SwinD-Net: a lightweight segmentation network for laparoscopic liver segmentation

The real-time requirement for image segmentation in laparoscopic surgical assistance systems is extremely high. Although traditional deep learning models can ensure high segmentation accuracy, they suffer from a large computational burden. In the practical setting of most hospitals, where powerful computing resources are lacking, these models cannot meet the real-time computational demands. We propose a novel network SwinD-Net based on Skip connections, incorporating Depthwise separable convolutions and Swin Transformer Blocks. To reduce computational overhead, we eliminate the skip connection in the first layer and reduce the number of channels in shallow feature maps. Additionally, we introduce Swin Transformer Blocks, which have a larger computational and parameter footprint, to extract global information and capture high-level semantic features. Through these modifications, our network achieves desirable performance while maintaining a lightweight design. We conduct experiments on the CholecSeg8k dataset to validate the effectiveness of our approach. Compared to other models, our approach achieves high accuracy while significantly reducing computational and parameter overhead. Specifically, our model requires only 98.82 M floating-point operations (FLOPs) and 0.52 M parameters, with an inference time of 47.49 ms per image on a CPU. Compared to the recently proposed lightweight segmentation network UNeXt, our model not only outperforms it in terms of the Dice metric but also has only 1/3 of the parameters and 1/22 of the FLOPs. In addition, our model achieves a 2.4 times faster inference speed than UNeXt, demonstrating comprehensive improvements in both accuracy and speed. Our model effectively reduces parameter count and computational complexity, improving the inference speed while maintaining comparable accuracy. The source code will be available at https://github.com/ouyangshuiming/SwinDNet.

Improving Image Segmentation with Contextual and Structural Similarity

Deep learning models for medical image segmentation are usually trained with voxel-wise losses, e.g., cross-entropy loss, focusing on unary supervision without considering inter-voxel relationships. This oversight potentially leads to semantically inconsistent predictions. Here, we propose a contextual similarity loss (CSL) and a structural similarity loss (SSL) to explicitly and efficiently incorporate inter-voxel relationships for improved performance. The CSL promotes consistency in predicted object categories for each image sub-region compared to ground truth. The SSL enforces compatibility between the predictions of voxel pairs by computing pair-wise distances between them, ensuring that voxels of the same class are close together whereas those from different classes are separated by a wide margin in the distribution space. The effectiveness of the CSL and SSL is evaluated using a clinical cone-beam computed tomography (CBCT) dataset of patients with various craniomaxillofacial (CMF) deformities and a public pancreas dataset. Experimental results show that the CSL and SSL outperform state-of-the-art regional loss functions in preserving segmentation semantics.

Detection method of viral pneumonia imaging features based on CT scan images in COVID-19 case study

This study aims to automatically analyze and extract abnormalities in the lung field due to Coronavirus Disease 2019 (COVID-19). Types of abnormalities that can be detected are Ground Glass Opacity (GGO) and consolidation. The proposed method can also identify the location of the abnormality in the lung field, that is, the central and peripheral lung area. The location and type of these abnormalities affect the severity and confidence level of a patient suffering from COVID-19. The detection results using the proposed method are compared with the results of manual detection by radiologists. From the experimental results, the proposed system can provide an average error of 0.059 for the severity score and 0.069 for the confidence level. This method has been implemented in a web-based application for general users.•A method to detect the appearance of viral pneumonia imaging features, namely Ground Glass Opacity (GGO) and consolidation on the chest Computed Tomography (CT) scan images.•This method can separate the lung field to the right lung and the left lung, and it also can identify the detected imaging feature's location in the central or peripheral of the lung field.•Severity level and confidence level of the patient's suffering are measured.

MEF-UNet: An end-to-end ultrasound image segmentation algorithm based on multi-scale feature extraction and fusion

Ultrasound image segmentation is a challenging task due to the complexity of lesion types, fuzzy boundaries, and low-contrast images along with the presence of noises and artifacts. To address these issues, we propose an end-to-end multi-scale feature extraction and fusion network (MEF-UNet) for the automatic segmentation of ultrasound images. Specifically, we first design a selective feature extraction encoder, including detail extraction stage and structure extraction stage, to precisely capture the edge details and overall shape features of the lesions. In order to enhance the representation capacity of contextual information, we develop a context information storage module in the skip-connection section, responsible for integrating information from adjacent two-layer feature maps. In addition, we design a multi-scale feature fusion module in the decoder section to merge feature maps with different scales. Experimental results indicate that our MEF-UNet can significantly improve the segmentation results in both quantitative analysis and visual effects.

Active learning for left ventricle segmentation in echocardiography

BACKGROUND AND OBJECTIVE

Training deep learning models for medical image segmentation require large annotated datasets, which can be expensive and time-consuming to create. Active learning is a promising approach to reduce this burden by strategically selecting the most informative samples for segmentation. This study investigates the use of active learning for efficient left ventricle segmentation in echocardiography with sparse expert annotations.

METHODS

We adapt and evaluate various sampling techniques, demonstrating their effectiveness in judiciously selecting samples for segmentation. Additionally, we introduce a novel strategy, Optimised Representativeness Sampling, which combines feature-based outliers with the most representative samples to enhance annotation efficiency.

RESULTS

Our findings demonstrate a substantial reduction in annotation costs, achieving a remarkable 99% upper bound performance while utilising only 20% of the labelled data. This equates to a reduction of 1680 images needing annotation within our dataset. When applied to a publicly available dataset, our approach yielded a remarkable 70% reduction in required annotation efforts, representing a significant advancement compared to baseline active learning strategies, which achieved only a 50% reduction. Our experiments highlight the nuanced performance of diverse sampling strategies across datasets within the same domain.

CONCLUSIONS

The study provides a cost-effective approach to tackle the challenges of limited expert annotations in echocardiography. By introducing a distinct dataset, made publicly available for research purposes, our work contributes to the field's understanding of efficient annotation strategies in medical image segmentation.

Quantifying the scale of erosion along major coastal aquifers of Pakistan using geospatial and machine learning approaches

Insufficient freshwater recharge and climate change resulted in seawater intrusion in most of the coastal aquifers in Pakistan. Coastal aquifers represent diverse landcover types with varying spectral properties, making it challenging to extract information about their state hence, such investigation requires a combination of geospatial tools. This study aims to monitor erosion along the major coastal aquifers of Pakistan and propose an approach that combines data fusion into the machine and deep learning image segmentation architectures for the erosion and accretion assessment in seascapes. The analysis demonstrated the image segmentation U-Net with EfficientNet backbone achieved the highest F1 score of 0.93, while ResNet101 achieved the lowest F1 score of 0.77. Resultant erosion maps indicated that Sandspit experiencing erosion at 3.14 km2 area. Indus delta is showing erosion, approximately 143 km2 of land over the past 30 years. Sonmiani has undergone substantial erosion with 52.2 km2 land. Miani Hor has experienced erosion up to 298 km2, Bhuri creek has eroded over 4.11 km2, east Phitii creek over 3.30 km2, and Waddi creek over 3.082 km2 land. Tummi creek demonstrates erosion, at 7.12 km2 of land, and East Khalri creek near Keti Bandar has undergone a measured loss of 5.2 km2 land linked with quantified reduction in the vertical sediment flow from 50 (billion cubic meters) to 10 BCM. Our analysis suggests that intense erosions are primarily a result of reduced sediment flow and climate change. Addressing this issue needs to be prioritized coastal management and climate change mitigation framework in Pakistan to safeguard communities. Leveraging emerging solutions, such as loss and damage financing and the integration of nature-based solutions (NbS), should be prioritized for the revival of the coastal aquifers.

GMAlignNet: multi-scale lightweight brain tumor image segmentation with enhanced semantic information consistency

Although the U-shaped architecture, represented by UNet, has become a major network model for brain tumor segmentation, the repeated convolution and sampling operations can easily lead to the loss of crucial information. Additionally, directly fusing features from different levels without distinction can easily result in feature misalignment, affecting segmentation accuracy. On the other hand, traditional convolutional blocks used for feature extraction cannot capture the abundant multi-scale information present in brain tumor images. This paper proposes a multi-scale feature-aligned segmentation model called GMAlignNet that fully utilizes Ghost convolution to solve these problems. Ghost Hierarchical Decoupled Fusion Unit and Ghost Hierarchical Decoupled Unit are used instead of standard convolutions in the encoding and decoding paths. This transformation replaces the holistic learning of volume structures by traditional convolutional blocks with multi-level learning on a specific view, facilitating the acquisition of abundant multi-scale contextual information through low-cost operations. Furthermore, a feature alignment unit is proposed that can utilize semantic information flow to guide the recovery of upsampled features. It performs pixel-level semantic information correction on misaligned features due to feature fusion. The proposed method is also employed to optimize three classic networks, namely DMFNet, HDCNet, and 3D UNet, demonstrating its effectiveness in automatic brain tumor segmentation. The proposed network model was applied to the BraTS 2018 dataset, and the results indicate that the proposed GMAlignNet achieved Dice coefficients of 81.65%, 90.07%, and 85.16% for enhancing tumor, whole tumor, and tumor core segmentation, respectively. Moreover, with only 0.29M parameters and 26.88G FLOPs, it demonstrates better potential in terms of computational efficiency and possesses the advantages of lightweight. Extensive experiments on the BraTS 2018, BraTS 2019, and BraTS 2020 datasets suggest that the proposed model exhibits better potential in handling edge details and contour recognition.

