Efficient Segmentation Method Research Articles

An Efficient Point Cloud Semantic Segmentation Method Based on Bilateral Enhancement and Random Sampling

Point cloud semantic segmentation is of utmost importance in practical applications. However, most existing methods have evolved to be incredibly intricate, leading to a rise in complexity that has made them increasingly impractical for real-world utilization. The escalating complexity of these methods has resulted in a deterioration in their efficiency and ease of implementation, making them less suitable for use in time-sensitive and resource-constrained environments. Towards this issue, we propose an efficient and lightweight segmentation method, able to achieve a remarkable performance in terms of both segmentation accuracy, training speed, and space consumption. Specifically, we first propose to adopt random sampling to replace the original one to obtain more efficiency. Moreover, a lightweight decoding module and an improved bilateral enhancement (BAE) module are developed to further improve the performance. The proposed method achieved a 73.6% and 60.7% mIoU on the S3DIS and Semantickitti datasets, respectively. In the future, the random sampling and the proposed BAE module can be adopted in a more concise and lightweight network to achieve faster and more-accurate point cloud segmentation.

Segmentation of hard exudate lesions in color fundus image using two-stage CNN-based methods

The presence of hard exudate (EX) lesions is an early clinical symptom of Diabetic Retinopathy (DR); its accurate segmentation is essential for diagnosis and treatment. Automatic segmentation of EX lesions is challenging because they have a wide range of sizes, shapes, and brightness, and can be confused with soft exudate lesions. In this paper, we present and assess an efficient segmentation method for interactively segmenting EX lesions in fundus images from DR patients. Our approach consisted of two main stages: (1) the first stage generates an automatic segmentation using a CNN-based method. (2) If the obtained segmentation is suboptimal, the second stage is used to refine each inaccurately segmented region using a CNN-based interactive segmentation method. To train and evaluate method performance, we used two public datasets (IDRiD and DDR) and two local datasets from two medical centers (EHos and EWT) containing a wide range of EX lesions. Several state-of-the art CNN-based models and the proposed method were trained and evaluated on IDRiD, DDR and EHos datasets using the area under the precision–recall curve (AUPR), the dice similarity coefficient (Dice) and the intersection over union (IoU). In addition, EWT dataset was used to evaluate the methods without retraining the CNN models. The evaluation results using IDRiD dataset showed that the proposed method achieved AUPR, mean Dice and mean IoU scores of 0.893, 76.6% and 62.4%, respectively, which are comparable to the state-of-the-art method on EX segmentation. When evaluated on the DDR dataset with large portion of small EX lesions, the proposed method achieved AUPR, mean Dice and mean IoU scores of 0.693, 60.8% and 45.5%, yielding state-of-the-art performance. Furthermore, with the main evaluation metric being AUPR, the evaluation results on two datasets, EHos and ETW, with scores of 0.689 and 0.667 demonstrated superiority of the proposed method over the other segmentation methods. The proposed two-stage CNN-based segmentation method is an useful approach for accurately segmenting EX lesions in color fundus images, enabling lesion quantification to aid experts in treating patients with DR. The source code, the trained CNN models and the data are publicly available at: https://github.com/dvquang2000/Interactive_HardExudates_Segmentation.

An efficient segmentation method based on semi-supervised learning for seafloor monitoring in Pujada Bay, Philippines

The semantic segmentation of marine images makes it easier to describe seafloor scenes and track marine organisms. However, creating human-annotated datasets for image segmentation requires significant time and effort. Therefore, this study proposes a semisupervised learning approach that combines the mean-teacher model and U-Net architecture for segmenting seafloor images obtained from the Philippines. The proposed method performs segmentation for categories, including corals, sea urchins, sea stars, and seagrass. Traditional manual labelling methods are used for coral, sea urchins, and sea stars. For the seagrass category, which is challenging to label manually, we used the K-means clustering algorithm to obtain corresponding labelled datasets based on the characteristics of such images and evaluated the feasibility. Compared with the U-Net-based supervised method, the semi-supervised method used in this study achieved good results and accuracy values, even with fewer labelled images.

