Retinal disorders, such as diabetic retinopathy, glaucoma, macular edema, and vein occlusions, are significant contributors to global vision impairment. These conditions frequently remain symptomless until patients suffer severe vision deterioration, underscoring the critical importance of early diagnosis. Fundus images serve as a valuable resource for identifying the initial indicators of these ailments, particularly by examining various characteristics of retinal blood vessels, such as their length, width, tortuosity, and branching patterns. Traditionally, healthcare practitioners often rely on manual retinal vessel segmentation, a process that is both time-consuming and intricate, demanding specialized expertise. However, this approach poses a notable challenge since its precision and consistency heavily rely on the availability of highly skilled professionals. To surmount these challenges, there is an urgent demand for an automatic and efficient method for retinal vessel segmentation and classification employing computer vision techniques, which form the foundation of biomedical imaging. Numerous researchers have put forth techniques for blood vessel segmentation, broadly categorized into machine learning, filtering-based, and model-based methods. Machine learning methods categorize pixels as either vessels or non-vessels, employing classifiers trained on hand-annotated images. Subsequently, these techniques extract features using 7D feature vectors and apply neural network classification. Additional post-processing steps are used to bridge gaps and eliminate isolated pixels. On the other hand, filtering-based approaches employ morphological operators within morphological image processing, capitalizing on predefined shapes to filter out objects from the background. However, this technique often treats larger blood vessels as cohesive structures. Model-based methods leverage vessel models to identify retinal blood vessels, but they are sensitive to parameter selection, necessitating careful choices to simultaneously detect thin and large vessels effectively. Our proposed research endeavors to conduct a thorough and empirical evaluation of the effectiveness of automated segmentation and classification techniques for identifying eye-related diseases, particularly diabetic retinopathy and glaucoma. This evaluation will involve various retinal image datasets, including DRIVE, REVIEW, STARE, HRF, and DRION. The methodologies under consideration encompass machine learning, filtering-based, and model-based approaches, with performance assessment based on a range of metrics, including true positive rate (TPR), true negative rate (TNR), positive predictive value (PPV), negative predictive value (NPV), false discovery rate (FDR), Matthews's correlation coefficient (MCC), and accuracy (ACC). The primary objective of this research is to scrutinize, assess, and compare the design and performance of different segmentation and classification techniques, encompassing both supervised and unsupervised learning methods. To attain this objective, we will refine existing techniques and develop new ones, ensuring a more streamlined and computationally efficient approach.

The destructive geohazard of landslides produces significant economic and environmental damages and social effects. State-of-the-art advances in landslide detection and monitoring are made possible through the integration of increased Earth Observation (EO) technologies and Deep Learning (DL) methods with traditional mapping methods. This assessment examines the EO and DL union for landslide detection by summarizing knowledge from more than 500 scholarly works. The research included examinations of studies that combined satellite remote sensing information, including Synthetic Aperture Radar (SAR) and multispectral imaging, with up-to-date Deep Learning models, particularly Convolutional Neural Networks (CNNs) and their U-Net versions. The research categorizes the examined studies into groups based on their methodological development, spatial extent, and validation techniques. Real-time EO data monitoring capabilities become more extensive through their use, but DL models perform automated feature recognition, which enhances accuracy in detection tasks. The research faces three critical problems: the deficiency of training data quantity for building stable models, the need to improve understanding of AI's predictions, and its capacity to function across diverse geographical landscapes. We introduce a combined approach that uses multi-source EO data alongside DL models incorporating physical laws to improve the evaluation and transferability between different platforms. Incorporating explainable AI (XAI) technology and active learning methods reduces the uninterpretable aspects of deep learning models, thereby improving the trustworthiness of automated landslide maps. The review highlights the need for a common agreement on datasets, benchmark standards, and interdisciplinary team efforts to advance the research topic. Research efforts in the future must combine semi-supervised learning approaches with synthetic data creation and real-time hazardous event predictions to optimise EO-DL framework deployments regarding landslide danger management. This study integrates EO and AI analysis methods to develop future landslide surveillance systems that aid in reducing disasters amid the current acceleration of climate change.