An Intuitionistic Fuzzy C-Means and Local Information-Based DCT Filtering for Fast Brain MRI Segmentation

Structural and photometric anomalies in the brain magnetic resonance images (MRIs) affect the segmentation performance. Moreover, a sudden change in intensity between two boundaries of the brain tissues makes it prone to data uncertainty, resulting in the misclassification of the pixels lying near the cluster boundaries. The discrete cosine transform (DCT) domain-based filtering is an effective way to deal with structural and photometric anomalies, while the intuitionistic fuzzy C-means (IFCM) clustering can handle the uncertainty using the intuitionistic fuzzy set (IFS) theory. In this background, we propose two novel approaches, namely, the DCT-based intuitionistic fuzzy C-means (DCT-IFCM) and the DCT-based local information IFCM (DCT-LIFCM), which effectively deal with the Rician and Gaussian noises and also handle the data uncertainty problem to provide high segmentation accuracy. The DCT-IFCM approach performs the histogram-based segmentation, while the DCT-LIFCM uses the pixel-wise computation to include the spatial information. Although the DCT-LIFCM delivers slightly better performance than the DCT-IFCM, the latter is very fast in providing equally high segmentation accuracy. An exhaustive performance analysis is provided to demonstrate the superior performance of the proposed algorithms compared with the state-of-the-art algorithms, including those based on the DCT-based filtering approach and the IFS theory.

Deep learning automatically assesses 2-µm laser-induced skin damage OCT images

The present study proposed a noninvasive, automated, in vivo assessment method based on optical coherence tomography (OCT) and deep learning techniques to qualitatively and quantitatively analyze the biological effects of 2-µm laser-induced skin damage at different irradiation doses. Different doses of 2-µm laser irradiation established a mouse skin damage model, after which the skin-damaged tissues were imaged non-invasively in vivo using OCT. The acquired images were preprocessed to construct the dataset required for deep learning. The deep learning models used were U-Net, DeepLabV3+, PSP-Net, and HR-Net, and the trained models were used to segment the damage images and further quantify the damage volume of mouse skin under different irradiation doses. The comparison of the qualitative and quantitative results of the four network models showed that HR-Net had the best performance, the highest agreement between the segmentation results and real values, and the smallest error in the quantitative assessment of the damage volume. Based on HR-Net to segment the damage image and quantify the damage volume, the irradiation doses 5.41, 9.55, 13.05, 20.85, 32.71, 52.92, 76.71, and 97.24 J/cm² corresponded to a damage volume of 4.58, 12.56, 16.74, 20.88, 24.52, 30.75, 34.13, and 37.32 mm³. The damage volume increased in a radiation dose-dependent manner.

CardSegNet: An adaptive hybrid CNN-vision transformer model for heart region segmentation in cardiac MRI

Cardiovascular MRI (CMRI) is a non-invasive imaging technique adopted for assessing the blood circulatory system's structure and function. Precise image segmentation is required to measure cardiac parameters and diagnose abnormalities through CMRI data. Because of anatomical heterogeneity and image variations, cardiac image segmentation is a challenging task. Quantification of cardiac parameters requires high-performance segmentation of the left ventricle (LV), right ventricle (RV), and left ventricle myocardium from the background. The first proposed solution here is to manually segment the regions, which is a time-consuming and error-prone procedure. In this context, many semi- or fully automatic solutions have been proposed recently, among which deep learning-based methods have revealed high performance in segmenting regions in CMRI data. In this study, a self-adaptive multi attention (SMA) module is introduced to adaptively leverage multiple attention mechanisms for better segmentation. The convolutional-based position and channel attention mechanisms with a patch tokenization-based vision transformer (ViT)-based attention mechanism in a hybrid and end-to-end manner are integrated into the SMA. The CNN- and ViT-based attentions mine the short- and long-range dependencies for more precise segmentation. The SMA module is applied in an encoder-decoder structure with a ResNet50 backbone named CardSegNet. Furthermore, a deep supervision method with multi-loss functions is introduced to the CardSegNet optimizer to reduce overfitting and enhance the model's performance. The proposed model is validated on the ACDC2017 (n=100), M&Ms (n=321), and a local dataset (n=22) using the 10-fold cross-validation method with promising segmentation results, demonstrating its outperformance versus its counterparts.

Usformer: A small network for left atrium segmentation of 3D LGE MRI

Left atrial (LA) fibrosis plays a vital role as a mediator in the progression of atrial fibrillation. 3D late gadolinium-enhancement (LGE) MRI has been proven effective in identifying LA fibrosis. Image analysis of 3D LA LGE involves manual segmentation of the LA wall, which is both lengthy and challenging. Automated segmentation poses challenges owing to the diverse intensities in data from various vendors, the limited contrast between LA and surrounding tissues, and the intricate anatomical structures of the LA. Current approaches relying on 3D networks are computationally intensive since 3D LGE MRIs and the networks are large. Regarding this issue, most researchers came up with two-stage methods: initially identifying the LA center using a scaled-down version of the MRIs and subsequently cropping the full-resolution MRIs around the LA center for final segmentation. We propose a lightweight transformer-based 3D architecture, Usformer, designed to precisely segment LA volume in a single stage, eliminating error propagation associated with suboptimal two-stage training. The transposed attention facilitates capturing the global context in large 3D volumes without significant computation requirements. Usformer outperforms the state-of-the-art supervised learning methods in terms of accuracy and speed. First, with the smallest Hausdorff Distance (HD) and Average Symmetric Surface Distance (ASSD), it achieved a dice score of 93.1% and 92.0% in the 2018 Atrial Segmentation Challenge and our local institutional dataset, respectively. Second, the number of parameters and computation complexity are largely reduced by 2.8x and 3.8x, respectively. Moreover, Usformer does not require a large dataset. When only 16 labeled MRI scans are used for training, Usformer achieves a 92.1% dice score in the challenge dataset. The proposed Usformer delineates the boundaries of the LA wall relatively accurately, which may assist in the clinical translation of LA LGE for planning catheter ablation of atrial fibrillation.

Numerical subgrid Bi-cubic methods of partial differential equations in image segmentation

Image segmentation is a core research in the image processing and computer vision. In this paper, we suggest a Bi-cubic spline phase transition potential and elaborate a Bi-Cubic spline phase transition potential development. In the image segmentation, we develop the new approach to apply the novel computational fluid dynamics in the boundary with subgrid. The numerical subgrid Bi-cubic method with Bi-Cubic spline for minimizing the piecewise constant energy functional is very efficient, robust and fast in the image segmentation with a multispecies multiphase segmentation models. The subgrid Bi-cubic spline is applied on the boundary with subgrid and the regular grid is applied on the non-boundary. The model generates a multispecies multiphase distribution with Bi-Cubic spline and we can extract the image segments with multispecies multiphase. Finally, we analyze the models and show the numerical results. Numerical results are presented with OCR (Optical Character Recognition) and the medical image.

ComPRePS: An Automated Cloud-based Image Analysis tool to democratize AI in Digital Pathology

Artificial intelligence (AI) has extensive applications in a wide range of disciplines including healthcare and clinical practice. Advances in high-resolution whole-slide brightfield microscopy allow for the digitization of histologically stained tissue sections, producing gigapixel-scale whole-slide images (WSI). The significant improvement in computing and revolution of deep neural network (DNN)-based AI technologies over the last decade allow us to integrate massively parallelized computational power, cutting-edge AI algorithms, and big data storage, management, and processing. Applied to WSIs, AI has created opportunities for improved disease diagnostics and prognostics with the ultimate goal of enhancing precision medicine and resulting patient care. The National Institutes of Health (NIH) has recognized the importance of developing standardized principles for data management and discovery for the advancement of science and proposed the Findable, Accessible, Interoperable, Reusable, (FAIR) Data Principles1 with the goal of building a modernized biomedical data resource ecosystem to establish collaborative research communities. In line with this mission and to democratize AI-based image analysis in digital pathology, we propose ComPRePS: an end-to-end automated Computational Renal Pathology Suite which combines massive scalability, on-demand cloud computing, and an easy-to-use web-based user interface for data upload, storage, management, slide-level visualization, and domain expert interaction. Moreover, our platform is equipped with both in-house and collaborator developed sophisticated AI algorithms in the back-end server for image analysis to identify clinically relevant micro-anatomic functional tissue units (FTU) and to extract image features.

