Residual U-Net Architecture Research Articles

A Unified Pipeline for Simultaneous Brain Tumor Classification and Segmentation Using Fine-Tuned CNN and Residual UNet Architecture.

In this paper, I present a comprehensive pipeline integrating a Fine-Tuned Convolutional Neural Network (FT-CNN) and a Residual-UNet (RUNet) architecture for the automated analysis of MRI brain scans. The proposed system addresses the dual challenges of brain tumor classification and segmentation, which are crucial tasks in medical image analysis for precise diagnosis and treatment planning. Initially, the pipeline preprocesses the FigShare brain MRI image dataset, comprising 3064 images, by normalizing and resizing them to achieve uniformity and compatibility with the model. The FT-CNN model then classifies the preprocessed images into distinct tumor types: glioma, meningioma, and pituitary tumor. Following classification, the RUNet model performs pixel-level segmentation to delineate tumor regions within the MRI scans. The FT-CNN leverages the VGG19 architecture, pre-trained on large datasets and fine-tuned for specific tumor classification tasks. Features extracted from MRI images are used to train the FT-CNN, demonstrating robust performance in discriminating between tumor types. Subsequently, the RUNet model, inspired by the U-Net design and enhanced with residual blocks, effectively segments tumors by combining high-resolution spatial information from the encoding path with context-rich features from the bottleneck. My experimental results indicate that the integrated pipeline achieves high accuracy in both classification (96%) and segmentation tasks (98%), showcasing its potential for clinical applications in brain tumor diagnosis. For the classification task, the metrics involved are loss, accuracy, confusion matrix, and classification report, while for the segmentation task, the metrics used are loss, accuracy, Dice coefficient, intersection over union, and Jaccard distance. To further validate the generalizability and robustness of the integrated pipeline, I evaluated the model on two additional datasets. The first dataset consists of 7023 images for classification tasks, expanding to a four-class dataset. The second dataset contains approximately 3929 images for both classification and segmentation tasks, including a binary classification scenario. The model demonstrated robust performance, achieving 95% accuracy on the four-class task and high accuracy (96%) in the binary classification and segmentation tasks, with a Dice coefficient of 95%.

RhizoNet segments plant roots to assess biomass and growth for enabling self-driving labs

Flatbed scanners are commonly used for root analysis, but typical manual segmentation methods are time-consuming and prone to errors, especially in large-scale, multi-plant studies. Furthermore, the complex nature of root structures combined with noisy backgrounds in images complicates automated analysis. Addressing these challenges, this article introduces RhizoNet, a deep learning-based workflow to semantically segment plant root scans. Utilizing a sophisticated Residual U-Net architecture, RhizoNet enhances prediction accuracy and employs a convex hull operation for delineation of the primary root component. Its main objective is to accurately segment root biomass and monitor its growth over time. RhizoNet processes color scans of plants grown in a hydroponic system known as EcoFAB, subjected to specific nutritional treatments. The root detection model using RhizoNet demonstrates strong generalization in the validation tests of all experiments despite variable treatments. The main contributions are the standardization of root segmentation and phenotyping, systematic and accelerated analysis of thousands of images, significantly aiding in the precise assessment of root growth dynamics under varying plant conditions, and offering a path toward self-driving labs.

An enhanced deep learning approach for vascular wall fracture analysis

This work outlines an efficient deep learning approach for analyzing vascular wall fractures using experimental data with openly accessible source codes (https://doi.org/10.25835/weuhha72) for reproduction. Vascular disease remains the primary cause of death globally to this day. Tissue damage in these vascular disorders is closely tied to how the diseases develop, which requires careful study. Therefore, the scientific community has dedicated significant efforts to capture the properties of vessel wall fractures. The symmetry-constrained compact tension (symconCT) test combined with digital image correlation (DIC) enabled the study of tissue fracture in various aorta specimens under different conditions. Main purpose of the experiments was to investigate the displacement and strain field ahead of the crack tip. These experimental data were to support the development and verification of computational models. The FEM model used the DIC information for the material parameters identification. Traditionally, the analysis of fracture processes in biological tissues involves extensive computational and experimental efforts due to the complex nature of tissue behavior under stress. These high costs have posed significant challenges, demanding efficient solutions to accelerate research progress and reduce embedded costs. Deep learning techniques have shown promise in overcoming these challenges by learning to indicate patterns and relationships between the input and label data. In this study, we integrate deep learning methodologies with the attention residual U-Net architecture to predict fracture responses in porcine aorta specimens, enhanced with a Monte Carlo dropout technique. By training the network on a sufficient amount of data, the model learns to capture the features influencing fracture progression. These parameterized datasets consist of pictures describing the evolution of tissue fracture path along with the DIC measurements. The integration of deep learning should not only enhance the predictive accuracy, but also significantly reduce the computational and experimental burden, thereby enabling a more efficient analysis of fracture response.

