Automatic Corneal Ulcer Segmentation Combining Gaussian Mixture Modeling and Otsu Method.

Manual segmentation in lumbar paraspinal muscle studies is time-consuming, subject to variability, and limits comprehensive evaluation. To solve the problems, we aim to develop a deep learning algorithm for the automatic segmentation and multi-parameter quantification of eight paraspinal muscles in lumbar CT. We collected CT images covering the range from the superior endplate of the L1 vertebra to the inferior endplate of the S1 vertebra. The dataset was partitioned into training, validation, and test sets in a ratio of 7:1:2. Six convolutional neural networks were employed to automatically segment the bilateral psoas major, quadratus lumborum, erector spinae, and multifidus across all slices. Model performance was evaluated using the Dice Similarity Coefficient (DSC), Hausdorff Distance (HD), and mean Intersection over Union (mIoU). Muscle parameters including cross-sectional area, muscle volume, fat infiltration, CT density, and paraspinal muscle index were calculated. The Intraclass Correlation Coefficient [ICC(3,1)] was used to assess agreement between model's segmentations and manual ground truth. 100 lumbar CT scans (age, 55.02 ± 16.2; 62 female) were used for model development. The mean DSC across all six models ranged from 0.754 to 0.903. TransUNet demonstrated the best overall (DSC, 0.903; 95% CI: 0.902, 0.913) (HD, 9.934mm; 95% CI: 9.815, 11.902) (mIoU, 0.841; 95% CI: 0.839, 0.850). For individual muscles segmented by TransUNet, the highest DSC was 0.936 (95% CI: 0.934-0.938) for the right multifidus. The mean ICC for all parameters was 0.931. The highest ICC value was 0.999 (p < 0.001). The deep learning tool enables automatic and accurate segmentation of eight lumbar paraspinal muscles. Muscle parameter measurements derived from model's segmentations and manual ground truth exhibit high agreement. Accurate segmentation and quantification of paraspinal muscle parameters facilitates large-scale epidemiological studies of paraspinal muscles and spine-related diseases, overcoming the limitations in manual methods.

Automatic segmentation of the choroid on optical coherence tomography (OCT) images helps ophthalmologists in diagnosing eye pathologies. Compared to manual segmentations, it is faster and is not affected by human errors. The presence of the large speckle noise in the OCT images limits the automatic segmentation and interpretation of them. To solve this problem, a new curvelet transform-based K-SVD method is proposed in this study. Furthermore, the dataset was manually segmented by a retinal ophthalmologist to draw a comparison with the proposed automatic segmentation technique. In this study, curvelet transform-based K-SVD dictionary learning and Lucy-Richardson algorithm were used to remove the speckle noise from OCT images. The Outer/Inner Choroidal Boundaries (O/ICB) were determined utilizing graph theory. The area between ICB and outer choroidal boundary was considered as the choroidal region. The proposed method was evaluated on our dataset and the average dice similarity coefficient (DSC) was calculated to be 92.14% ± 3.30% between automatic and manual segmented regions. Moreover, by applying the latest presented open-source algorithm by Mazzaferri et al. on our dataset, the mean DSC was calculated to be 55.75% ± 14.54%. A significant similarity was observed between automatic and manual segmentations. Automatic segmentation of the choroidal layer could be also utilized in large-scale quantitative studies of the choroid.

Cast shadow affects computer vision tasks such as image segmentation, object detection and tracking since objects and shadows share the same visual motion characteristics. This unavoidable problem decreases video surveillance system performance. The basic idea of this paper is to exploit the evidence that shadows darken the surface which they are cast upon. For this reason, we propose a simple and accurate method for detection of moving cast shadows based on chromatic properties in RGB, HSV and YUV color spaces. The method requires no a priori assumptions regarding the scene or lighting source. Starting from a normalization step, we apply canny filter to detect the boundary between self-shadow and cast shadow. This treatment is devoted only for the first sequence. Then, we separate between background and moving objects using an improved version of Gaussian mixture model. In order to remove these unwanted shadows completely, we use three change estimators calculated according to the intensity ratio in HSV color space, chromaticity properties in RGB color space, and brightness ratio in YUV color space. Only pixels that satisfy threshold of the three estimators are labeled as shadow and will be removed. Experiments carried out on various video databases prove that the proposed system is robust and efficient and can precisely remove shadows for a wide class of environment and without any assumptions. Experimental results also show that our approach outperforms existing methods and can run in real-time systems.