SUnet: A multi-organ segmentation network based on multiple attention

Organ segmentation in abdominal or thoracic computed tomography (CT) images plays a crucial role in medical diagnosis as it enables doctors to locate and evaluate organ abnormalities quickly, thereby guiding surgical planning, and aiding treatment decision-making. This paper proposes a novel and efficient medical image segmentation method called SUnet for multi-organ segmentation in the abdomen and thorax. SUnet is a fully attention-based neural network. Firstly, an efficient spatial reduction attention (ESRA) module is introduced not only to extract image features better, but also to reduce overall model parameters, and to alleviate overfitting. Secondly, SUnet's multiple attention-based feature fusion module enables effective cross-scale feature integration. Additionally, an enhanced attention gate (EAG) module is considered by using grouped convolution and residual connections, providing richer semantic features. We evaluate the performance of the proposed model on synapse multiple organ segmentation dataset and automated cardiac diagnostic challenge dataset. SUnet achieves an average Dice of 84.29% and 92.25% on these two datasets, respectively, outperforming other models of similar complexity and size, and achieving state-of-the-art results.

Metal Abrasive Image Segmentation Algorithm Based on K-means Clustering

Metal abrasive image segmentation is one of the important image processing tasks in the industrial field. However, due to the complex color and texture characteristics of metal abrasive images, as well as difficult factors such as noise and lighting changes, traditional image segmentation methods often fail to achieve high accuracy and stability. In order to solve this problem, a metal abrasive image segmentation algorithm based on K-means clustering is proposed. The algorithm applies the K-means clustering algorithm to the image segmentation of metal abrasive particles, and realizes the separation of metal abrasive grains from the background by clustering the color features of image pixels. Experimental results show that our algorithm shows good accuracy and stability in the metal abrasive image segmentation task, and has high efficiency. Therefore, the algorithm provides an accurate and efficient method for metal abrasive image segmentation.

MixSeg: a lightweight and accurate mix structure network for semantic segmentation of apple leaf disease in complex environments.

Semantic segmentation is effective in dealing with complex environments. However, the most popular semantic segmentation methods are usually based on a single structure, they are inefficient and inaccurate. In this work, we propose a mix structure network called MixSeg, which fully combines the advantages of convolutional neural network, Transformer, and multi-layer perception architectures. Specifically, MixSeg is an end-to-end semantic segmentation network, consisting of an encoder and a decoder. In the encoder, the Mix Transformer is designed to model globally and inject local bias into the model with less computational cost. The position indexer is developed to dynamically index absolute position information on the feature map. The local optimization module is designed to optimize the segmentation effect of the model on local edges and details. In the decoder, shallow and deep features are fused to output accurate segmentation results. Taking the apple leaf disease segmentation task in the real scene as an example, the segmentation effect of the MixSeg is verified. The experimental results show that MixSeg has the best segmentation effect and the lowest parameters and floating point operations compared with the mainstream semantic segmentation methods on small datasets. On apple alternaria blotch and apple grey spot leaf image datasets, the most lightweight MixSeg-T achieves 98.22%, 98.09% intersection over union for leaf segmentation and 87.40%, 86.20% intersection over union for disease segmentation. Thus, the performance of MixSeg demonstrates that it can provide a more efficient and stable method for accurate segmentation of leaves and diseases in complex environments.

An Efficient Brain Tumor Segmentation Method Based on Adaptive Moving Self-Organizing Map and Fuzzy K-Mean Clustering.

Brain tumors in Magnetic resonance image segmentation is challenging research. With the advent of a new era and research into machine learning, tumor detection and segmentation generated significant interest in the research world. This research presents an efficient tumor detection and segmentation technique using an adaptive moving self-organizing map and Fuzzyk-mean clustering (AMSOM-FKM). The proposed method mainly focused on tumor segmentation using extraction of the tumor region. AMSOM is an artificial neural technique whose training is unsupervised. This research utilized the online Kaggle Brats-18 brain tumor dataset. This dataset consisted of 1691 images. The dataset was partitioned into 70% training, 20% testing, and 10% validation. The proposed model was based on various phases: (a) removal of noise, (b) selection of feature attributes, (c) image classification, and (d) tumor segmentation. At first, the MR images were normalized using the Wiener filtering method, and the Gray level co-occurrences matrix (GLCM) was used to extract the relevant feature attributes. The tumor images were separated from non-tumor images using the AMSOM classification approach. At last, the FKM was used to distinguish the tumor region from the surrounding tissue. The proposed AMSOM-FKM technique and existing methods, i.e., Fuzzy-C-means and K-mean (FMFCM), hybrid self-organization mapping-FKM, were implemented over MATLAB and compared based on comparison parameters, i.e., sensitivity, precision, accuracy, and similarity index values. The proposed technique achieved more than 10% better results than existing methods.