Vascular changes of the choroid and their correlations with visual acuity in diabetic retinopathy.

To investigate changes in the choroidal vasculature and their correlations with visual acuity in diabetic retinopathy (DR). The cohort was composed of 225 eyes from 225 subjects, including 60 eyes from 60 subjects with healthy control, 55 eyes from 55 subjects without DR, 46 eyes from 46 subjects with nonproliferative diabetic retinopathy (NPDR), 21 eyes from 21 subjects with proliferative diabetic retinopathy (PDR), and 43 eyes from 43 subjects with clinically significant macular edema (CSME). Swept-source optical coherence tomography (SS-OCT) was used to image the eyes with a 12-mm radial line scan protocol. The parameters for 6-mm diameters of region centered on the macular fovea were analyzed. Initially, a custom deep learning algorithm based on a modified residual U-Net architecture was utilized for choroidal boundary segmentation. Subsequently, the SS-OCT image was binarized and the Niblack-based automatic local threshold algorithm was employed to calibrate subfoveal choroidal thickness (SFCT), luminal area (LA), and stromal area (SA) by determining the distance between the two boundaries. Finally, the ratio of LA and total choroidal area (SA + LA) was defined as the choroidal vascularity index (CVI). The choroidal parameters in five groups were compared, and correlations of the choroidal parameters with age, gender, duration of diabetes mellitus (DM), glycated hemoglobin (HbA1c), fasting blood sugar, SFCT and best-corrected visual acuity (BCVA) were analyzed. The CVI, SFCT, LA, and SA values of patients with DR were found to be significantly lower compared to both healthy patients and patients without DR (P< 0.05). The SFCT was significantly higher in NPDR group compared to the No DR group (P < 0.001). Additionally, the SFCT was lower in the PDR group compared to the NPDR group (P = 0.014). Furthermore, there was a gradual decrease in CVI with progression of diabetic retinopathy, reaching its lowest value in the PDR group. However, the CVI of the CSME group exhibited a marginally closer proximity to that of the NPDR group. The multivariate regression analysis revealed a positive correlation between CVI and the duration of DM as well as LA (P < 0.05). The results of both univariate and multivariate regression analyses demonstrated a significant positive correlation between CVI and BCVA (P = 0.003). Choroidal vascular alterations, especially decreased CVI, occurred in patients with DR. The CVI decreased with duration of DM and was correlated with visual impairment, indicating that the CVI might be a reliable imaging biomarker to monitor the progression of DR.

A generalizable deep voxel-guided morphometry algorithm for the detection of subtle lesion dynamics in multiple sclerosis

IntroductionMultiple sclerosis (MS) is a chronic neurological disorder characterized by the progressive loss of myelin and axonal structures in the central nervous system. Accurate detection and monitoring of MS-related changes in brain structures are crucial for disease management and treatment evaluation. We propose a deep learning algorithm for creating Voxel-Guided Morphometry (VGM) maps from longitudinal MRI brain volumes for analyzing MS disease activity. Our approach focuses on developing a generalizable model that can effectively be applied to unseen datasets.MethodsLongitudinal MS patient high-resolution 3D T1-weighted follow-up imaging from three different MRI systems were analyzed. We employed a 3D residual U-Net architecture with attention mechanisms. The U-Net serves as the backbone, enabling spatial feature extraction from MRI volumes. Attention mechanisms are integrated to enhance the model's ability to capture relevant information and highlight salient regions. Furthermore, we incorporate image normalization by histogram matching and resampling techniques to improve the networks' ability to generalize to unseen datasets from different MRI systems across imaging centers. This ensures robust performance across diverse data sources.ResultsNumerous experiments were conducted using a dataset of 71 longitudinal MRI brain volumes of MS patients. Our approach demonstrated a significant improvement of 4.3% in mean absolute error (MAE) against the state-of-the-art (SOTA) method. Furthermore, the algorithm's generalizability was evaluated on two unseen datasets (n = 116) with an average improvement of 4.2% in MAE over the SOTA approach.DiscussionResults confirm that the proposed approach is fast and robust and has the potential for broader clinical applicability.

