Papers by Keyword: Segmentation

Abstract: This paper presents the outcomes of the project for applications of 3D printing technology in healthcare. The anatomical 3D modelling was carried out, and the 3D digital anatomical models were developed from the CT scan medical images, through medical images acquisition, segmentation, preparation, 3D printing and post-processing processes. Furthermore, the 3D digital models were converted into 3D physical models through Fused Deposition Modeling (FDM) and Stereolithography (SLA) 3D printing technologies. The segmentation and preparation processes were performed by employing 3D Slicer V5.8.0 and Meshmixer Autodesk V3.5software, respectively. The 3D digital models were prepared for printing using GrabCAD print V1.88 and Preform V3.36.0 software for FDM and SLA 3D printing technologies, respectively. During the models’ printing preparations, the printing parameters’ settings were performed, and the G-Codes were generated, which then sent to the printers. The printed models are to be used for training and research at University of Namibia. In addition to manual segmentation, AI-based segmentation which is an automated segmentation was also performed, and the models generated from the two segmentation methods were compared.

Abstract: The construction of Yogyakarta International Airport (YIA), driven by the limitations of land use at the former Adisutjipto Airport, is a key focal point of this research. The development of YIA has led to significant land cover changes in 2016 to 2021, transforming predominantly agricultural and fisheries land into built-up areas. Numerous studies have utilized remote sensing data to analyze land cover and land use (LULC) changes in Kulonprogo Regency, applying a range of remote sensing analytical methods. The most substantial LULC changes have been observed in Temon District, where agricultural land has sharply decreased, coinciding with the expansion of built-up areas. This study aims to further examine land use changes in Temon District, employing object-based classification techniques to enhance the accuracy of land cover analysis. In this study, OBIA classified land cover with 80% accuracy for the 2016 image (scale 100, shape 0.7, compactness 0.7) and 86% for the 2021 image (scale 100, shape 0.9, compactness 0.2). The most significant change was a 537-hectare reduction in paddy field.

Abstract: In the recent era- very frequently people come across health issues due to consumption of poor-quality food items- which leads to issues such as food poisoning, vomiting, diarrhea, etc., For a full development of fruits and vegetables, all the nutrients are necessary during its growth. But due to circumstances like soil defects, infections, water scarcity, waterlogging, etc., the vegetables & fruits gets infected with some diseases. So there arises a necessity of a system which inspects for any presence of disease in fruits & vegetables, with reduced manual intervention. This paper provides a detailed overview of a system developed using the Python programming language. Its aim is to recognize and classify various fruits and vegetables, while also identifying any diseases affecting them and determining the specific type of infection. In order to recognize the details accurately, the system is designed to use convolutional neural networks (CNN) and the results are displayed using computer vision techniques. The analysis, implementation, and future improvements of the proposed system are briefed in this paper. For this, we have used Anaconda navigator software (Jupyter notebook, IDLE).

Abstract: Automated retinal vascular network detection and analysis using digital retinal images continue to play a major role in the field of biomedicine for the diagnosis and management of various forms of human ailments like hypertension, diabetic retinopathy, retinopathy of prematurity, glaucoma and cardiovascular diseases. Although several literature have implemented different automatic approaches of detecting blood vessels in the retinal and also determining their tortuous states, the results obtained show that there are needs for further investigation on more efficient ways to detect and characterize the blood vessel network tortuosity states. This paper implements the use of an adaptive thresholding method based on local spatial relational variance (LSRV) for the detection of the retinal vascular networks. The suitability of a multi-layer perceptron artificial neural network (MLP-ANN) technique for the tortuosity characterization of retinal blood vascular networks is also presented in this paper. Some vessel geometric features of detected vessels are fed into ANN classifier for the automatic classification of the retinal vascular networks as being tortuous vessels or normal vessels. Experimental studies conducted on DRIVE and STARE databases show that the vascular network detection results obtained from the method implemented in this paper detects large and thin vascular networks in the retina. In comparison to preious methods in the literature, the proposed method for vascular network segmentation achieved better performance than several methods in the literature with a mean accuracy value of 95.04% and mean sensitivity value of 75.16% on DRIVE and mean accuracy value of 94.02% and average sensitivity value of 76.55% on STARE with computational processing time of 4.5 seconds and 9.4 seconds on DRIVE and STARE respectively. The MLP-ANN method proposed for the vascular network tortuosity characterization achieves promising accuracy rates of 77.5%, 80%, 83.33%, 85%, 86.67% and 100% for varying training sample sizes.

Abstract: The fourth most common form of cancer among women is cervical cancer with 569,847new cases and 311,365 reported deaths worldwide in 2018. Cervical cancer is classified as the third leading cause of cancer among women in Malaysia, with approximately 1,682 new cervical cases and about 944 deaths occurred in 2018. Cervical cancer can be detected early by cervical cancer screening. Papanicolaou test, also known as Pap smear test is conducted to detect cancer or precancer in the cervix. The disadvantage of this conventional method is that the sample of microscopic images will risk blurring effects, noise, shadow, lighting and artefact problems. The diagnostic microscopic observation performed by a microbiologist is normally time-consuming and may produce inaccurate results even by experienced hands. Thus, correct diagnosis information is essential to assist physicians to analyze the condition of the patients. In this study, an automatedsegmentation system is proposed to be used as it is more accurate and faster compared to the conventional technique. Using the proposed method in this paper, the image was enhanced by applying a median filter and Partial Contrast Stretching. A segmentation method based on mathematical morphology was performed to segment the nucleus in the Pap smear images. Image Quality Assessment (IQA) which measures the accuracy, sensitivity and specificity were used to prove the effectiveness of the proposed method. The results of the numerical simulation indicate that the proposed method shows a higher percentage of accuracy and specificity with 93.66% and 95.54% respectively compared to Otsu, Niblack and Wolf methods. As a conclusion, the percentage of sensitivity is slightly lower, with 89.20% compared to Otsu and Wolf methods. The results presented here may facilitate improvements in the detection performance in comparison to the existing methods.