A deep learning-based approach for fully automated segmentation and quantitative analysis of muscle fibers in pig skeletal muscle

Muscle fiber properties exert a significant influence on pork quality, with cross-sectional area (CSA) being a crucial parameter closely associated with various meat quality indicators, such as shear force. Effectively identifying and segmenting muscle fibers in a robust manner constitutes a vital initial step in determining CSA. This step is highly intricate and time-consuming, necessitating an accurate and automated analytical approach. One limitation of existing methods is their tendency to perform well on high signal-to-noise ratio images of intact, healthy muscle fibers but their lack of validation on more complex image datasets featuring significant morphological changes, such as the presence of ice crystals. In this study, we undertake the fully automatic segmentation of muscle fiber microscopic images stained with myosin adenosine triphosphate (mATPase) activity using a deep learning architecture known as SOLOv2. Our objective is to efficiently derive accurate measurements of muscle fiber size and distribution. Tests conducted on actual images demonstrate that our method adeptly handles the intricate task of muscle fiber segmentation, yielding quantitative results amenable to statistical analysis and displaying reliability comparable to manual analysis.

US2Mask: Image-to-mask generation learning via a conditional GAN for cardiac ultrasound image segmentation

Cardiac ultrasound (US) image segmentation is vital for evaluating clinical indices, but it often demands a large dataset and expert annotations, resulting in high costs for deep learning algorithms. To address this, our study presents a framework utilizing artificial intelligence generation technology to produce multi-class RGB masks for cardiac US image segmentation. The proposed approach directly performs semantic segmentation of the heart's main structures in US images from various scanning modes. Additionally, we introduce a novel learning approach based on conditional generative adversarial networks (CGAN) for cardiac US image segmentation, incorporating a conditional input and paired RGB masks. Experimental results from three cardiac US image datasets with diverse scan modes demonstrate that our approach outperforms several state-of-the-art models, showcasing improvements in five commonly used segmentation metrics, with lower noise sensitivity. Source code is available at https://github.com/energy588/US2mask.

Fully automatic integration of dental CBCT images and full-arch intraoral impressions with stitching error correction via individual tooth segmentation and identification

We present a fully automated method of integrating intraoral scan (IOS) and dental cone-beam computerized tomography (CBCT) images into one image by complementing each image's weaknesses. Dental CBCT alone may not be able to delineate precise details of the tooth surface due to limited image resolution and various CBCT artifacts, including metal-induced artifacts. IOS is very accurate for the scanning of narrow areas, but it produces cumulative stitching errors during full-arch scanning. The proposed method is intended not only to compensate the low-quality of CBCT-derived tooth surfaces with IOS, but also to correct the cumulative stitching errors of IOS across the entire dental arch. Moreover, the integration provides both gingival structure of IOS and tooth roots of CBCT in one image. The proposed fully automated method consists of four parts; (i) individual tooth segmentation and identification module for IOS data (TSIM-IOS); (ii) individual tooth segmentation and identification module for CBCT data (TSIM-CBCT); (iii) global-to-local tooth registration between IOS and CBCT; and (iv) stitching error correction for full-arch IOS. The experimental results show that the proposed method achieved landmark and surface distance errors of 112.4μm and 301.7μm, respectively.

Pelvic bone tumor segmentation fusion algorithm based on fully convolutional neural network and conditional random field

Background and objective

Pelvic bone tumors represent a harmful orthopedic condition, encompassing both benign and malignant forms. Addressing the issue of limited accuracy in current machine learning algorithms for bone tumor image segmentation, we have developed an enhanced bone tumor image segmentation algorithm. This algorithm is built upon an improved full convolutional neural network, incorporating both the fully convolutional neural network (FCNN-4s) and a conditional random field (CRF) to achieve more precise segmentation.

Methodology

The enhanced fully convolutional neural network (FCNN-4s) was employed to conduct initial segmentation on preprocessed images. Following each convolutional layer, batch normalization layers were introduced to expedite network training convergence and enhance the accuracy of the trained model. Subsequently, a fully connected conditional random field (CRF) was integrated to fine-tune the segmentation results, refining the boundaries of pelvic bone tumors and achieving high-quality segmentation.

Results

The experimental outcomes demonstrate a significant enhancement in segmentation accuracy and stability when compared to the conventional convolutional neural network bone tumor image segmentation algorithm. The algorithm achieves an average Dice coefficient of 93.31 %, indicating superior performance in real-time operations.

Conclusion

In contrast to the conventional convolutional neural network segmentation algorithm, the algorithm presented in this paper boasts a more intricate structure, proficiently addressing issues of over-segmentation and under-segmentation in pelvic bone tumor segmentation. This segmentation model exhibits superior real-time performance, robust stability, and is capable of achieving heightened segmentation accuracy.

Visual Prompting Based Incremental Learning for Semantic Segmentation of Multiplex Immuno-Flourescence Microscopy Imagery

Deep learning approaches are state-of-the-art for semantic segmentation of medical images, but unlike many deep learning applications, medical segmentation is characterized by small amounts of annotated training data. Thus, while mainstream deep learning approaches focus on performance in domains with large training sets, researchers in the medical imaging field must apply new methods in creative ways to meet the more constrained requirements of medical datasets. We propose a framework for incrementally fine-tuning a multi-class segmentation of a high-resolution multiplex (multi-channel) immuno-flourescence image of a rat brain section, using a minimal amount of labelling from a human expert. Our framework begins with a modified Swin-UNet architecture that treats each biomarker in the multiplex image separately and learns an initial "global" segmentation (pre-training). This is followed by incremental learning and refinement of each class using a very limited amount of additional labeled data provided by a human expert for each region and its surroundings. This incremental learning utilizes the multi-class weights as an initialization and uses the additional labels to steer the network and optimize it for each region in the image. In this way, an expert can identify errors in the multi-class segmentation and rapidly correct them by supplying the model with additional annotations hand-picked from the region. In addition to increasing the speed of annotation and reducing the amount of labelling, we show that our proposed method outperforms a traditional multi-class segmentation by a large margin.

Optic disc and cup segmentation for glaucoma detection using Attention U-Net incorporating residual mechanism

Glaucoma is a common eye disease that can cause blindness. Accurate detection of the optic disc and cup disc is crucial for glaucoma diagnosis. Algorithm models based on artificial intelligence can assist doctors in improving detection performance. In this article, U-Net is used as the backbone network, and the attention and residual modules are integrated to construct an end-to-end convolutional neural network model for optic disc and cup disc segmentation. The U-Net backbone is used to infer the basic position information of optic disc and cup disc, the attention module enhances the model's ability to represent and extract features of optic disc and cup disc, and the residual module alleviates gradient disappearance or explosion that may occur during feature representation of the neural network. The proposed model is trained and tested on the DRISHTI-GS1 dataset. Results show that compared with the original U-Net method, our model can more effectively separate optic disc and cup disc in terms of overlap error, sensitivity, and specificity.

Deep learning-based automatic pipeline for 3D needle localization on intra-procedural 3D MRI

PURPOSE

Accurate and rapid needle localization on 3D magnetic resonance imaging (MRI) is critical for MRI-guided percutaneous interventions. The current workflow requires manual needle localization on 3D MRI, which is time-consuming and cumbersome. Automatic methods using 2D deep learning networks for needle segmentation require manual image plane localization, while 3D networks are challenged by the need for sufficient training datasets. This work aimed to develop an automatic deep learning-based pipeline for accurate and rapid 3D needle localization on in vivo intra-procedural 3D MRI using a limited training dataset.

METHODS

The proposed automatic pipeline adopted Shifted Window (Swin) Transformers and employed a coarse-to-fine segmentation strategy: (1) initial 3D needle feature segmentation with 3D Swin UNEt TRansfomer (UNETR); (2) generation of a 2D reformatted image containing the needle feature; (3) fine 2D needle feature segmentation with 2D Swin Transformer and calculation of 3D needle tip position and axis orientation. Pre-training and data augmentation were performed to improve network training. The pipeline was evaluated via cross-validation with 49 in vivo intra-procedural 3D MR images from preclinical pig experiments. The needle tip and axis localization errors were compared with human intra-reader variation using the Wilcoxon signed rank test, with p < 0.05 considered significant.