A Temporal Downscaling Model for Gridded Geophysical Data with Enhanced Residual U-Net

Temporal downscaling of gridded geophysical data is essential for improving climate models, weather forecasting, and environmental assessments. However, existing methods often cannot accurately capture multi-scale temporal features, affecting their accuracy and reliability. To address this issue, we introduce an Enhanced Residual U-Net architecture for temporal downscaling. The architecture, which incorporates residual blocks, allows for deeper network structures without the risk of overfitting or vanishing gradients, thus capturing more complex temporal dependencies. The U-Net design inherently can capture multi-scale features, making it ideal for simulating various temporal dynamics. Moreover, we implement a flow regularization technique with advection loss to ensure that the model adheres to physical laws governing geophysical fields. Our experimental results across various variables within the ERA5 dataset demonstrate an improvement in downscaling accuracy, outperforming other methods.

EAMR-Net: A multiscale effective spatial and cross-channel attention network for retinal vessel segmentation.

Delineation of retinal vessels in fundus images is essential for detecting a range of eye disorders. An automated technique for vessel segmentation can assist clinicians and enhance the efficiency of the diagnostic process. Traditional methods fail to extract multiscale information, discard unnecessary information, and delineate thin vessels. In this paper, a novel residual U-Net architecture that incorporates multi-scale feature learning and effective attention is proposed to delineate the retinal vessels precisely. Since drop block regularization performs better than drop out in preventing overfitting, drop block was used in this study. A multi-scale feature learning module was added instead of a skip connection to learn multi-scale features. A novel effective attention block was proposed and integrated with the decoder block to obtain precise spatial and channel information. Experimental findings indicated that the proposed model exhibited outstanding performance in retinal vessel delineation. The sensitivities achieved for DRIVE, STARE, and CHASE_DB datasets were 0.8293, 0.8151 and 0.8084, respectively.

CAT-Seg: cascaded medical assistive tool integrating residual attention mechanisms and Squeeze-Net for 3D MRI biventricular segmentation

Cardiac image segmentation is a critical step in the early detection of cardiovascular disease. The segmentation of the biventricular is a prerequisite for evaluating cardiac function in cardiac magnetic resonance imaging (CMRI). In this paper, a cascaded model CAT-Seg is proposed for segmentation of 3D-CMRI volumes. CAT-Seg addresses the problem of biventricular confusion with other regions and localized the region of interest (ROI) to reduce the scope of processing. A modified DeepLabv3+ variant integrating SqueezeNet (SqueezeDeepLabv3+) is proposed as a part of CAT-Seg. SqueezeDeepLabv3+ handles the different shapes of the biventricular through the different cardiac phases, as the biventricular only accounts for small portion of the volume slices. Also, CAT-Seg presents a segmentation approach that integrates attention mechanisms into 3D Residual UNet architecture (3D-ResUNet) called 3D-ARU to improve the segmentation results of the three major structures (left ventricle (LV), Myocardium (Myo), and right ventricle (RV)). The integration of the spatial attention mechanism into ResUNet handles the fuzzy edges of the three structures. The proposed model achieves promising results in training and testing with the Automatic Cardiac Diagnosis Challenge (ACDC 2017) dataset and the external validation using MyoPs. CAT-Seg demonstrates competitive performance with state-of-the-art models. On ACDC 2017, CAT-Seg is able to segment LV, Myo, and RV with an average minimum dice symmetry coefficient (DSC) performance gap of 1.165%, 4.36%, and 3.115% respectively. The average maximum improvement in terms of DSC in segmenting LV, Myo and RV is 4.395%, 6.84% and 7.315% respectively. On MyoPs external validation, CAT-Seg outperformed the state-of-the-art in segmenting LV, Myo, and RV with an average minimum performance gap of 6.13%, 5.44%, and 2.912% respectively.

Deep learning in computed tomography super resolution using multi-modality data training.