Quantification of diffusion restriction lesions in sporadic Creutzfeldt-Jakob disease (sCJD) may provide information of the disease burden. We aim to develop an automatic segmentation model for sCJD and to evaluate the volume of disease extent as a prognostic marker for overall survival. Fifty-six patients (mean age ± SD, 61.2 ± 9.9 years) were included from February 2000 to July 2020. A threshold-based segmentation was used to obtain abnormal signal intensity masks. Segmented volumes were compared with the visual grade. The Dice similarity coefficient was calculated to measure the similarity between the automatic vs. manual segmentation. Cox proportional hazards regression analysis was performed to evaluate the volume of disease extent as a prognostic marker. The automatic segmentation showed good correlation with the visual grading. The cortical lesion volumes significantly increased as the visual grade aggravated (extensive: 112.9 ± 73.2; moderate: 45.4 ± 30.4; minimal involvement: 29.6 ± 18.1 mm3) (P < 0.001). The deep gray matter lesion volumes were significantly higher for positive than for negative involvement of the deep gray matter (5.6 ± 4.6 mm3 vs. 1.0 ± 1.3 mm3, P < 0.001). The mean Dice similarity coefficients were 0.90 and 0.94 for cortical and deep gray matter lesions, respectively. However, the volume of disease extent was not associated with worse overall survival (cortical extent: P = 0.07; deep gray matter extent: P = 0.12).

Attention-guided multi-scale learning network for automatic prostate and tumor segmentation on MRI

Separate renal function assessment is important in clinical decision making. The single-photon emission computed tomography is commonly used for the assessment although radioactive, tedious and of high cost. This study aimed to automatically assess the separate renal function using plain CT images and artificial intelligence methods, including deep learning-based automatic segmentation and radiomics modeling. We performed a retrospective study on 281 patients with nephrarctia or hydronephrosis from two centers (Training set: 159 patients from Center I; Test set: 122 patients from Center II). The renal parenchyma and hydronephrosis regions in plain CT images were automatically segmented using deep learning-based U-Net transformers (UNETR). Radiomic features were extracted from the two regions and used to build radiomic signature using the ElasticNet, then further combined with clinical characteristics using multivariable logistic regression to obtain an integrated model. The automatic segmentation was evaluated using the dice similarity coefficient (DSC). The mean DSC of automatic kidney segmentation based on UNETR was 0.894 and 0.881 in the training and test sets. The average time of automatic and manual segmentation was 3.4s/case and 1477.9s/case. The AUC of radiomic signature was 0.778 in the training set and 0.801 in the test set. The AUC of the integrated model was 0.792 and 0.825 in the training and test sets. It is feasible to assess the renal function of each kidney separately using plain CT and AI methods. Our method can minimize the radiation risk, improve the diagnostic efficiency and reduce the costs.

To explore the feasibility of using the bidirectional local distance based medical similarity index (MSI) to evaluate automatic segmentation on medical images. Taking the intermediate risk clinical target volume for nasopharyngeal carcinoma manually segmented by an experience radiation oncologist as region of interest, using Atlas-based and deep-learning-based methods to obtain automatic segmentation respectively, and calculated multiple MSI and Dice similarity coefficient (DSC) between manual segmentation and automatic segmentation. Then the difference between MSI and DSC was comparatively analyzed. DSC values for Atlas-based and deep-learning-based automatic segmentation were 0.73 and 0.84 respectively. MSI values for them varied between 0.29~0.78 and 0.44~0.91 under different inside-outside-level. It is feasible to use MSI to evaluate the results of automatic segmentation. By setting the penalty coefficient, it can reflect phenomena such as under-delineation and over-delineation, and improve the sensitivity of medical image contour similarity evaluation.

Combinations of data augmentation methods and deep learning architectures for automatic pancreas segmentation on CT images are proposed and evaluated. Images from a public CT dataset of pancreas segmentation were used to evaluate the models. Baseline U-net and deep U-net were chosen for the deep learning models of pancreas segmentation. Methods of data augmentation included conventional methods, mixup, and random image cropping and patching (RICAP). Ten combinations of the deep learning models and the data augmentation methods were evaluated. Four-fold cross validation was performed to train and evaluate these models with data augmentation methods. The dice similarity coefficient (DSC) was calculated between automatic segmentation results and manually annotated labels and these were visually assessed by two radiologists. The performance of the deep U-net was better than that of the baseline U-net with mean DSC of 0.703–0.789 and 0.686–0.748, respectively. In both baseline U-net and deep U-net, the methods with data augmentation performed better than methods with no data augmentation, and mixup and RICAP were more useful than the conventional method. The best mean DSC was obtained using a combination of deep U-net, mixup, and RICAP, and the two radiologists scored the results from this model as good or perfect in 76 and 74 of the 82 cases.

