and the average symmetric surface distance (ASSD) as evaluation metrics. We compared these results with those obtained using the Res-U-Net, Dense U-Net, U-Net++, Swin-Unet, and H-Dense U-Net models. The MDU-Net model showed DSC values of 0.933, 0.830, 0.833, 0.953, and 0.917 in the scalp, skull, CSF, white matter, and gray matter, respectively. The corresponding HD values were 2.37, 2.89, 2.13, 1.52, and 1.53mm, respectively. The ASSD values were 0.50, 1.63, 1.28, 0.26, and 0.27mm, respectively. Comparing these results with other models revealed that the MDU-Net model demonstrated the best performance in terms of the DSC values for the scalp, CSF, white matter, and gray matter. When compared with the H-Dense U-Net model, which showed the highest performance among the other models, the MDU-Net model showed substantial improvements in the HD view, particularly in the gray matter region, with a difference of approximately 9%. In addition, in terms of the ASSD, the MDU-Net model outperformed the H-Dense U-Net model, showing an approximately 7% improvements in the white matter and approximately 9% improvements in the gray matter. Compared with existing models in terms of DSC, HD, and ASSD, the proposed MDU-Net model demonstrated the best performance on average and showed its potential to enhance the accuracy of automatic segmentation for head MRI scans.

Comparative study of sub-second temporal resolution 4D-MRI and 4D-CT for target motion assessment in a phantom model

To develop and investigate the feasibility of sub-second temporal resolution volumetric T1-weighted four-dimensional (4D-) MRI in comparison with 4D-CT for respiratory-correlated motion assessment using an MRI/CT-compatible phantom. Sub-second high temporal resolution (0.5 s) gradient-echo T1-weighted 4D-MRI was developed using a volumetric acquisition scheme with compressed sensing. An MRI/CT-compatible motion phantom (simulated liver tumor) with three sinusoidal movements of amplitudes and two respiratory patterns was introduced and imaged with 4D-MRI and 4D-CT to investigate the geometric accuracy of the target movement. The geometric accuracy, including centroid position, volume, similarity index of dice similarity coefficient (DSC), and Hausdorff distance (HD), was systematically evaluated. Proposed 4D-MRI achieved a similar geometric accuracy compared with 4D-CT regarding the centroid position, volume, and similarity index. The observed position differences of the absolute average centroid were within 0.08 cm in 4D-MRI and 0.03 cm in 4D-CT, less than the 1-pixel resolution for each modality. The observed volume difference in 4D-MRI/4D-CT was within 0.73 cm3 (4.5%)/0.29 cm3 (2.1%) for a large target and 0.06 cm3 (11.3%)/0.04 cm3 (11.6%) for a small target. The observed DSC values for 4D-MRI/4D-CT were at least 0.93/0.95 for the large target and 0.83/0.84 for the small target. The maximum HD values were 0.25 cm/0.31 cm for the large target and 0.21 cm/0.15 cm for the small target. Although 4D-CT potentially exhibit superior numerical accuracy in phantom studies, the proposed high temporal resolution 4D-MRI demonstrates sub-millimetre geometric accuracy comparable to that of 4D-CT. These findings suggest that the 4D-MRI technique is a viable option for characterizing motion and generating phase-dependent internal target volumes within the realm of radiotherapy.

DAWTran: dynamic adaptive windowing transformer network for pneumothorax segmentation with implicit feature alignment

Objective. This study aims to address the significant challenges posed by pneumothorax segmentation in computed tomography images due to the resemblance between pneumothorax regions and gas-containing structures such as the trachea and bronchus. Approach. We introduce a novel dynamic adaptive windowing transformer (DAWTran) network incorporating implicit feature alignment for precise pneumothorax segmentation. The DAWTran network consists of an encoder module, which employs a DAWTran, and a decoder module. We have proposed a unique dynamic adaptive windowing strategy that enables multi-head self-attention to effectively capture multi-scale information. The decoder module incorporates an implicit feature alignment function to minimize information deviation. Moreover, we utilize a hybrid loss function to address the imbalance between positive and negative samples. Main results. Our experimental results demonstrate that the DAWTran network significantly improves the segmentation performance. Specifically, it achieves a higher dice similarity coefficient (DSC) of 91.35% (a larger DSC value implies better performance), showing an increase of 2.21% compared to the TransUNet method. Meanwhile, it significantly reduces the Hausdorff distance (HD) to 8.06 mm (a smaller HD value implies better performance), reflecting a reduction of 29.92% in comparison to the TransUNet method. Incorporating the dynamic adaptive windowing (DAW) mechanism has proven to enhance DAWTran’s performance, leading to a 4.53% increase in DSC and a 15.85% reduction in HD as compared to SwinUnet. The application of the implicit feature alignment (IFA) further improves the segmentation accuracy, increasing the DSC by an additional 0.11% and reducing the HD by another 10.01% compared to the model only employing DAW. Significance. These results highlight the potential of the DAWTran network for accurate pneumothorax segmentation in clinical applications, suggesting that it could be an invaluable tool in improving the precision and effectiveness of diagnosis and treatment in related healthcare scenarios. The improved segmentation performance with the inclusion of DAW and IFA validates the effectiveness of our proposed model and its components.

Auto-segmentation of thoracic organs in CT scans of breast cancer patients using a 3D U-net cascaded into 2D patchGANs

The quality of organ volume delineation significantly influences the efficacy of radiotherapy treatment for breast cancer patients. This study introduces a novel method for auto-segmentation of the breasts, lungs and heart. The proposed pipeline leverages a multi-class 3D U-Net with a pre-trained ResNet(2+1)D-18 encoder branch, cascaded with a 2D PatchGAN mask correction model for each class. This approach requires a single 3D model, providing a relatively efficient solution. The models were trained and evaluated on 70 thoracic DICOM datasets belonging to breast cancer patients. The evaluation demonstrated state-of-the-art segmentation performance, with mean Dice similarity coefficient values ranging from 0.89 to 0.98, Hausdorff distance values ranging from 2.25 to 8.68 mm, and mean surface distance values ranging from 0.62 to 2.79 mm. These results underscore the pipeline’s potential to enhance breast cancer diagnosis and treatment strategies, with possible applications in other medical sectors utilizing auto-segmentation.

3D Content Encryption Using Multi-Level Chaotic Maps

Nowadays, 3D content is becoming an essential part of multimedia applications, when the 3D content is not protected, hackers may attack and steal it. This paper introduces a proposed scheme that provides high protection for 3D content by implementing multiple levels of security with preserving the original size using weight factor (w). First level of security is implemented by encrypting the texture map based on a 2D Logistic chaotic map. Second level is implemented by shuffling vertices (confusion) based on a 1D Tent chaotic map. Third level is implemented by modifying the vertices values (diffusion) based on a 3D Lorenz chaotic map. Results illustrate that the proposed scheme is completely deform the entire 3D content according to Hausdorff Distance (HD) approximately around 100 after the encryption process. It provides a high security against brute force attack because it has large key space equal to 10165 and secret key sensitivity using NPCR near 99:6% and UACI near 33:4%. The histogram and HD indicate the decrypted 3D content is identical to the origin where HD values approximate zero.