One of the limitations in leveraging the potential of artificial intelligence in X-ray imaging is the limited availability of annotated training data. As X-ray and CT shares similar imaging physics, one could achieve cross-domain data sharing, so to generate labeled synthetic X-ray images from annotated CT volumes as digitally reconstructed radiographs (DRRs). To account for the lower resolution of CT and the CT-generated DRRs as compared to the real X-ray images, we propose the use of super-resolution (SR) techniques to enhance the CT resolution before DRRgeneration. As spatial resolution can be defined by the modulation transfer function kernel in CT physics, we propose to train a SR network using paired low-resolution (LR) and high-resolution (HR) images by varying the kernel's shape and cutoff frequency. This is different to previous deep learning-based SR techniques on RGB and medical images which focused on refining the sampling grid. Instead of generating LR images by bicubic interpolation, we aim to create realistic multi-detector CT (MDCT) like LR images from HR cone-beam CT (CBCT)scans. We propose and evaluate the use of a SR U-Net for the mapping between LR and HR CBCT image slices. We reconstructed paired LR and HR training volumes from the same CT scans with small in-plane sampling grid size of . We used the residual U-Net architecture to train two models. SRUN : trained with kernel-based LR images, and SRUN : trained with bicubic downsampled data as baseline. Both models are trained on one CBCT dataset (n = 13 391). The performance of both models was then evaluated on unseen kernel-based and interpolation-based LR CBCT images (n = 10 950), and also on MDCT images (n = 1392). Five-fold cross validation and ablation study were performed to find the optimal hyperparameters. Both SRUN and SRUN models show significant improvements (p-value 0.05) in mean absolute error (MAE), peak signal-to-noise ratio (PSNR) and structural similarity index measures (SSIMs) on unseen CBCT images. Also, the improvement percentages in MAE, PSNR, and SSIM by SRUN is larger than SRUN . For SRUN , MAE is reduced by 14%, and PSNR and SSIMs increased by 6 and 8%, respectively. To conclude, SRUN outperforms SRUN , which the former generates sharper images when tested with kernel-based LR CBCT images as well as cross-modality LR MDCTdata. Our proposed method showed better performance than the baseline interpolation approach on unseen LR CBCT. We showed that the frequency behavior of the used data is important for learning the SR features. Additionally, we showed cross-modality resolution improvements to LR MDCT images. Our approach is, therefore, a first and essential step in enabling realistic high spatial resolution CT-generated DRRs for deep learningtraining.

A multiscale residual U-net architecture for super-resolution ultrasonic phased array imaging from full matrix capture data.

Ultrasonic phased array imaging using full-matrix capture (FMC) has raised great interest among various communities, including the nondestructive testing community, as it makes full use of the echo space to provide preferable visualization performance of inhomogeneities. The conventional way of FMC data postprocessing for imaging is through beamforming approaches, such as delay-and-sum, which suffers from limited imaging resolution and contrast-to-noise ratio. To tackle these difficulties, we propose a deep learning (DL)-based image forming approach, termed FMC-Net, to reconstruct high-quality ultrasonic images directly from FMC data. Benefitting from the remarkable capability of DL to approximate nonlinear mapping, the developed FMC-Net automatically models the underlying nonlinear wave-matter interactions; thus, it is trained end-to-end to link the FMC data to the spatial distribution of the acoustic scattering coefficient of the inspected object. Specifically, the FMC-Net is an encoder-decoder architecture composed of multiscale residual modules that make local perception at different scales for the transmitter-receiver pair combinations in the FMC data. We numerically and experimentally compared the DL imaging results to the total focusing method and wavenumber algorithm and demonstrated that the proposed FMC-Net remarkably outperforms conventional methods in terms of exceeding resolution limit and visualizing subwavelength defects. It is expected that the proposed DL approach can benefit a variety of ultrasonic array imaging applications.

Multimodal Brain Tumor Detection Using Loss Aware Residual UNet Model

This research introduces a novel approach to multimodal brain tumor detection using the Residual UNet architecture. By integrating various imaging techniques, such as MRI and CT, the study offers a comprehensive perspective on brain anomalies. The Residual UNet architecture, an enhancement of the conventional UNet, is tailor-made for biomedical image segmentation. The architecture's residual connections optimize deeper network training, making it suitable for detecting intricate patterns in multimodal brain images. This paper details a structured approach that amalgamates advanced imaging techniques and machine learning to develop an efficient tumor detection system. The study utilizes the BRATS brain tumor MRI datasets and incorporates sophisticated preprocessing, data augmentation, and the innovative Loss-Aware ResUNet model for optimal results. It outperformed other top models with a peak accuracy of 97%.