Heterogeneous intensity distribution inside the prostate gland, significant variations in prostate shape, size, inter dataset contrast variations, and imaging artifacts like shadow regions and speckle in Trans Rectal Ultrasound (TRUS) images challenge computer aided automatic or semi-automatic segmentation of the prostate. In this paper, we propose a supervised learning schema based on random forest for automatic initialization and propagation of statistical shape and appearance model. Parametric representation of the statistical model of shape and appearance is derived from principal component analysis (PCA) of the probability distribution inside the prostate and PCA of the contour landmarks obtained from the training images. Unlike traditional statistical models of shape and intensity priors, the appearance model in this paper is derived from the posterior probabilities obtained from random forest classification. This probabilistic information is then used for the initialization and propagation of the statistical model. The proposed method achieves mean Dice Similarity Coefficient (DSC) value of 0.96±0.01, with a mean segmentation time of 0.67±0.02 seconds when validated with 24 images from 6 datasets with considerable shape, size, and intensity variations, in a leave-one-patient-out validation framework. The model achieves statistically significant t-test p-value<0.0001 in mean DSC and mean mean absolute distance (MAD) values compared to traditional statistical models of shape and intensity priors.

Retinal vessel segmentation is part of the morphological extraction of retinal blood vessels that plays an essential role in medical image processing. Manual segmentation is possible to do, but it is time-consuming and requires special operators. Moreover, the possibility of variability between operators is vast. This study aims to answer the shortcomings of the manual segmentation process by automatically segmenting retinal blood vessels. The main contribution of this study is the use of a simple method to iteratively segment retinal blood vessels. All processes in the segmentation are simulated using Matlab. The algorithm was evaluated by comparing the results of the automatic segmentation with 20 manually segmented images from the STARE dataset. The result show specificity 98.13%, accuracy 93.60%, sensitivity 56.42%, precision 80.48%, and the dice coefficient 64.06%. In conclusion, the automatic retinal blood vessel image segmentation process worked well.

Purpose:To investigate the feasibility of using geodesic curves to track fuzzy borders between liver and chest wall and to evaluate how it could improve the performance of automatic liver segmentation on treatment planning CT images.Methods:The performance of automatic liver segmentation usually suffers from the fuzzy borders between liver and the adjacent chest wall due to similar HU values on non‐contrast‐enhanced planning CT images. To address this issue, geodesic curves were used to track these fuzzy borders. We first constructed a horizontal gradient map on the coronal‐view images. After automatically identifying one starting and one ending point, a minimal distance map (MDM) was constructed by evolving a front starting from an infinitesimal circle around the ending point to every pixel on the image. The value of each point in MDM represents the minimal distance from the point to the ending point, which combines both arc length and gradient magnitude along the path. The front evolution was numerically solved by fast‐marching method. We coupled this method with our automatic liver segmentation method, in which an initial contour was firstly estimated by adaptive thresholding and then a distance‐regularized geodesic active contour model was used to further refine the contour. Besides visual assessment, dice similarity coefficient (DSC) was calculated to quantitatively compare the computer‐generated contours with manual outlines.Results:This study included five patients with 597 CT slices who received SBRT liver treatment. We observed clear separation between liver and chest wall after delineation. The results of automatic segmentation were in excellent agreement with manual outlines, yielding all DSCs higher than 0.9 with mean of 0.93.Conclusion:The preliminary results demonstrate that geodesic curve can be used to track fuzzy borders between liver and chest wall, and thus to improve the performance of automatic liver segmentation on planning CT images.