Performance of alternative manual and automated deep learning segmentation techniques for the prediction of benign and malignant lung nodules.

We aim to evaluate the performance of radiomic biopsy (RB), best-fit bounding box (BB), and a deep-learning-based segmentation method called no-new-U-Net (nnU-Net), compared to the standard full manual (FM) segmentation method for predicting benign and malignant lung nodules using a computed tomography (CT) radiomic machine learning model. A total of 188 CT scans of lung nodules from 2 institutions were used for our study. One radiologist identified and delineated all 188 lung nodules, whereas a second radiologist segmented a subset () of these nodules. Both radiologists employed FM and RB segmentation methods. BB segmentations were generated computationally from the FM segmentations. The nnU-Net, a deep-learning-based segmentation method, performed automatic nodule detection and segmentation. The time radiologists took to perform segmentations was recorded. Radiomic features were extracted from each segmentation method, and models to predict benign and malignant lung nodules were developed. The Kruskal-Wallis and DeLong tests were used to compare segmentation times and areas under the curve (AUC), respectively. For the delineation of the FM, RB, and BB segmentations, the two radiologists required a median time (IQR) of 113 (54 to 251.5), 21 (9.25 to 38), and 16 (12 to 64.25) s, respectively (). In dataset 1, the mean AUC (95% CI) of the FM, RB, BB, and nnU-Net model were 0.964 (0.96 to 0.968), 0.985 (0.983 to 0.987), 0.961 (0.956 to 0.965), and 0.878 (0.869 to 0.888). In dataset 2, the mean AUC (95% CI) of the FM, RB, BB, and nnU-Net model were 0.717 (0.705 to 0.729), 0.919 (0.913 to 0.924), 0.699 (0.687 to 0.711), and 0.644 (0.632 to 0.657). Radiomic biopsy-based models outperformed FM and BB models in prediction of benign and malignant lung nodules in two independent datasets while deep-learning segmentation-based models performed similarly to FM and BB. RB could be a more efficient segmentation method, but further validation is needed.

3D City Reconstruction: A Novel Method for Semantic Segmentation and Building Monomer Construction Using Oblique Photography

Existing 3D city reconstruction via oblique photography can only produce surface models, lacking semantic information about the urban environment and the ability to incorporate all individual buildings. Here, we propose a method for the semantic segmentation of 3D model data from oblique photography and for building monomer construction and implementation. Mesh data were converted into and mapped as point sets clustered to form superpoint sets via rough geometric segmentation, facilitating subsequent feature extractions. In the local neighborhood computation of semantic segmentation, a neighborhood search method based on geodesic distances, improved the rationality of the neighborhood. In addition, feature information was retained via the superpoint sets. Considering the practical requirements of large-scale 3D datasets, this study offers a robust and efficient segmentation method that combines traditional random forest and Markov random field models to segment 3D scene semantics. To address the need for modeling individual and unique buildings, our methodology utilized 3D mesh data of buildings as a data source for specific contour extraction. Model monomer construction and building contour extractions were based on mesh model slices and assessments of geometric similarity, which allowed the simultaneous and automatic achievement of these two processes.

Cooperative Swarm Intelligence Algorithms for Adaptive Multilevel Thresholding Segmentation of COVID-19 CT-Scan Images