H-Seg: a horizontal reconstruction volume segmentation method for accurate depth estimation in a computer-generated hologram.

In this work, we introduce a novel approach for depth estimation in a computer-generated hologram by employing horizontal segmentation of the reconstruction volume instead of conventional vertical segmentation. The reconstruction volume is divided into horizontal slices and each slice is processed using a residual U-net architecture to identify in-focus lines, enabling determination of the slice's intersection with the 3D scene. The individual slice results are then combined to generate a dense depth map of the scene. Our experiments demonstrate the effectiveness of our method, with improved accuracy, faster processing times, lower graphics processing unit (GPU) utilization, and smoother predicted depth maps than existing state-of-the-art models.

Improving Road Segmentation by Combining Satellite Images and LiDAR Data with a Feature-Wise Fusion Strategy

Numerous deep learning techniques have been explored in pursuit of achieving precise road segmentation; nonetheless, this task continues to present a significant challenge. Exposing shadows and the obstruction of objects are the most important difficulties associated with road segmentation using optical image data alone. By incorporating additional data sources, such as LiDAR data, the accuracy of road segmentation can be improved in areas where optical images are insufficient to segment roads properly. The missing information in spectral data due to the object blockage and shadow effect can be compensated by the integration of 2D and 3D information. This study proposes a feature-wise fusion strategy of optical images and point clouds to enhance the road segmentation performance of a deep learning model. For this purpose, high-resolution satellite images and airborne LiDAR point cloud collected over Florida, USA, were used. Eigenvalue-based and geometric 3D property-based features were calculated based on the LiDAR data. These optical images and LiDAR-based features were used together to train, end-to-end, a deep residual U-Net architecture. In this strategy, the high-level features generated from optical images were concatenated with the LiDAR-based features before the final convolution layer. The consistency of the proposed strategy was evaluated using ResNet backbones with a different number of layers. According to the obtained results, the proposed fusion strategy improved the prediction capacity of the U-Net models with different ResNet backbones. Regardless of the backbone, all models showed enhancement in prediction statistics by 1% to 5%. The combination of optical images and LiDAR point cloud in the deep learning model has increased the prediction performance and provided the integrity of road geometry in woodland and shadowed areas.

Semantic Segmentation of Digestive Abnormalities from WCE Images by Using AttResU-Net Architecture

Colorectal cancer is one of the most common malignancies and the leading cause of cancer death worldwide. Wireless capsule endoscopy is currently the most frequent method for detecting precancerous digestive diseases. Thus, precise and early polyps segmentation has significant clinical value in reducing the probability of cancer development. However, the manual examination is a time-consuming and tedious task for doctors. Therefore, scientists have proposed many computational techniques to automatically segment the anomalies from endoscopic images. In this paper, we present an end-to-end 2D attention residual U-Net architecture (AttResU-Net), which concurrently integrates the attention mechanism and residual units into U-Net for further polyp and bleeding segmentation performance enhancement. To reduce outside areas in an input image while emphasizing salient features, AttResU-Net inserts a sequence of attention units among related downsampling and upsampling steps. On the other hand, the residual block propagates information across layers, allowing for the construction of a deeper neural network capable of solving the vanishing gradient issue in each encoder. This improves the channel interdependencies while lowering the computational cost. Multiple publicly available datasets were employed in this work, to evaluate and verify the proposed method. Our highest-performing model was AttResU-Net, on the MICCAI 2017 WCE dataset, which achieved an accuracy of 99.16%, a Dice coefficient of 94.91%, and a Jaccard index of 90.32%. The experiment findings show that the proposed AttResU-Net overcomes its baselines and provides performance comparable to existing polyp segmentation approaches.

Improved Agricultural Field Segmentation in Satellite Imagery Using TL-ResUNet Architecture

Currently, there is a growing population around the world, and this is particularly true in developing countries, where food security is becoming a major problem. Therefore, agricultural land monitoring, land use classification and analysis, and achieving high yields through efficient land use are important research topics in precision agriculture. Deep learning-based algorithms for the classification of satellite images provide more reliable and accurate results than traditional classification algorithms. In this study, we propose a transfer learning based residual UNet architecture (TL-ResUNet) model, which is a semantic segmentation deep neural network model of land cover classification and segmentation using satellite images. The proposed model combines the strengths of residual network, transfer learning, and UNet architecture. We tested the model on public datasets such as DeepGlobe, and the results showed that our proposed model outperforms the classic models initiated with random weights and pre-trained ImageNet coefficients. The TL-ResUNet model outperforms other models on several metrics commonly used as accuracy and performance measures for semantic segmentation tasks. Particularly, we obtained an IoU score of 0.81 on the validation subset of the DeepGlobe dataset for the TL-ResUNet model.