RESULTS

The average end-to-end computational time for the pipeline was 6 s per 3D volume. The median Dice scores of the 3D Swin UNETR and 2D Swin Transformer in the pipeline were 0.80 and 0.93, respectively. The median 3D needle tip and axis localization errors were 1.48 mm (1.09 pixels) and 0.98°, respectively. Needle tip localization errors were significantly smaller than human intra-reader variation (median 1.70 mm; p < 0.01).

CONCLUSION

The proposed automatic pipeline achieved rapid pixel-level 3D needle localization on intra-procedural 3D MRI without requiring a large 3D training dataset and has the potential to assist MRI-guided percutaneous interventions.

PARSEG: a computationally efficient approach for statistical validation of botanical seeds' images

Human recognition and automated image validation are the most widely used approaches to validate the output of binary segmentation methods but, as the number of pixels in an image easily exceeds several million, they become highly demanding from both practical and computational standpoint. We propose a method, called PARSEG, which stands for PArtitioning, Random Selection, Estimation, and Generalization; being the basic steps within this procedure. Suggested method enables us to perform statistical validation of binary images by selecting the minimum number of pixels from the original image to be used for validation without deteriorating the effectiveness of the validation procedure. It utilizes binary classifiers to accomplish image validation and selects the optimal sample of pixels according to a specific objective function. As a result, the computational complexity of the validation experiment is substantially reduced. The procedure's effectiveness is illustrated by considering images composed of approximately 13 million pixels from the field of seed recognition. PARSEG provides roughly the same precision of the validation process when extended to the entire image, but it utilizes only about 4% of the original number of pixels, thus reducing, by about 90%, the computing time required to validate a binary segmented image.

ResDAC-Net: a novel pancreas segmentation model utilizing residual double asymmetric spatial kernels

The pancreas not only is situated in a complex abdominal background but is also surrounded by other abdominal organs and adipose tissue, resulting in blurred organ boundaries. Accurate segmentation of pancreatic tissue is crucial for computer-aided diagnosis systems, as it can be used for surgical planning, navigation, and assessment of organs. In the light of this, the current paper proposes a novel Residual Double Asymmetric Convolution Network (ResDAC-Net) model. Firstly, newly designed ResDAC blocks are used to highlight pancreatic features. Secondly, the feature fusion between adjacent encoding layers fully utilizes the low-level and deep-level features extracted by the ResDAC blocks. Finally, parallel dilated convolutions are employed to increase the receptive field to capture multiscale spatial information. ResDAC-Net is highly compatible to the existing state-of-the-art models, according to three (out of four) evaluation metrics, including the two main ones used for segmentation performance evaluation (i.e., DSC and Jaccard index).

Light-M: An efficient lightweight medical image segmentation framework for resource-constrained IoMT

The Internet of Medical Things (IoMT) is being incorporated into current healthcare systems. This technology intends to connect patients, IoMT devices, and hospitals over mobile networks, allowing for more secure, quick, and convenient health monitoring and intelligent healthcare services. However, existing intelligent healthcare applications typically rely on large-scale AI models, and standard IoMT devices have significant resource constraints. To alleviate this paradox, in this paper, we propose a Knowledge Distillation (KD)-based IoMT end-edge-cloud orchestrated architecture for medical image segmentation tasks, called Light-M, aiming to deploy a lightweight medical model in resource-constrained IoMT devices. Specifically, Light-M trains a large teacher model in the cloud server and employs computation in local nodes through imitation of the performance of the teacher model using knowledge distillation. Light-M contains two KD strategies: (1) active exploration and passive transfer (AEPT) and (2) self-attention-based inter-class feature variation (AIFV) distillation for the medical image segmentation task. The AEPT encourages the student model to learn undiscovered knowledge/features of the teacher model without additional feature layers, aiming to explore new features and outperform the teacher. To improve the distinguishability of the student for different classes, the student learns the self-attention-based feature variation (AIFV) between classes. Since the proposed AEPT and AIFV only appear in the training process, our framework does not involve any additional computation burden for a student model during the segmentation task deployment. Extensive experiments on cardiac images and public real-scene datasets demonstrate that our approach improves student model learning representations and outperforms state-of-the-art methods by combining two knowledge distillation strategies. Moreover, when deployed on the IoT device, the distilled student model takes only 29.6 ms for one sample at the inference step.

MicroSegNet: A deep learning approach for prostate segmentation on micro-ultrasound images

Micro-ultrasound (micro-US) is a novel 29-MHz ultrasound technique that provides 3-4 times higher resolution than traditional ultrasound, potentially enabling low-cost, accurate diagnosis of prostate cancer. Accurate prostate segmentation is crucial for prostate volume measurement, cancer diagnosis, prostate biopsy, and treatment planning. However, prostate segmentation on micro-US is challenging due to artifacts and indistinct borders between the prostate, bladder, and urethra in the midline. This paper presents MicroSegNet, a multi-scale annotation-guided transformer UNet model designed specifically to tackle these challenges. During the training process, MicroSegNet focuses more on regions that are hard to segment (hard regions), characterized by discrepancies between expert and non-expert annotations. We achieve this by proposing an annotation-guided binary cross entropy (AG-BCE) loss that assigns a larger weight to prediction errors in hard regions and a lower weight to prediction errors in easy regions. The AG-BCE loss was seamlessly integrated into the training process through the utilization of multi-scale deep supervision, enabling MicroSegNet to capture global contextual dependencies and local information at various scales. We trained our model using micro-US images from 55 patients, followed by evaluation on 20 patients. Our MicroSegNet model achieved a Dice coefficient of 0.939 and a Hausdorff distance of 2.02 mm, outperforming several state-of-the-art segmentation methods, as well as three human annotators with different experience levels. Our code is publicly available at https://github.com/mirthAI/MicroSegNet and our dataset is publicly available at https://zenodo.org/records/10475293.

CrossU-Net: Dual-modality cross-attention U-Net for segmentation of precancerous lesions in gastric cancer

Gastric precancerous lesions (GPL) significantly elevate the risk of gastric cancer, and precise diagnosis and timely intervention are critical for patient survival. Due to the elusive pathological features of precancerous lesions, the early detection rate is less than 10%, which hinders lesion localization and diagnosis. In this paper, we provide a GPL pathological dataset and propose a novel method for improving the segmentation accuracy on a limited-scale dataset, namely RGB and Hyperspectral dual-modal pathological image Cross-attention U-Net (CrossU-Net). Specifically, we present a self-supervised pre-training model for hyperspectral images to serve downstream segmentation tasks. Secondly, we design a dual-stream U-Net-based network to extract features from different modal images. To promote information exchange between spatial information in RGB images and spectral information in hyperspectral images, we customize the cross-attention mechanism between the two networks. Furthermore, we use an intermediate agent in this mechanism to improve computational efficiency. Finally, we add a distillation loss to align predicted results for both branches, improving network generalization. Experimental results show that our CrossU-Net achieves accuracy and Dice of 96.53% and 91.62%, respectively, for GPL lesion segmentation, providing a promising spectral research approach for the localization and subsequent quantitative analysis of pathological features in early diagnosis.

Choroid plexus volume in multiple sclerosis can be estimated on structural MRI avoiding contrast injection

We compared choroid plexus (ChP) manual segmentation on non-contrast-enhanced (non-CE) sequences and reference standard CE T1- weighted (T1w) sequences in 61 multiple sclerosis patients prospectively included. ChP was separately segmented on T1w, T2-weighted (T2w) fluid-attenuated inversion-recovery (FLAIR), and CE-T1w sequences. Inter-rater variability assessed on 10 subjects showed high reproducibility between sequences measured by intraclass correlation coefficient (T1w 0.93, FLAIR 0.93, CE-T1w 0.99). CE-T1w showed higher signal-to-noise ratio and contrast-to-noise ratio (CE-T1w 23.77 and 18.49, T1w 13.73 and 7.44, FLAIR 13.09 and 10.77, respectively). Manual segmentation of ChP resulted 3.073 ± 0.563 mL (mean ± standard deviation) on T1w, 3.787 ± 0.679 mL on FLAIR, and 2.984 ± 0.506 mL on CE-T1w images, with an error of 28.02 ± 19.02% for FLAIR and 3.52 ± 12.61% for T1w. FLAIR overestimated ChP volume compared to CE-T1w (p < 0.001). The Dice similarity coefficient of CE-T1w versus T1w and FLAIR was 0.67 ± 0.05 and 0.68 ± 0.05, respectively. Spatial error distribution per slice was calculated after nonlinear coregistration to the standard MNI152 space and showed a heterogeneous profile along the ChP especially near the fornix and the hippocampus. Quantitative analyses suggest T1w as a surrogate of CE-T1w to estimate ChP volume.Relevance statement To estimate the ChP volume, CE-T1w can be replaced by non-CE T1w sequences because the error is acceptable, while FLAIR overestimates the ChP volume. This encourages the development of automatic tools for ChP segmentation, also improving the understanding of the role of the ChP volume in multiple sclerosis, promoting longitudinal studies.Key points • CE-T1w sequences are considered the reference standard for ChP manual segmentation.• FLAIR sequences showed a higher CNR than T1w sequences but overestimated the ChP volume.• Non-CE T1w sequences can be a surrogate of CE-T1w sequences for manual segmentation of ChP.