Abstract: Breast cancer occurs as a result of erratic growth and proliferation cells that originate in the breast. In this paper, the classifiers were used to identify the abnormalities on mammograms to get the region of interest (ROI). Before classifier based segmentation, noise, pectoral muscles, and tags were removed for a successful segmentation process. Then the proposed approach extracted the brightest regions using modified k-means. From the extracted brightest regions, shape and texture features were extracted and given to classifiers (KNN and SVM) and marked as ROI only those non-overlapping abnormal regions. The ROIs obtained using the proposed classifier-based segmentation algorithm was compared with the ground truth annotated by the radiologists. The datasets used to evaluate the performance of the proposed algorithm was public (MIAS) and local datasets (BGH and DADC).

Abstract: In this paper, a modified adaptive K-means (MAKM) method is proposed to extract the region of interest (ROI) from the local and public datasets. The local image datasets are collected from Bethezata General Hospital (BGH) and the public datasets are from Mammographic Image Analysis Society (MIAS). The same image number is used for both datasets, 112 are abnormal and 208 are normal. Two texture features (GLCM and Gabor) from ROIs and one CNN based extracted features are considered in the experiment. CNN features are extracted using Inception-V3 pre-trained model after simple preprocessing and cropping. The quality of the features are evaluated individually and by fusing features to one another and five classifiers (SVM, KNN, MLP, RF, and NB) are used to measure the descriptive power of the features using cross-validation. The proposed approach was first evaluated on the local dataset and then applied to the public dataset. The results of the classifiers are measured using accuracy, sensitivity, specificity, kappa, computation time and AUC. The experimental analysis made using GLCM features from the two datasets indicates that GLCM features from BGH dataset outperformed that of MIAS dataset in all five classifiers. However, Gabor features from the two datasets scored the best result with two classifiers (SVM and MLP). For BGH and MIAS, SVM scored an accuracy of 99%, 97.46%, the sensitivity of 99.48%, 96.26% and specificity of 98.16%, 100% respectively. And MLP achieved an accuracy of 97%, 87.64%, the sensitivity of 97.40%, 96.65% and specificity of 96.26%, 75.73% respectively. Relatively maximum performance is achieved for feature fusion between Gabor and CNN based extracted features using MLP classifier. However, KNN, MLP, RF, and NB classifiers achieved almost 100% performance for GLCM texture features and SVM scored an accuracy of 96.88%, the sensitivity of 97.14% and specificity of 96.36%. As compared to other classifiers, NB has scored the least computation time in all experiments.

Abstract: Segmentation of blood vessels in the retinal is a crucial step in the diagnosis of eye diseases such as diabetic retinopathy and glaucoma. This paper presents a supervised method for automatic segmentation of blood vessels in retinal images. The proposed method based on a hybrid combination between Gray-Level and Moment Invariant techniques. There are four steps involved, whereas preprocessing, feature extraction, classification, and post-processing. In the preprocessing, three stages are performed include vessel central light reflex removal, background homogenization, and vessel enhancement. The 7-D vector feature extraction was performed to compute that compose of gray-level and moment invariants-based features for pixel representation. The decision tree is used for classification step that characterized the pixel based on vessels and non-vessels. The final step is the post-processing which will remove the small artifacts appears after classification process. The proposed method was compared to the Vascular Tree method and Morphological method. Based on the objective evaluation, the proposed method achieved (sensitivity = 98.589, specificity = 55.544 and accuracy = 96.197).

Abstract: Measurement of surface deformation is a key component for mapping intraoperative and preoperative image data in image-guided surgery. In this study, we segment CT-scanned images and then use a coherent point drift algorithm for the estimation of surface deformation. To extract surface points, the segmentation is based on the intensity of the image data. The registration of two point sets is considered as a probability density estimation problem in an expectation-maximization framework. Experimental results show that surface deformation between two point sets can be obtained based on the obtained geometric transformation.

Abstract: Comparison of quantitative methods for the segmentation of nuclei images is performed. A sizeable variability is typical for biological specimens and it induces the elemental uncertainty in experimental data. To remove the variability from comparison we have proposed and built the special test objects of two types simulating the nuclei images. I-type objects are test patterns as sets of circles and squares of specified dimensions in chromium films on a glass plate. II-type objects are images with brightness differentials that simulate diffractive blurring and are built up with MathCad programming environment. Test objects enable objective comparison of characteristics obtained in the course of their quantitative optical-and-structural analysis using various algorithms and programs. We found that the efficiency of the segmentation algorithm depends on diffractive blurring of the image. Specifics of Otsu’s algorithm and local algorithm of brightness gradient are analyzed for finding the segmentation threshold of digital images modeling transmission Feulgen-stained cell nuclei specimens with diffractive blurring. The performed calculations revealed that the border of geometrooptical image practically coincide with the points of inflection on the intensity distribution graph in a test-object image space. Computational experiments show that quantitative results of the morphometric image study defined by the various segmentation algorithms vary within 5%. It is established that the threshold-identifying algorithm based on the brightness gradient is preferable in the image cytometry.