The Coronavirus Disease 2019 (COVID-19) is widespread throughout the world and poses a serious threat to public health and safety. A COVID-19 infection can be recognized using computed tomography (CT) scans. To enhance the categorization, some image segmentation techniques are presented to extract regions of interest from COVID-19 CT images. Multi-level thresholding (MLT) is one of the simplest and most effective image segmentation approaches, especially for grayscale images like CT scan images. Traditional image segmentation methods use histogram approaches; however, these approaches encounter some limitations. Now, swarm intelligence inspired meta-heuristic algorithms have been applied to resolve MLT, deemed an NP-hard optimization task. Despite the advantages of using meta-heuristics to solve global optimization tasks, each approach has its own drawbacks. However, the common flaw for most meta-heuristic algorithms is that they are unable to maintain the diversity of their population during the search, which means they might not always converge to the global optimum. This study proposes a cooperative swarm intelligence-based MLT image segmentation approach that hybridizes the advantages of parallel meta-heuristics and MLT for developing an efficient image segmentation method for COVID-19 CT images. An efficient cooperative model-based meta-heuristic called the CPGH is developed based on three practical algorithms: particle swarm optimization (PSO), grey wolf optimizer (GWO), and Harris hawks optimization (HHO). In the cooperative model, the applied algorithms are executed concurrently, and a number of potential solutions are moved across their populations through a procedure called migration after a set number of generations. The CPGH model can solve the image segmentation problem using MLT image segmentation. The proposed CPGH is evaluated using three objective functions, cross-entropy, Otsu’s, and Tsallis, over the COVID-19 CT images selected from open-sourced datasets. Various evaluation metrics covering peak signal-to-noise ratio (PSNR), structural similarity index (SSIM), and universal quality image index (UQI) were employed to quantify the segmentation quality. The overall ranking results of the segmentation quality metrics indicate that the performance of the proposed CPGH is better than conventional PSO, GWO, and HHO algorithms and other state-of-the-art methods for MLT image segmentation. On the tested COVID-19 CT images, the CPGH offered an average PSNR of 24.8062, SSIM of 0.8818, and UQI of 0.9097 using 20 thresholds.

A solution to the stagnation of multi-verse optimization: An efficient method for breast cancer pathologic images segmentation

The multi-verse optimizer (MVO) algorithm has been applied to image segmentation, feature selection, engineering problems, and many other fields. MVO, like other metaheuristic algorithms, still has shortcomings, such as poor convergence speed and quickly falling into local optimum. To address these concerns, this paper proposes CBQMVO, extending the original MVO algorithm with three strategies: covariance matrix adaptation strategy (CMAES), biogeography-based learning strategy (BLS), quasi-reflected and quasi-opposition strategy (QROS). CMAES can make the algorithm approach quickly the current local optimal solution and accelerate the convergence. BLS can enrich the population’s diversity to discourage prematurely and assist the algorithm in jumping out of the local optimum. QROS can increase the probability of search particles falling near the optimal solution. A set of experiments were conducted to evaluate the performance of the CBQMVO. First, the original algorithm comparison experiment on IEEE CEC2014 includes strategy comparison, dimension comparison, exploration/exploitation balance, and population diversity experiments. Then, the advanced algorithm comparison experiment was carried out on IEEE CEC 2014. Furthermore, the champion algorithm comparison experiment was conducted on IEEE CEC2017 and IEEE CEC2020. A series of comparative experimental data demonstrate that CBQMVO has high performance, especially on some unimodal and complex competition functions. In addition, this paper also applied CBQMVO to implement Renyi’s entropy multilevel threshold image segmentation based on the non-local mean 2D histogram (RMIS-2D) on breast cancer pathologic images. Compared with other metaheuristic algorithms and Kapur’s entropy image segmentation, the proposed scheme in this paper has a better segmentation effect.

An Efficient Multilevel Threshold Segmentation Method for Breast Cancer Imaging Based on Metaheuristics Algorithms: Analysis and Validations

Breast cancer is a hazardous disease that should be seriously tackled to reduce its danger in all aspects of the world. Therefore, several imaging ways to detect this disease were considered, but the produced images need to be accurately processed to effectively detect it. Image segmentation is an indispensable step in image processing to segment the homogenous regions that have similar features such as brightness, color, texture, contrast, form, and size. Several techniques like region-based, threshold-based, edge-based, and feature-based clustering have been developed for image segmentation; however, thresholding, which is divided into two classes: bilevel and multilevel, won the highest attention by the researchers due to its simplicity, ease of use and accuracy. The multilevel thresholding-based image segmentation is difficult to be tackled using traditional techniques, especially with increasing the threshold level; therefore, the researchers pay attention to the metaheuristic algorithms which could overcome several hard problems in a reasonable time. In this paper, a new hybrid metaheuristic algorithm based on integrating the jellyfish search algorithm with an effective improvement method is proposed for segmenting the color images of breast cancer, namely the hybrid jellyfish search algorithm HJSO. Experiments are extensively performed to appear the superiority of the proposed algorithm, including validating its performance using various breast cancer images and conducting an extensive comparison with several rival algorithms to explore its effectiveness. The experimental findings, including various performance metrics like fitness values, CPU time, Peak signal-to-noise ratio (PSNR), standard deviation, Features similarity index (FSIM), and Structural similarity index (SSIM), totally show the efficiency of HJSO.