Brain tumors are a primary factor causing cancer-related deaths globally, and their classification remains a significant research challenge due to the variability in tumor intensity, size, and shape, as well as the similar appearances of different tumor types. Accurate differentiation is further complicated by these factors, making diagnosis difficult even with advanced imaging techniques such as magnetic resonance imaging (MRI). Recent techniques in artificial intelligence (AI), in particular deep learning (DL), have improved the speed and accuracy of medical image analysis, but they still face challenges like overfitting and the need for large annotated datasets. This study addresses these challenges by presenting two approaches for brain tumor classification using MRI images. The first approach involves fine-tuning transfer learning cutting-edge models, including SEResNet, ConvNeXtBase, and ResNet101V2, with global average pooling 2D and dropout layers to minimize overfitting and reduce the need for extensive preprocessing. The second approach leverages the Vision Transformer (ViT), optimized with the AdamW optimizer and extensive data augmentation. Experiments on the BT-Large-4C dataset demonstrate that SEResNet achieves the highest accuracy of 97.96%, surpassing ViT’s 95.4%. These results suggest that fine-tuning and transfer learning models are more effective at addressing the challenges of overfitting and dataset limitations, ultimately outperforming the Vision Transformer and existing state-of-the-art techniques in brain tumor classification.

Breast cancer was a prevalent form of cancer worldwide. Thermography, a method for diagnosing breast cancer, involves recording the thermal patterns of the breast. This article explores the use of a convolutional neural network (CNN) algorithm to extract features from a dataset of thermographic images. Initially, the CNN network was used to extract a feature vector from the images. Subsequently, machine learning techniques can be used for image classification. This study utilizes four classification methods, namely Fully connected neural network (FCnet), support vector machine (SVM), classification linear model (CLINEAR), and KNN, to classify breast cancer from thermographic images. The accuracy rates achieved by the FCnet, SVM, CLINEAR, and k-nearest neighbors (KNN) algorithms were 94.2%, 95.0%, 95.0%, and 94.1%, respectively. Furthermore, the reliability parameters for these classifiers were computed as 92.1%, 97.5%, 96.5%, and 91.2%, while their respective sensitivities were calculated as 95.5%, 94.1%, 90.4%, and 93.2%. These findings can assist experts in developing an expert system for breast cancer diagnosis.

Named Entity Recognition (NER), a core task in Information Extraction (IE) alongside Relation Extraction (RE), identifies and extracts entities like place and person names in various domains. NER has improved business processes in both public and private sectors but remains underutilized in government institutions, especially in developing countries like Indonesia. This study examines which government fields have utilized NER over the past five years, evaluates system performance, identifies common methods, highlights countries with significant adoption, and outlines current challenges. Over 64 international studies from 15 countries were selected using PRISMA 2020 guidelines. The findings are synthesized into a preliminary ontology design for Government NER.

Monitoring marine biodiversity is a challenge in some vulnerable and difficult-to-access habitats, such as underwater caves. Underwater caves are a great focus of biodiversity, concentrating a large number of species in their environment. However, most of the sessile species that live on the rocky walls are very vulnerable, and they are often threatened by different pressures. The use of these spaces as a destination for recreational divers can cause different impacts on the benthic habitat. In this work, we propose a methodology based on video recordings of cave walls and image analysis with deep learning algorithms to estimate the spatial density of structuring species in a study area. We propose a combination of automatic frame overlap detection, estimation of the actual extent of surface cover, and semantic segmentation of the main 10 species of corals and sponges to obtain species density maps. These maps can be the data source for monitoring biodiversity over time. In this paper, we analyzed the performance of three different semantic segmentation algorithms and backbones for this task and found that the Mask R-CNN model with the Xception101 backbone achieves the best accuracy, with an average segmentation accuracy of 82%.