Use of multifidelity training data and transfer learning for efficient construction of subsurface flow surrogate models

In subsurface flow settings, data assimilation/history matching presents computational challenges because many high-fidelity models must be simulated. Various deep-learning-based surrogate modeling techniques have been developed to reduce the simulation costs associated with these applications. However, to construct data-driven surrogate models, several thousand high-fidelity simulation runs may be required to provide training samples, and these computations can make training prohibitively expensive. To address this issue, in this work we present a framework where most of the training simulations are performed on coarsened (low-fidelity) geomodels. These models are constructed using a flow-based upscaling method. The framework entails the use of a transfer-learning procedure, incorporated within an existing recurrent residual U-Net architecture, in which network training is accomplished in three steps. In the first step, where the bulk of the training is performed, only low-fidelity simulation results are used. The second and third steps, in which the output layer is trained and the overall network is fine-tuned, require a relatively small number of high-fidelity simulations. Here we use 2500 low-fidelity runs and 200 high-fidelity runs, which leads to about a 90% reduction in training simulation costs. The method is applied for two-phase subsurface flow in 3D channelized systems, with flow driven by wells. The surrogate model trained with multifidelity data is shown to be nearly as accurate as a reference surrogate trained with only high-fidelity data in predicting dynamic pressure and saturation fields in new geomodels. Importantly, the network provides results that are significantly more accurate than the low-fidelity simulations used for most of the training. The multifidelity surrogate is also applied for history matching using an ensemble-based procedure, where accuracy relative to reference results is again demonstrated.

Vascular Changes of the Choroid and Their Correlations With Visual Acuity in Pathological Myopia.

Photoreceptor loss plays a role in visual impairment in pathological myopia. As the nutrition and oxygen demands of photoreceptors are mainly supported by the choroidal vessels, we aimed to investigate changes in the choroidal vasculature and their correlations with visual acuity in pathological myopia. The cohort was composed of 136 eyes from 80 participants, including 42 eyes from 21 participants with emmetropia/low myopia, 48 eyes from 26 participants with simple high myopia, and 46 eyes from 33 participants with pathological myopia. Swept-source optical coherence tomography (OCT) was used to image the eyes with a 12-mm radial line scan protocol. The parameters for 6-mm diameters of macula area centered on the fovea were analyzed. A custom deep learning algorithm based on a modified residual U-Net architecture was used to segment the choroidal boundaries. Then, the distance between the two boundaries was determined and choroidal thickness (CT), luminal area (LA), and stromal area (SA) were demarcated based on Niblack's auto-local threshold algorithm after binarization of the OCT images. Finally, the ratio of LA and total choroidal area was defined as the choroidal vascularity index (CVI). The choroidal parameters in three groups were compared, and correlations of the choroidal parameters with age, gender, axial length, and best-corrected visual acuity (BCVA) were analyzed. The CVI, CT, LA, and SA values were lower in pathological myopia than in emmetropia/low myopia and simple high myopia (P < 0.05). The CT, LA, and SA values were lower in simple high myopia than in emmetropia/low myopia (P < 0.05), whereas there was no difference between the CVIs in the emmetropia/low myopia and high myopia groups (P > 0.05). The CVI was nonlinear with increases in axial length (AL), and there was a critical AL flexion point, approximately 27.26 mm; however, the CT, LA, and SA were negatively correlated with AL. Further analysis showed that only younger subjects (40 years old or less) showed significant AL flexion points. Simple and multiple regression models showed that the CVI was correlated with BCVA (P < 0.05). Choroidal vascular alterations, especially decreased CVI, occurred in patients with pathological myopia. The CVI decreased with axial elongation beyond the flexion point and was correlated with visual impairment, indicating that the CVI might be a reliable imaging biomarker to monitor the progression of pathological myopia.