A novel breast cancer detection architecture based on a CNN-CBR system for mammogram classification

This paper presents a novel framework for breast cancer detection using mammogram images. The proposed solution aims to output an explainable classification from a mammogram image. The classification approach uses a Case-Based Reasoning system (CBR). CBR accuracy strongly depends on the quality of the extracted features. To achieve relevant classification, we propose a pipeline that includes image enhancement and data augmentation to improve the quality of extracted features and provide a final diagnosis. An efficient segmentation method based on a U-Net architecture is used to extract Regions of interest (RoI) from mammograms. The purpose is to combine deep learning (DL) with CBR to improve classification accuracy. DL provides accurate mammogram segmentation, while CBR gives an explainable and accurate classification. The proposed approach was tested on the CBIS-DDSM dataset and achieved high performance with an accuracy (Acc) of 86.71 % and a recall of 91.34 %, outperforming some well-known machine learning (ML) and DL approaches.

An efficient multilevel image thresholding method based on improved heap-based optimizer

Image segmentation is the process of separating pixels of an image into multiple classes, enabling the analysis of objects in the image. Multilevel thresholding (MTH) is a method used to perform this task, and the problem is to obtain an optimal threshold that properly segments each image. Methods such as the Kapur entropy or the Otsu method, which can be used as objective functions to determine the optimal threshold, are efficient in determining the best threshold for bi-level thresholding; however, they are not effective for MTH due to their high computational cost. This paper integrates an efficient method for MTH image segmentation called the heap-based optimizer (HBO) with opposition-based learning termed improved heap-based optimizer (IHBO) to solve the problem of high computational cost for MTH and overcome the weaknesses of the original HBO. The IHBO was proposed to improve the convergence rate and local search efficiency of search agents of the basic HBO, the IHBO is applied to solve the problem of MTH using the Otsu and Kapur methods as objective functions. The performance of the IHBO-based method was evaluated on the CEC’2020 test suite and compared against seven well-known metaheuristic algorithms including the basic HBO, salp swarm algorithm, moth flame optimization, gray wolf optimization, sine cosine algorithm, harmony search optimization, and electromagnetism optimization. The experimental results revealed that the proposed IHBO algorithm outperformed the counterparts in terms of the fitness values as well as other performance indicators, such as the structural similarity index (SSIM), feature similarity index (FSIM), peak signal-to-noise ratio. Therefore, the IHBO algorithm was found to be superior to other segmentation methods for MTH image segmentation.

PseudoSegRT: efficient pseudo-labelling for intraoperative OCT segmentation

PurposeRobotic ophthalmic microsurgery has significant potential to help improve the success of challenging procedures and overcome the physical limitations of the surgeon. Intraoperative optical coherence tomography (iOCT) has been reported for the visualisation of ophthalmic surgical manoeuvres, where deep learning methods can be used for real-time tissue segmentation and surgical tool tracking. However, many of these methods rely heavily on labelled datasets, where producing annotated segmentation datasets is a time-consuming and tedious task.MethodsTo address this challenge, we propose a robust and efficient semi-supervised method for boundary segmentation in retinal OCT to guide a robotic surgical system. The proposed method uses U-Net as the base model and implements a pseudo-labelling strategy which combines the labelled data with unlabelled OCT scans during training. After training, the model is optimised and accelerated with the use of TensorRT.ResultsCompared with fully supervised learning, the pseudo-labelling method can improve the generalisability of the model and show better performance for unseen data from a different distribution using only 2% of labelled training samples. The accelerated GPU inference takes less than 1 millisecond per frame with FP16 precision.ConclusionOur approach demonstrates the potential of using pseudo-labelling strategies in real-time OCT segmentation tasks to guide robotic systems. Furthermore, the accelerated GPU inference of our network is highly promising for segmenting OCT images and guiding the position of a surgical tool (e.g. needle) for sub-retinal injections.