LAMA: Lesion-Aware Mixup Augmentation for Skin Lesion Segmentation

Deep learning can exceed dermatologists' diagnostic accuracy in experimental image environments. However, inaccurate segmentation of images with multiple skin lesions can be seen with current methods. Thus, information present in multiple-lesion images, available to specialists, is not retrievable by machine learning. While skin lesion images generally capture a single lesion, there may be cases in which a patient's skin variation may be identified as skin lesions, leading to multiple false positive segmentations in a single image. Conversely, image segmentation methods may find only one region and may not capture multiple lesions in an image. To remedy these problems, we propose a novel and effective data augmentation technique for skin lesion segmentation in dermoscopic images with multiple lesions. The lesion-aware mixup augmentation (LAMA) method generates a synthetic multi-lesion image by mixing two or more lesion images from the training set. We used the publicly available International Skin Imaging Collaboration (ISIC) 2017 Challenge skin lesion segmentation dataset to train the deep neural network with the proposed LAMA method. As none of the previous skin lesion datasets (including ISIC 2017) has considered multiple lesions per image, we created a new multi-lesion (MuLe) segmentation dataset utilizing publicly available ISIC 2020 skin lesion images with multiple lesions per image. MuLe was used as a test set to evaluate the effectiveness of the proposed method. Our test results show that the proposed method improved the Jaccard score 8.3% from 0.687 to 0.744 and the Dice score 5% from 0.7923 to 0.8321 over a baseline model on MuLe test images. On the single-lesion ISIC 2017 test images, LAMA improved the baseline model's segmentation performance by 0.08%, raising the Jaccard score from 0.7947 to 0.8013 and the Dice score 0.6% from 0.8714 to 0.8766. The experimental results showed that LAMA improved the segmentation accuracy on both single-lesion and multi-lesion dermoscopic images. The proposed LAMA technique warrants further study.

Highly accurate and effective deep neural networks in pathological diagnosis of prostate cancer

PURPOSE

To established an AI system to make the pathological diagnosis of prostate cancer.

METHODS

Prostate histopathological whole mount (WM) sections from patients underwent robot-assisted laparoscopic prostatectomy were prepared. All the prostate WM pathological sections were converted to digital image data and marked with different colors on the basis of the ISUP Gleason grade group. The image was then fed into a segmentation algorithm. We chose modified U-Net as our fundamental network architecture.

RESULTS

172 patients were involved in this study. 896 pieces of prostate WM pathological sections from 160 patients, in which 826 pieces of WM sections from 148 patients were assigned to the training set randomly. After image segmentation there were totally 2,138,895 patches, of which 1,646,535 patches were valid for training. The other WM section was arranged for testing. Based on the whole image testing, AI and pathologists presented the same answers among 21 of 22 pieces of sections. To evaluate the diagnostic results at the pixel level, we anticipated correct cancer or non-cancer diagnose from this AI system. The area under the ROC curve as 96.8%. The value of pixel accuracy of three methods (binary analysis, clinically oriented analysis and analysis for different ISUP Gleason grade) were 96.93%, 95.43% and 93.88%, respectively. The value of frequency weighted IoU were 94.32%, 92.13% and 90.21%, respectively.

CONCLUSIONS

This AI system is able to assist pathologists to make a final diagnosis, indicating the great potential and a wide-range of applications of AI in the medical field.

Segmentation of liver and liver lesions using deep learning

Segmentation of organs and lesions could be employed for the express purpose of dosimetry in nuclear medicine, assisted image interpretations, and mass image processing studies. Deep leaning created liver and liver lesion segmentation on clinical 3D MRI data has not been fully addressed in previous experiments. To this end, the required data were collected from 128 patients, including their T1w and T2w MRI images, and ground truth labels of the liver and liver lesions were generated. The collection of 110 T1w-T2w MRI image sets was divided, with 94 designated for training and 16 for validation. Furthermore, 18 more datasets were separately allocated for use as hold-out test datasets. The T1w and T2w MRI images were preprocessed into a two-channel format so that they were used as inputs to the deep learning model based on the Isensee 2017 network. To calculate the final Dice coefficient of the network performance on test datasets, the binary average of T1w and T2w predicted images was used. The deep learning model could segment all 18 test cases, with an average Dice coefficient of 88% for the liver and 53% for the liver tumor. Liver segmentation was carried out with rather a high accuracy; this could be achieved for liver dosimetry during systemic or selective radiation therapies as well as for attenuation correction in PET/MRI scanners. Nevertheless, the delineation of liver lesions was not optimal; therefore, tumor detection was not practical by the proposed method on clinical data.

ASD-Net: a novel U-Net based asymmetric spatial-channel convolution network for precise kidney and kidney tumor image segmentation

Early intervention in tumors can greatly improve human survival rates. With the development of deep learning technology, automatic image segmentation has taken a prominent role in the field of medical image analysis. Manually segmenting kidneys on CT images is a tedious task, and due to the diversity of these images and varying technical skills of professionals, segmentation results can be inconsistent. To address this problem, a novel ASD-Net network is proposed in this paper for kidney and kidney tumor segmentation tasks. First, the proposed network employs newly designed Adaptive Spatial-channel Convolution Optimization (ASCO) blocks to capture anisotropic information in the images. Then, other newly designed blocks, i.e., Dense Dilated Enhancement Convolution (DDEC) blocks, are utilized to enhance feature propagation and reuse it across the network, thereby improving its segmentation accuracy. To allow the network to segment complex and small kidney tumors more effectively, the Atrous Spatial Pyramid Pooling (ASPP) module is incorporated in its middle layer. With its generalized pyramid feature, this module enables the network to better capture and understand context information at various scales within the images. In addition to this, the concurrent spatial and channel squeeze & excitation (scSE) attention mechanism is adopted to better comprehend and manage context information in the images. Additional encoding layers are also added to the base (U-Net) and connected to the original encoding layer through skip connections. The resultant enhanced U-Net structure allows for better extraction and merging of high-level and low-level features, further boosting the network's ability to restore segmentation details. In addition, the combined Binary Cross Entropy (BCE)-Dice loss is utilized as the network's loss function. Experiments, conducted on the KiTS19 dataset, demonstrate that the proposed ASD-Net network outperforms the existing segmentation networks according to all evaluation metrics used, except for recall in the case of kidney tumor segmentation, where it takes the second place after Attention-UNet.

CRCS: An automatic image processing pipeline for hormone level analysis of Cushing's disease

Due to the abnormal secretion of adreno-cortico-tropic-hormone (ACTH) by tumors, Cushing's disease leads to hypercortisonemia, a precursor to a series of metabolic disorders and serious complications. Cushing's disease has high recurrence rate, short recurrence time and undiscovered recurrence reason after surgical resection. Qualitative or quantitative automatic image analysis of histology images can potentially in providing insights into Cushing's disease, but still no software has been available to the best of our knowledge. In this study, we propose a quantitative image analysis-based pipeline CRCS, which aims to explore the relationship between the expression level of ACTH in normal cell tissues adjacent to tumor cells and the postoperative prognosis of patients. CRCS mainly consists of image-level clustering, cluster-level multi-modal image registration, patch-level image classification and pixel-level image segmentation on the whole slide imaging (WSI). On both image registration and classification tasks, our method CRCS achieves state-of-the-art performance compared to recently published methods on our collected benchmark dataset. In addition, CRCS achieves an accuracy of 0.83 for postoperative prognosis of 12 cases. CRCS demonstrates great potential for instrumenting automatic diagnosis and treatment for Cushing's disease.

CDRIME-MTIS: An enhanced rime optimization-driven multi-threshold segmentation for COVID-19 X-ray images

To improve the detection of COVID-19, this paper researches and proposes an effective swarm intelligence algorithm-driven multi-threshold image segmentation (MTIS) method. First, this paper proposes a novel RIME structure integrating the Co-adaptive hunting and dispersed foraging strategies, called CDRIME. Specifically, the Co-adaptive hunting strategy works in coordination with the basic search rules of RIME at the individual level, which not only facilitates the algorithm to explore the global optimal solution but also enriches the population diversity to a certain extent. The dispersed foraging strategy further enriches the population diversity to help the algorithm break the limitation of local search and thus obtain better convergence. Then, on this basis, a new multi-threshold image segmentation method is proposed by combining the 2D non-local histogram with 2D Kapur entropy, called CDRIME-MTIS. Finally, the results of experiments based on IEEE CEC2017, IEEE CEC2019, and IEEE CEC2022 demonstrate that CDRIME has superior performance than some other basic, advanced, and state-of-the-art algorithms in terms of global search, convergence performance, and escape from local optimality. Meanwhile, the segmentation experiments on COVID-19 X-ray images demonstrate that CDRIME is more advantageous than RIME and other peers in terms of segmentation effect and adaptability to different threshold levels. In conclusion, the proposed CDRIME significantly enhances the global optimization performance and image segmentation of RIME and has great potential to improve COVID-19 diagnosis.