SABOS‐Net: Self‐supervised attention based network for automatic organ segmentation of head and neck CT images

AbstractThe segmentation of Organs At Risk (OAR) in Computed Tomography (CT) images is an essential part of the planning phase of radiation treatment to avoid the adverse effects of cancer radiotherapy treatment. Accurate segmentation is a tedious task in the head and neck region due to a large number of small and sensitive organs and the low contrast of CT images. Deep learning‐based automatic contouring algorithms can ease this task even when the organs have irregular shapes and size variations. This paper proposes a fully automatic deep learning‐based self‐supervised 3D Residual UNet architecture with CBAM(Convolution Block Attention Mechanism) for the organ segmentation in head and neck CT images. The Model Genesis structure and image context restoration techniques are used for self‐supervision, which can help the network learn image features from unlabeled data, hence solving the annotated medical data scarcity problem in deep networks. A new loss function is applied for training by integrating Focal loss, Tversky loss, and Cross‐entropy loss. The proposed model outperforms the state‐of‐the‐art methods in terms of dice similarity coefficient in segmenting the organs. Our self‐supervised model could achieve a 4% increase in the dice score of Chiasm, which is a small organ that is present only in a very few CT slices. The proposed model exhibited better accuracy for 5 out of 7 OARs than the recent state‐of‐the‐art models. The proposed model could simultaneously segment all seven organs in an average time of 0.02 s. The source code of this work is made available at https://github.com/seeniafrancis/SABOSNet.

Auto-detection and segmentation of involved lymph nodes in HPV-associated oropharyngeal cancer using a convolutional deep learning neural network

PurposeSegmentation of involved lymph nodes on head and neck computed tomography (HN-CT) scans is necessary for the radiotherapy planning of early-stage human papilloma virus (HPV) associated oropharynx cancers (OPC). We aimed to train a deep learning convolutional neural network (DL-CNN) to segment involved lymph nodes on HN-CT scans. MethodsGround-truth segmentation of involved nodes was performed on pre-surgical HN-CT scans for 90 patients who underwent levels II-IV neck dissection for node-positive HPV-OPC (training/validation [n = 70] and testing [n = 20]). A 5-fold cross validation approach was used to train 5 DL-CNN sub-models based on a residual U-net architecture. Validation and testing segmentation masks were compared to ground-truth masks using predetermined metrics. A lymph auto-detection model to discriminate between “node-positive” and “node-negative” HN-CT scans was developed by thresholding segmentation model outputs and evaluated using the area under the receiver operating characteristic curve (AUC). ResultsIn the DL-CNN validation phase, all sub-models yielded segmentation masks with median Dice ≥ 0.90 and median volume similarity score of ≥ 0.95. In the testing phase, the DL-CNN produced consensus segmentation masks with median Dice of 0.92 (IQR, 0.89–0.95), median volume similarity of 0.97 (IQR, 0.94–0.99), and median Hausdorff distance of 4.52 mm (IQR, 1.22–8.38). The detection model achieved an AUC of 0.98. ConclusionThe results from this single-institution study demonstrate the successful automation of lymph node segmentation for patients with HPV-OPC using a DL-CNN. Future studies, including validation with an external dataset, are necessary to clarify its role in the larger radiation oncology treatment planning workflow.

Dual U-Net-Based Conditional Generative Adversarial Network for Blood Vessel Segmentation with Reduced Cerebral MR Training Volumes.

Segmenting vessels in brain images is a critical step for many medical interventions and diagnoses of illnesses. Recent advances in artificial intelligence provide better models, achieving a human-like level of expertise in many tasks. In this paper, we present a new approach to segment Time-of-Flight Magnetic Resonance Angiography (TOF-MRA) images, relying on fewer training samples than state-of-the-art methods. We propose a conditional generative adversarial network with an adapted generator based on a concatenated U-Net with a residual U-Net architecture (UUr-cGAN) to carry out blood vessel segmentation in TOF-MRA images, relying on data augmentation to diminish the drawback of having few volumes at disposal for training the model, while preventing overfitting by using regularization techniques. The proposed model achieves 89.52% precision and 87.23% in Dice score on average from the cross-validated experiment for brain blood vessel segmentation tasks, which is similar to other state-of-the-art methods while using considerably fewer training samples. UUr-cGAN extracts important features from small datasets while preventing overfitting compared to other CNN-based methods and still achieve a relatively good performance in image segmentation tasks such as brain blood vessels from TOF-MRA.