Landslide susceptibility prediction using slope unit-based machine learning models considering the heterogeneity of conditioning factors

To perform landslide susceptibility prediction (LSP), it is important to select appropriate mapping unit and landslide-related conditioning factors. The efficient and automatic multi-scale segmentation (MSS) method proposed by the authors promotes the application of slope units. However, LSP modeling based on these slope units has not been performed. Moreover, the heterogeneity of conditioning factors in slope units is neglected, leading to incomplete input variables of LSP modeling. In this study, the slope units extracted by the MSS method are used to construct LSP modeling, and the heterogeneity of conditioning factors is represented by the internal variations of conditioning factors within slope unit using the descriptive statistics features of mean, standard deviation and range. Thus, slope units-based machine learning models considering internal variations of conditioning factors (variant slope-machine learning) are proposed. The Chongyi County is selected as the case study and is divided into 53,055 slope units. Fifteen original slope unit-based conditioning factors are expanded to 38 slope unit-based conditioning factors through considering their internal variations. Random forest (RF) and multi-layer perceptron (MLP) machine learning models are used to construct variant Slope-RF and Slope-MLP models. Meanwhile, the Slope-RF and Slope-MLP models without considering the internal variations of conditioning factors, and conventional grid units-based machine learning (Grid-RF and MLP) models are built for comparisons through the LSP performance assessments. Results show that the variant Slope-machine learning models have higher LSP performances than Slope-machine learning models; LSP results of variant Slope-machine learning models have stronger directivity and practical application than Grid-machine learning models. It is concluded that slope units extracted by MSS method can be appropriate for LSP modeling, and the heterogeneity of conditioning factors within slope units can more comprehensively reflect the relationships between conditioning factors and landslides. The research results have important reference significance for land use and landslide prevention. ➢ Slope units extracted by multi-scale segmentation method are appropriate for LSP; ➢ Heterogeneity of conditioning factors are reflected by mean, range and standard deviation values; ➢ Variant Slope-machine learning models have higher LSP accuracy than Slope-machine learning models; ➢ Slope-machine learning models have stronger engineering practicability than Grid-machine learning models.

An early detection and segmentation of Brain Tumor using Deep Neural Network

BackgroundMagnetic resonance image (MRI) brain tumor segmentation is crucial and important in the medical field, which can help in diagnosis and prognosis, overall growth predictions, Tumor density measures, and care plans needed for patients. The difficulty in segmenting brain Tumors is primarily because of the wide range of structures, shapes, frequency, position, and visual appeal of Tumors, like intensity, contrast, and visual variation. With recent advancements in Deep Neural Networks (DNN) for image classification tasks, intelligent medical image segmentation is an exciting direction for Brain Tumor research. DNN requires a lot of time & processing capabilities to train because of only some gradient diffusion difficulty and its complication.MethodsTo overcome the gradient issue of DNN, this research work provides an efficient method for brain Tumor segmentation based on the Improved Residual Network (ResNet). Existing ResNet can be improved by maintaining the details of all the available connection links or by improving projection shortcuts. These details are fed to later phases, due to which improved ResNet achieves higher precision and can speed up the learning process.ResultsThe proposed improved Resnet address all three main components of existing ResNet: the flow of information through the network layers, the residual building block, and the projection shortcut. This approach minimizes computational costs and speeds up the process.ConclusionAn experimental analysis of the BRATS 2020 MRI sample data reveals that the proposed methodology achieves competitive performance over the traditional methods like CNN and Fully Convolution Neural Network (FCN) in more than 10% improved accuracy, recall, and f-measure.