An automated method for tendon image segmentation on ultrasound using grey-level co-occurrence matrix features and hidden Gaussian Markov random fields

BACKGROUND

Despite knowledge of qualitative changes that occur on ultrasound in tendinopathy, there is currently no objective and reliable means to quantify the severity or prognosis of tendinopathy on ultrasound.

OBJECTIVE

The primary objective of this study is to produce a quantitative and automated means of inferring potential structural changes in tendinopathy by developing and implementing an algorithm which performs a texture based segmentation of tendon ultrasound (US) images.

METHOD

A model-based segmentation approach is used which combines Gaussian mixture models, Markov random field theory and grey-level co-occurrence (GLCM) features. The algorithm is trained and tested on 49 longitudinal B-mode ultrasound images of the Achilles tendons which are labelled as tendinopathic (24) or healthy (25). Hyperparameters are tuned, using a training set of 25 images, to optimise a decision tree based classification of the images from texture class proportions. We segment and classify the remaining test images using the decision tree.

RESULTS

Our approach successfully detects a difference in the texture profiles of tendinopathic and healthy tendons, with 22/24 of the test images accurately classified based on a simple texture proportion cut-off threshold. Results for the tendinopathic images are also collated to gain insight into the topology of structural changes that occur with tendinopathy. It is evident that distinct textures, which are predominantly present in tendinopathic tendons, appear most commonly near the transverse boundary of the tendon, though there was a large variability among diseased tendons.

CONCLUSION

The GLCM based segmentation of tendons under ultrasound resulted in distinct segmentations between healthy and tendinopathic tendons and provides a potential tool to objectively quantify damage in tendinopathy.

Accurate multilevel thresholding image segmentation via oppositional Snake Optimization algorithm: Real cases with liver disease

Liver-related diseases significantly contribute to global mortality rates. Accurate segmentation of liver disease from CT scans is essential for early diagnosis and treatment selection, particularly in computer-aided diagnosis (CAD) systems. To address challenges posed by inconsistent liver presence and unclear boundaries, an enhanced Snake Optimization (SO) algorithm is proposed that integrates with opposition-based learning (OBL) called (SO-OBL), proving effective in global optimization and multilevel image segmentation. Experiments using CEC'2022 test functions compare SO-OBL with eleven recent and state-of-the-art metaheuristic algorithms, demonstrating its superior performance. Additionally, an advanced liver disease segmentation model based on SO-OBL incorporates an optimized multilevel thresholding technique, leveraging Otsu's function. Notable segmentation metric results, including FSIM = 0.947, SSIM = 0.941, PSNR = 24.876, MSE = 236.88, and execution time = 0.281, underscore the model's efficiency and potential for accurate diagnosis in CAD systems.

Comparative evaluation of uncertainty estimation and decomposition methods on liver segmentation

PURPOSE

Deep neural networks need to be able to indicate error likelihood via reliable estimates of their predictive uncertainty when used in high-risk scenarios, such as medical decision support. This work contributes a systematic overview of state-of-the-art approaches for decomposing predictive uncertainty into aleatoric and epistemic components, and a comprehensive comparison for Bayesian neural networks (BNNs) between mutual information decomposition and the explicit modelling of both uncertainty types via an additional loss-attenuating neuron.

METHODS

Experiments are performed in the context of liver segmentation in CT scans. The quality of the uncertainty decomposition in the resulting uncertainty maps is qualitatively evaluated, and quantitative behaviour of decomposed uncertainties is systematically compared for different experiment settings with varying training set sizes, label noise, and distribution shifts.

RESULTS

Our results show the mutual information decomposition to robustly yield meaningful aleatoric and epistemic uncertainty estimates, while the activation of the loss-attenuating neuron appears noisier with non-trivial convergence properties. We found that the addition of a heteroscedastic neuron does not significantly improve segmentation performance or calibration, while slightly improving the quality of uncertainty estimates.

CONCLUSIONS

Mutual information decomposition is simple to implement, has mathematically pleasing properties, and yields meaningful uncertainty estimates that behave as expected under controlled changes to our data set. The additional extension of BNNs with loss-attenuating neurons provides no improvement in terms of segmentation performance or calibration in our setting, but marginal benefits regarding the quality of decomposed uncertainties.

Fetal Ultrasound Segmentation and Measurements Using Appearance and Shape Prior Based Density Regression with Deep CNN and Robust Ellipse Fitting

Accurately segmenting the structure of the fetal head (FH) and performing biometry measurements, including head circumference (HC) estimation, stands as a vital requirement for addressing abnormal fetal growth during pregnancy under the expertise of experienced radiologists using ultrasound (US) images. However, accurate segmentation and measurement is a challenging task due to image artifact, incomplete ellipse fitting, and fluctuations due to FH dimensions over different trimesters. Also, it is highly time-consuming due to the absence of specialized features, which leads to low segmentation accuracy. To address these challenging tasks, we propose an automatic density regression approach to incorporate appearance and shape priors into the deep learning-based network model (DR-ASPnet) with robust ellipse fitting using fetal US images. Initially, we employed multiple pre-processing steps to remove unwanted distortions, variable fluctuations, and a clear view of significant features from the US images. Then some form of augmentation operation is applied to increase the diversity of the dataset. Next, we proposed the hierarchical density regression deep convolutional neural network (HDR-DCNN) model, which involves three network models to determine the complex location of FH for accurate segmentation during the training and testing processes. Then, we used post-processing operations using contrast enhancement filtering with a morphological operation model to smooth the region and remove unnecessary artifacts from the segmentation results. After post-processing, we applied the smoothed segmented result to the robust ellipse fitting-based least square (REFLS) method for HC estimation. Experimental results of the DR-ASPnet model obtain 98.86% dice similarity coefficient (DSC) as segmentation accuracy, and it also obtains 1.67 mm absolute distance (AD) as measurement accuracy compared to other state-of-the-art methods. Finally, we achieved a 0.99 correlation coefficient (CC) in estimating the measured and predicted HC values on the HC18 dataset.

The critical mixed transport process in remediation agent radial injection into contaminated aquifer plumes

Accurately depicting the subsurface mixing of radially injected remediation agents with contaminated plumes remains paramount yet challenging for understanding and simulating reactive transport. To address this, the present research employed the mixing dynamics of a potassium permanganate plume injected into a pre-existing contaminated plume. Through combining colour deconvolution and thresholding, we effectively isolated local mixing values within the Gaussian annular narrow mixing zone from the noise of mixed double-plume images. Key findings revealed increasing injection rate promotes plume mixing while adding xanthan gum to increase fluid viscosity moderates interface mixing, reducing mixing zone width by 25.3% and 37.4% for 100 mg/L and 400 mg/L xanthan gum, respectively. Grain size is pivotal, with a 30% increase in mixing areas observed in coarse-grained sands over medium-grained sands. Balancing sufficient mixing and preventing contaminated plume growth is essential for effective remediation. Injection rates below 5 mL/min may suppress contaminated plume expansion, albeit at the possible cost of protracted remediation durations. For the attainment of optimal remediation, it's imperative to harmonize robust mixing processes with the mitigation of contaminated plume expansion - a balance that adding xanthan gum during the initial injection phase seems poised to achieve (xanthan gum optimized the average mixing index (AMI)). These findings provide valuable insights into groundwater plume mixing, supporting effective remediation strategies.

Rice pest dataset supports the construction of smart farming systems

Rice holds a significant position in the global food supply chain, particularly in Asian, African, and Latin American countries. However, rice pests and diseases cause significant damage to the supply and growth of the rice cultivation industry. Therefore, this article provides a high-quality dataset that has been reviewed by agricultural experts. The dataset is well-suited to support the development of automation systems and smart farming practices. It plays a vital role in facilitating the automatic construction, detection, and classification of rice diseases. However, challenges arise due to the diversity of the dataset collected from various sources, varying in terms of disease types and sizes. This necessitates support for upgrading and enhancing the dataset through various operations in data processing, preprocessing, and statistical analysis. The dataset is provided completely free of charge and has been rigorously evaluated by agricultural experts, making it a reliable resource for system development, research, and communication needs.