Skin cancer is the most common type of cancer, a life-threatening condition that leads to serious health problems if not detected early, and its incidence is increasing worldwide. In recent years, computer vision and decision support systems have been used for disease detection in dermoscopic images. Furthermore, it has been observed that data representation methods affect the detection performance of these models, and the effect of color information on transformer-based models has not been sufficiently investigated. This study used the International Skin Imaging Collaboration (ISIC) 2017 dataset consisting of RGB images, and these images were converted into the HSV, LAB, and YCbCr color spaces. Transformer-based models, including visual transformer (ViT), swin transformer, data efficient image transformer (DeiT), and label-free self-distillation (DINO), were used for benign and malignant classification. According to the classification performance results, RGB and HSV color spaces particularly in the DeiT and Swin models, stable and high accuracy values ​​were obtained. It was observed that the ViT and DINO models were more sensitive to color space transformations and achieved lower classification performance compared to other models. The highest performance in skin cancer classification was achieved with the DeiT model trained in the RGB color space, with the highest accuracy (0.7668). Furthermore, the explainability-based gradient-weighted class activation mapping (Grad-CAM) method was used to analyze where the models focused in image regions when making classification decisions. This study shows the effect and usability of color space transformations in transformer-based models for skin cancer classification and offers a comparative contribution to the literature.

Intraoperative ultrasound (iUS) is increasingly used in pediatric brain tumor surgery. However, accurate tumor tissue identification and localization within the patient with iUS remains challenging. Automatic tumor segmentation may overcome these difficulties by providing quicker, more consistent, and reliable image analysis. This study evaluated the feasibility of automatic 3-dimensional segmentation of pediatric brain tumors using navigated iUS. A retrospective analysis was conducted on 109 navigated iUS acquisitions from 60 pediatric patients who underwent neurosurgical tumor resection. 3-dimensional nnU-Net was trained on 79 preresectional iUS acquisitions with corresponding tumor segmentations. Evaluation was performed on 30 iUS acquisitions against segmentations validated by 2 neurosurgeons. Performance benchmarks were set at a minimum Dice Similarity Coefficient (DSC) of 0.85 and a maximum mean surface distance and 95th percentile Hausdorff distance of 5 mm. The trained algorithm predicted tumor segmentations for all acquisitions in the test set, which met the predefined benchmark values with a median DSC of 0.89 (IQR: 0.13), a median mean surface distance of 1.1 mm (IQR: 1.0 mm), and a median 95th percentile Hausdorff distance of 5.6 mm (IQR: 3.5 mm). The lowest quartile of DSC values consisted of 5 patients, including craniopharyngiomas (n = 3) and high-grade gliomas (n = 2). Oversegmentation of the lateral ventricle as tumor was observed in 2 patients. In 1 high-grade glioma case, edema was misclassified as tumor because of unclear tumor boundaries. This study demonstrates that automatic segmentation of pediatric brain tumors using navigated iUS is feasible and achieved high performance in most cases, with few outliers that showed moderate performance in tumor prediction. Future efforts will focus on real-time segmentation of residual tumor tissue during tumor resection. Ultimately, real-time visualization of automatically segmented tumors in 3 dimensions could improve intraoperative navigation and support neurosurgical decision making.

Malaria is a disease in human caused by Plasmodium parasite. Misdiagnosis and improperness in medical treatment towards this disease can lead to the death of patients. A number of digital image analysis-based research works to minimize the misdiagnosis has been developed. This research aims to segment the form of parasitic cells of Plasmodium using the combination of the red (R) channel in the RGB color space and saturation (S) channel in the HSV color space. The images used were parasitic cells of Plasmodium vivax, consisting of the phases of trophozoite, schizonts, and gametocytes. Each of which comprises ten samples. The process was begun by improving the image quality continued by separating the red channel in the RGB color space and by converting the image of RGB into the HSV image to obtain the saturation (S) channel. These two color channels were then combined in the form of a grayscale image. The further process was conducted by changing those grayscale image into the binary image using the Otsu method and the morphological operation to form it as the masking image. Of 30 tested images, the segmentation of the red (R) channel and saturation (S) channel combination achieved the accuracy of 93.33%, while the segmentation using saturation (S) channel produces result in the accuracy of 66.67%.

This paper proposes novel algorithms for color based segmentation and shape based matching of forest fires from video frames. Fire pixels from the sample images are collected and compose 3D point cloud in RGB color space, based on which Gaussian mixture model is used to build a more precise color model. According to new definition of flame pattern, possible fire region is further segmented with the help of HSV color space. The segmented results are labeled and the ratios of different color pixels are calculated. The candidate fire regions among consecutive frames are then matched using our defined overlapping degree and varying degree of two regions. From the matched candidate fire regions, dynamic features are presented including area, roundness, contour and color distribution, which can be taken as useful features for the recognition of forest flames. Our work has been tested with experiments and proved very valuable to predict forest fire and protect forest resources.