EBUNet: a fast and accurate semantic segmentation network with lightweight efficient bottleneck unit

It has been difficult to achieve a suitable balance between effectiveness and efficiency in lightweight semantic segmentation networks in recent years. The goal of this work is to present an efficient and reliable semantic segmentation method called EBUNet, which is aimed at achieving a favorable trade-off between inference speed and prediction accuracy. Initially, we develop an Efficient Bottleneck Unit (EBU) that employs depth-wise convolution and depth-wise dilated convolution to obtain adequate features with moderate computation costs. Then, we developed a novel Image Partition Attention Module (IPAM), which divides feature maps into subregions and generates attention weights based on them. As a third step, we developed a novel lightweight attention decoder with which to retrieve spatial information effectively. Extensive experiments show that our EBUNet achieves 73.4% mIou and 152 FPS on the Cityscapes dataset and 72.2% mIoU and 147 FPS on the Camvid dataset with only 1.57 M parameters. The results of the experiment confirm that the proposed model is capable of making a decent trade-off in terms of accuracy, inference, and model size. The source code of our EBUNet is available at (https://github.com/Skybird1101/EBUNet).

An efficient image segmentation method based on expectation maximization and Salp swarm algorithm.

Multilevel image thresholding using Expectation Maximization (EM) is an efficient method for image segmentation. However, it has two weaknesses: 1) EM is a greedy algorithm and cannot jump out of local optima. 2) it cannot guarantee the number of required classes while estimating the histogram by Gaussian Mixture Models (GMM). in this paper, to overcome these shortages, a novel thresholding approach by combining EM and Salp Swarm Algorithm (SSA) is developed. SSA suggests potential points to the EM algorithm to fly to a better position. Moreover, a new mechanism is considered to maintain the number of desired clusters. Twenty-four medical test images are selected and examined by standard metrics such as PSNR and FSIM. The proposed method is compared with the traditional EM algorithm, and an average improvement of 5.27% in PSNR values and 2.01% in FSIM values were recorded. Also, the proposed approach is compared with four existing segmentation techniques by using CT scan images that Qatar University has collected. Experimental results depict that the proposed method obtains the first rank in terms of PSNR and the second rank in terms of FSIM. It has been observed that the proposed technique performs better performance in the segmentation result compared to other considered state-of-the-art methods.

Dual center validation of deep learning for automated multi-label segmentation of thoracic anatomy in bedside chest radiographs

Background and ObjectivesBedside chest radiographs (CXRs) are challenging to interpret but important for monitoring cardiothoracic disease and invasive therapy devices in critical care and emergency medicine. Taking surrounding anatomy into account is likely to improve the diagnostic accuracy of artificial intelligence and bring its performance closer to that of a radiologist. Therefore, we aimed to develop a deep convolutional neural network for efficient automatic anatomy segmentation of bedside CXRs. MethodsTo improve the efficiency of the segmentation process, we introduced a "human-in-the-loop" segmentation workflow with an active learning approach, looking at five major anatomical structures in the chest (heart, lungs, mediastinum, trachea, and clavicles). This allowed us to decrease the time needed for segmentation by 32% and select the most complex cases to utilize human expert annotators efficiently. After annotation of 2,000 CXRs from different Level 1 medical centers at Charité – University Hospital Berlin, there was no relevant improvement in model performance, and the annotation process was stopped. A 5-layer U-ResNet was trained for 150 epochs using a combined soft Dice similarity coefficient (DSC) and cross-entropy as a loss function. DSC, Jaccard index (JI), Hausdorff distance (HD) in mm, and average symmetric surface distance (ASSD) in mm were used to assess model performance. External validation was performed using an independent external test dataset from Aachen University Hospital (n = 20). ResultsThe final training, validation, and testing dataset consisted of 1900/50/50 segmentation masks for each anatomical structure. Our model achieved a mean DSC/JI/HD/ASSD of 0.93/0.88/32.1/5.8 for the lung, 0.92/0.86/21.65/4.85 for the mediastinum, 0.91/0.84/11.83/1.35 for the clavicles, 0.9/0.85/9.6/2.19 for the trachea, and 0.88/0.8/31.74/8.73 for the heart. Validation using the external dataset showed an overall robust performance of our algorithm. ConclusionsUsing an efficient computer-aided segmentation method with active learning, our anatomy-based model achieves comparable performance to state-of-the-art approaches. Instead of only segmenting the non-overlapping portions of the organs, as previous studies did, a closer approximation to actual anatomy is achieved by segmenting along the natural anatomical borders. This novel anatomy approach could be useful for developing pathology models for accurate and quantifiable diagnosis.