MCNet: A multi-level context-aware network for the segmentation of adrenal gland in CT images

Accurate segmentation of the adrenal gland from abdominal computed tomography (CT) scans is a crucial step towards facilitating the computer-aided diagnosis of adrenal-related diseases such as essential hypertension and adrenal tumors. However, the small size of the adrenal gland, which occupies less than 1% of the abdominal CT slice, poses a significant challenge to accurate segmentation. To address this problem, we propose a novel multi-level context-aware network (MCNet) to segment adrenal glands in CT images. Our MCNet mainly consists of two components, i.e., the multi-level context aggregation (MCA) module and multi-level context guidance (MCG) module. Specifically, the MCA module employs multi-branch dilated convolutional layers to capture geometric information, which enables handling of changes in complex scenarios such as variations in the size and shape of objects. The MCG module, on the other hand, gathers valuable features from the shallow layer and leverages the complete utilization of feature information at different resolutions in various codec stages. Finally, we evaluate the performance of the MCNet on two CT datasets, including our clinical dataset (Ad-Seg) and a publicly available dataset known as Distorted Golden Standards (DGS), from different perspectives. Compared to ten other state-of-the-art segmentation methods, our MCNet achieves 71.34% and 75.29% of the best Dice similarity coefficient on the two datasets, respectively, which is at least 2.46% and 1.19% higher than other segmentation methods.

Hybrid representation-enhanced sampling for Bayesian active learning in musculoskeletal segmentation of lower extremities

PURPOSE

Manual annotations for training deep learning models in auto-segmentation are time-intensive. This study introduces a hybrid representation-enhanced sampling strategy that integrates both density and diversity criteria within an uncertainty-based Bayesian active learning (BAL) framework to reduce annotation efforts by selecting the most informative training samples.

METHODS

The experiments are performed on two lower extremity datasets of MRI and CT images, focusing on the segmentation of the femur, pelvis, sacrum, quadriceps femoris, hamstrings, adductors, sartorius, and iliopsoas, utilizing a U-net-based BAL framework. Our method selects uncertain samples with high density and diversity for manual revision, optimizing for maximal similarity to unlabeled instances and minimal similarity to existing training data. We assess the accuracy and efficiency using dice and a proposed metric called reduced annotation cost (RAC), respectively. We further evaluate the impact of various acquisition rules on BAL performance and design an ablation study for effectiveness estimation.

RESULTS

In MRI and CT datasets, our method was superior or comparable to existing ones, achieving a 0.8% dice and 1.0% RAC increase in CT (statistically significant), and a 0.8% dice and 1.1% RAC increase in MRI (not statistically significant) in volume-wise acquisition. Our ablation study indicates that combining density and diversity criteria enhances the efficiency of BAL in musculoskeletal segmentation compared to using either criterion alone.

CONCLUSION

Our sampling method is proven efficient in reducing annotation costs in image segmentation tasks. The combination of the proposed method and our BAL framework provides a semi-automatic way for efficient annotation of medical image datasets.

Deep learning for real-time multi-class segmentation of artefacts in lung ultrasound

Lung ultrasound (LUS) has emerged as a safe and cost-effective modality for assessing lung health, particularly during the COVID-19 pandemic. However, interpreting LUS images remains challenging due to its reliance on artefacts, leading to operator variability and limiting its practical uptake. To address this, we propose a deep learning pipeline for multi-class segmentation of objects (ribs, pleural line) and artefacts (A-lines, B-lines, B-line confluence) in ultrasound images of a lung training phantom. Lightweight models achieved a mean Dice Similarity Coefficient (DSC) of 0.74, requiring fewer than 500 training images. Applying this method in real-time, at up to 33.4 frames per second in inference, allows enhanced visualisation of these features in LUS images. This could be useful in providing LUS training and helping to address the skill gap. Moreover, the segmentation masks obtained from this model enable the development of explainable measures of disease severity, which have the potential to assist in the triage and management of patients. We suggest one such semi-quantitative measure called the B-line Artefact Score, which is related to the percentage of an intercostal space occupied by B-lines and in turn may be associated with the severity of a number of lung conditions. Moreover, we show how transfer learning could be used to train models for small datasets of clinical LUS images, identifying pathologies such as simple pleural effusions and lung consolidation with DSC values of 0.48 and 0.32 respectively. Finally, we demonstrate how such DL models could be translated into clinical practice, implementing the phantom model alongside a portable point-of-care ultrasound system, facilitating bedside assessment and improving the accessibility of LUS.

Accurate malocclusion tooth segmentation method based on a level set with adaptive edge feature enhancement

Objective

This study aimed to accurately segment teeth under complex oral conditions, including complex structural interference among adjacent teeth or malocclusion conditions, such as tooth rotation and displacement caused by dental crowding.

Study design

Cone-beam computed tomography (CBCT) images were obtained from 19 patients with complex oral conditions, and a three-step solution was proposed. This study used a global convex level-set model to extract bony tissue and developed a flexible curve extraction method for separating neighbouring teeth under complex structural interference. In addition, a local level-set model with adaptive edge feature enhancement was proposed to segment individual teeth precisely. This model adaptively enhances edge features based on the structure of the root boundary and accurately distinguishes between the close-contact root and alveolar bone resulting from tooth rotation or displacement.

Results

The experimental results showed that the average Dice similarity coefficient values for incisors, canines, premolars, and molars were 93.30%, 93.47%, 93.24%, and 93.89%, respectively, and the average tooth centroid distances were 0.66, 0.61, 0.87, and 0.80 mm, respectively.

Conclusion

The proposed method can effectively segment teeth without relying on highly precise annotated datasets, yielding satisfactory results even under complex structural interference between adjacent teeth or tooth rotation and displacement caused by dental crowding. It is more robust than the other methods and provides valuable data for further research and clinical practice.

Development of a semi-automatic segmentation technique based on mean magnetic resonance imaging intensity thresholding for volumetric quantification of plexiform neurofibromas

Rationale and objectives

Plexiform neurofibromas (PNs) are peripheral nerve tumors that occur in 25-50 % of patients with neurofibromatosis type 1. PNs may have complex, diffused, and irregular shapes. The objective of this work was to develop a volumetric quantification method for PNs as clinical assessment is currently based on unidimensional measurement.

Materials and methods

A semi-automatic segmentation technique based on mean magnetic resonance imaging (MRI) intensity thresholding (SSTMean) was developed and compared to a similar and previously published technique based on minimum image intensity thresholding (SSTMini). The performance (volume and computation time) of the two techniques was compared to manual tracings of 15 tumors of different locations, shapes, and sizes. Performance was also assessed using different MRI sequences. Reproducibility was assessed by inter-observer analysis.

Results

When compared to manual tracing, quantification performed with SSTMean was not significantly different (mean difference: 1.2 %), while volumes computed by SSTMini were significantly different (p < .0001, mean difference: 13.4 %). Volumes quantified by SSTMean were also significantly different than the ones assessed by SSTMini (p < .0001). Using SSTMean, volumes quantified with short TI inversion recovery, T1-, and T2-weighted imaging were not significantly different. Computation times used by SSTMean and SSTMini were significantly lower than for manual segmentation (p < .0001). The highest difference measured by two users was 8 cm3.

Conclusion

Our method showed accuracy compared to a current gold standard (manual tracing) and reproducibility between users. The refined segmentation threshold and the possibility to define multiple regions-of-interest to initiate segmentation may have contributed to its performance. The versatility and speed of our method may prove useful to better monitor volumetric changes in lesions of patients enrolled in clinical trials to assessing response to therapy.

Comparison of Otsu and an adapted Chan-Vese method to determine thyroid active volume using Monte Carlo generated SPECT images

BACKGROUND

The Otsu method and the Chan-Vese model are two methods proven to perform well in determining volumes of different organs and specific tissue fractions. This study aimed to compare the performance of the two methods regarding segmentation of active thyroid gland volumes, reflecting different clinical settings by varying the parameters: gland size, gland activity concentration, background activity concentration and gland activity concentration heterogeneity.

METHODS

A computed tomography was performed on three playdough thyroid phantoms with volumes 20, 35 and 50 ml. The image data were separated into playdough and water based on Hounsfield values. Sixty single photon emission computed tomography (SPECT) projections were simulated by Monte Carlo method with isotope Technetium-99 m ([Formula: see text]Tc). Linear combinations of SPECT images were made, generating 12 different combinations of volume and background: each with both homogeneous thyroid activity concentration and three hotspots of different relative activity concentrations (48 SPECT images in total). The relative background levels chosen were 5 %, 10 %, 15 % and 20 % of the phantom activity concentration and the hotspot activities were 100 % (homogeneous case) 150 %, 200 % and 250 %. Poisson noise, (coefficient of variation of 0.8 at a 20 % background level, scattering excluded), was added before reconstruction was done with the Monte Carlo-based SPECT reconstruction algorithm Sahlgrenska Academy reconstruction code (SARec). Two different segmentation algorithms were applied: Otsu's threshold selection method and an adaptation of the Chan-Vese model for active contours without edges; the results were evaluated concerning relative volume, mean absolute error and standard deviation per thyroid volume, as well as dice similarity coefficient.

RESULTS

Both methods segment the images well and deviate similarly from the true volumes. They seem to slightly overestimate small volumes and underestimate large ones. Different background levels affect the two methods similarly as well. However, the Chan-Vese model deviates less and paired t-testing showed significant difference between distributions of dice similarity coefficients (p-value [Formula: see text]).

CONCLUSIONS

The investigations indicate that the Chan-Vese model performs better and is slightly more robust, while being more challenging to implement and use clinically. There is a trade-off between performance and user-friendliness.

Evaluating artificial intelligence for comparative radiography

INTRODUCTION

Comparative radiography is a forensic identification and shortlisting technique based on the comparison of skeletal structures in ante-mortem and post-mortem images. The images (e.g., 2D radiographs or 3D computed tomographies) are manually superimposed and visually compared by a forensic practitioner. It requires a significant amount of time per comparison, limiting its utility in large comparison scenarios.

METHODS

We propose and validate a novel framework for automating the shortlisting of candidates using artificial intelligence. It is composed of (1) a segmentation method to delimit skeletal structures' silhouettes in radiographs, (2) a superposition method to generate the best simulated "radiographs" from 3D images according to the segmented radiographs, and (3) a decision-making method for shortlisting all candidates ranked according to a similarity metric.

MATERIAL

The dataset is composed of 180 computed tomographies and 180 radiographs where the frontal sinuses are visible. Frontal sinuses are the skeletal structure analyzed due to their high individualization capability.

RESULTS

Firstly, we validate two deep learning-based techniques for segmenting the frontal sinuses in radiographs, obtaining high-quality results. Secondly, we study the framework's shortlisting capability using both automatic segmentations and superimpositions. The obtained superimpositions, based only on the superimposition metric, allowed us to filter out 40% of the possible candidates in a completely automatic manner. Thirdly, we perform a reliability study by comparing 180 radiographs against 180 computed tomographies using manual segmentations. The results allowed us to filter out 73% of the possible candidates. Furthermore, the results are robust to inter- and intra-expert-related errors.

Lightweight colon polyp segmentation algorithm based on improved DeepLabV3

To address the problems that the current polyp segmentation model is complicated and the segmentation accuracy needs to be further improved, a lightweight polyp segmentation network model Li-DeepLabV3+ is proposed. Firstly, the optimized MobileNetV2 network is used as the backbone network to reduce the model complexity. Secondly, an improved simple pyramid pooling module is used to replace the original Atrous Spatial Pyramid Pooling structure, which improves the model training efficiency of the model while reducing the model parameters. Finally, to enhance the feature representation, in the feature fusion module, the low-level feature and the high-level feature are fused using the improved Unified Attention Fusion Module, which applies both channel and spatial attention to enrich the fused features, thus obtaining more boundary information. The model was combined with transfer learning for training and validation on the CVC-ClinicDB and Kvasir SEG datasets, and the generalization of the model was verified across the datasets. The experiment results show that the Li-DeepLabV3+ model has superior advantages in segmentation accuracy and segmentation speed, and has certain generalization abilities.

C-DARL: Contrastive diffusion adversarial representation learning for label-free blood vessel segmentation

Blood vessel segmentation in medical imaging is one of the essential steps for vascular disease diagnosis and interventional planning in a broad spectrum of clinical scenarios in image-based medicine and interventional medicine. Unfortunately, manual annotation of the vessel masks is challenging and resource-intensive due to subtle branches and complex structures. To overcome this issue, this paper presents a self-supervised vessel segmentation method, dubbed the contrastive diffusion adversarial representation learning (C-DARL) model. Our model is composed of a diffusion module and a generation module that learns the distribution of multi-domain blood vessel data by generating synthetic vessel images from diffusion latent. Moreover, we employ contrastive learning through a mask-based contrastive loss so that the model can learn more realistic vessel representations. To validate the efficacy, C-DARL is trained using various vessel datasets, including coronary angiograms, abdominal digital subtraction angiograms, and retinal imaging. Experimental results confirm that our model achieves performance improvement over baseline methods with noise robustness, suggesting the effectiveness of C-DARL for vessel segmentation.Our source code is available at https://github.com/boahK/MEDIA_CDARL.2.

Filter pruning for convolutional neural networks in semantic image segmentation

The remarkable performance of Convolutional Neural Networks (CNNs) has increased their use in real-time systems and devices with limited resources. Hence, compacting these networks while preserving accuracy has become necessary, leading to multiple compression methods. However, the majority require intensive iterative procedures and do not delve into the influence of the used data. To overcome these issues, this paper presents several contributions, framed in the context of explainable Artificial Intelligence (xAI): (a) two filter pruning methods for CNNs, which remove the less significant convolutional kernels; (b) a fine-tuning strategy to recover generalization; (c) a layer pruning approach for U-Net; and (d) an explanation of the relationship between performance and the used data. Filter and feature maps information are used in the pruning process: Principal Component Analysis (PCA) is combined with a next-convolution influence-metric, while the latter and the mean standard deviation are used in an importance score distribution-based method. The developed strategies are generic, and therefore applicable to different models. Experiments demonstrating their effectiveness are conducted over distinct CNNs and datasets, focusing mainly on semantic segmentation (using U-Net, DeepLabv3+, SegNet, and VGG-16 as highly representative models). Pruned U-Net on agricultural benchmarks achieves 98.7% parameters and 97.5% FLOPs drop, with a 0.35% gain in accuracy. DeepLabv3+ and SegNet on CamVid reach 46.5% and 72.4% parameters reduction and a 51.9% and 83.6% FLOPs drop respectively, with almost no decrease in accuracy. VGG-16 on CIFAR-10 obtains up to 86.5% parameter and 82.2% FLOPs decrease with a 0.78% accuracy gain.

Manual versus deep learning measurements to evaluate cumulus expansion of bovine oocytes and its relationship with embryo development in vitro

Cumulus expansion is an important indicator of oocyte maturation and has been suggested to be indicative of greater oocyte developmental capacity. Although multiple methods have been described to assess cumulus expansion, none of them is considered a gold standard. Additionally, these methods are subjective and time-consuming. In this manuscript, the reliability of three cumulus expansion measurement methods was assessed, and a deep learning model was created to automatically perform the measurement. Cumulus expansion of 232 cumulus-oocyte complexes was evaluated by three independent observers using three methods: (1) measurement of the cumulus area, (2) measurement of three distances between the zona pellucida and outer cumulus, and (3) scoring cumulus expansion on a 5-point Likert scale. The reliability of the methods was calculated in terms of intraclass-correlation coefficients (ICC) for both inter- and intra-observer agreements. The area method resulted in the best overall inter-observer agreement with an ICC of 0.89 versus 0.54 and 0.30 for the 3-distance and scoring methods, respectively. Therefore, the area method served as the base to create a deep learning model, AI-xpansion, which reaches a human-level performance in terms of average rank, bias and variance. To evaluate the accuracy of the methods, the results of cumulus expansion calculations were linked to embryonic development. Cumulus expansion had increased significantly in oocytes that achieved successful embryo development when measured by AI-xpansion, the area- or 3-distance method, while this was not the case for the scoring method. Measuring the area is the most reliable method to manually evaluate cumulus expansion, whilst deep learning automatically performs the calculation with human-level precision and high accuracy and could therefore be a valuable prospective tool for embryologists.

M-DDC: MRI based demyelinative diseases classification with U-Net segmentation and convolutional network

Childhood demyelinative diseases classification (DDC) with brain magnetic resonance imaging (MRI) is crucial to clinical diagnosis. But few attentions have been paid to DDC in the past. How to accurately differentiate pediatric-onset neuromyelitis optica spectrum disorder (NMOSD) from acute disseminated encephalomyelitis (ADEM) based on MRI is challenging in DDC. In this paper, a novel architecture M-DDC based on joint U-Net segmentation network and deep convolutional network is developed. The U-Net segmentation can provide pixel-level structure information, that helps the lesion areas location and size estimation. The classification branch in DDC can detect the regions of interest inside MRIs, including the white matter regions where lesions appear. The performance of the proposed method is evaluated on MRIs of 201 subjects recorded from the Children's Hospital of Zhejiang University School of Medicine. The comparisons show that the proposed DDC achieves the highest accuracy of 99.19% and dice of 71.1% for ADEM and NMOSD classification and segmentation, respectively.