Accurate segmentation of brain tumors in magnetic resonance imaging (MRI) remains a challenging task due to heterogeneous tumor structures, varying intensities across modalities, and limited annotated data. Deep learning has significantly advanced segmentation accuracy; however, it often suffers from sensitivity to hyperparameter settings and limited generalization. To overcome these challenges, bio-inspired metaheuristic algorithms have been increasingly employed to optimize various stages of the deep learning pipeline—including hyperparameter tuning, preprocessing, architectural design, and attention modulation. This review systematically examines developments from 2015 to 2025, focusing on the integration of nature-inspired optimization methods such as Particle Swarm Optimization (PSO), Genetic Algorithm (GA), Differential Evolution (DE), Grey Wolf Optimizer (GWO), Whale Optimization Algorithm (WOA), and novel hybrids including CJHBA and BioSwarmNet into deep learning-based brain tumor segmentation frameworks. A structured multi-query search strategy was executed using Publish or Perish across Google Scholar and Scopus databases. Following PRISMA guidelines, 3895 records were screened through automated filtering and manual eligibility checks, yielding a curated set of 106 primary studies. Through bibliometric mapping, methodological synthesis, and performance analysis, we highlight trends in algorithm usage, application domains (e.g., preprocessing, architecture search), and segmentation outcomes measured by metrics such as Dice Similarity Coefficient (DSC), Jaccard Index (JI), Hausdorff Distance (HD), and ASSD. Our findings demonstrate that bio-inspired optimization significantly enhances segmentation accuracy and robustness, particularly in multimodal settings involving FLAIR and T1CE modalities. The review concludes by identifying emerging research directions in hybrid optimization, real-time clinical applicability, and explainable AI, providing a roadmap for future exploration in this interdisciplinary domain.

This study explores the application of Convolutional Neural Networks (CNNs) for brain tumor segmentation, leveraging their ability to automatically extract hierarchical features from medical images. CNN architectures like U-Net, V-Net, and ResNet have shown significant promise in brain tumor classification, offering high precision in detecting tumor boundaries and classifying tumor types. Various benchmark datasets, such as BraTS, TCIA, Harvard, and Kaggle, provide annotated MRI images to evaluate these models. Performance metrics including Dice Similarity Coefficient (DSC), Intersection over Union, and accuracy are employed to assess the models’ effectiveness. The results demonstrate that CNN-based models, particularly U-Net, perform exceptionally well, with DSC scores exceeding 0.90 in most cases. However, challenges such as data imbalance, the need for large datasets, and high computational demands persist. Despite these limitations, CNNs, when combined with advanced techniques like transfer learning and data augmentation, offer robust solutions for brain tumor segmentation, showing promise for real-time clinical deployment. Further advancements are necessary to address generalization issues and enhance model efficiency, ensuring broader applicability in clinical settings.

Objective: This study aims to measure inter-religious harmony in the D.K.I Jakarta with the variables that become measuring instruments have three dimensions, namely tolerance, equality, and of cooperation. Theoretical framework: D.K.I Jakarta is considered a miniature barometer of religious harmony because of the plurality and heterogeneity of its people. One of the successes of developing religious harmony can be assessed by looking at the most basic indicators of harmony at the national and regional levels. Method/design/approach: in this study, the research was characterized as quantitative, descriptive and explanatory, with data collected through a survey. Villages were selected by multistage random sampling in 4 districts in D.K.I Jakarta Province with total sample of 400 respondents. Results and conclusion: The research findings show that the index of inter-religious harmony in the D.K.I Jakarta is 72.60, or in the high level of religious harmony category. For this reason, maintenance must continue improving to achieve a harmonious religious life. There are still sub-dimensions that need special attention, namely caring for the construction of other places of worship and being president or regional head regardless of religion. Research implications: Through this research, policymakers or other related agencies can use it to anticipate the emergence of conflicts. The results of this research can also be used as evaluation material to improve the quality of religious harmony in the future. Originality/value: in this study was to determine the level of religious harmony and obtain information on mapping the strengths and vulnerabilities of inter-religious relations in D.K.I Jakarta.

Penyakit malaria merupakan infeksi penyakit yang ditularkan oleh nyamuk baik dari manusia ataupun hewan lain melalui protozoa parasit yang biasa disebut plasmodium. Malaria bisa berakibat fatal jika terlambat dideteksi dikarenakan dapat mengakibatkan anemia akut, gagal ginjal, hingga berujung kematian. Infeksi malaria sebenarnya bisa dideteksi dini menggunakan sampel sel darah merah, dikarenakan sel darah merah yang terinfeksi parasit akan tampak pola atau bercak. Dalam penelitian ini menggunakan data set publik berupa gambar sel darah merah yang terinfeksi protozoa parasit dan tidak terinfeksi. Beberapa peneliti sudah melakukan penelitian untuk mengklasifikasikan penyakit malaria dari sampel sel darah merah. Pada umumnya peneliti sebelumnya menggunakan metode image processing dengan mengubah gambar menjadi negatif untuk kemudian di klasifikasikan menggunakan metode klasifikasi tertentu. Di dalam penelitian ini penulis mencoba mengembangkan metode lain yaitu dengan melakukan pre-prosessing terlebih dahulu terhadap data set yang ada yaitu dengan mengubah ukuran gambar menjadi ukuran PIXEL 100x100, untuk kemudian dilakukan image augmentation untuk memperbanyak data set sehingga kemungkinan akurasi naik menjadi lebih tinggi. Proses selanjutnya yaitu mengubah dimensi gambar yang semula berdimensi tiga menjadi 1 dimensi dengan proses fitur reduksi menggunakan PCA (Principle Component Analysis) hasil dari fitur reduksi tersebut selanjutnya diklasifikasikan menggunakan metode Convolutional Neural Network (CNN). Hasil dari penelitian menunjukkan peningkatan akurasi yang lebih baik menjadi 98,30% dengan menggunakan metode tersebut.Kata kunci: image processing, CNN Classifier, Malaria, image augmentation, PCA

In the envisioned landscape of medical imaging in 2044, this research explores the integration of advanced AI techniques, specifically reinforcement learning (RL) and meta-learning, to address persistent challenges in disease diagnosis and treatment planning. Leveraging vast amounts of imaging data, deep learning models have demonstrated significant advancements in tasks such as tumor detection and organ segmentation. However, existing approaches often face limitations in adapting to evolving patient characteristics and data scarcity. By incorporating principles from RL and meta-learning, this study aims to develop dynamic, adaptive AI systems capable of optimizing imaging protocols, enhancing diagnostic accuracy, and personalizing treatment strategies for individual patients. The research conducts a comprehensive review of existing literature on RL and meta-learning in healthcare proposes novel methodologies for integrating these techniques into medical imaging workflows, and evaluates their efficacy through empirical studies and clinical validation. The ultimate goal is to contribute to the advancement of medical imaging technologies, paving the way for more personalized and efficient healthcare solutions in the future

Heart failure is a disease that is one of the most crucial in the world. Researchers have used several machine learning techniques to assist health professionals in the diagnosis of heart failure. K-NN is a technique of supervised learning algorithm that has been successfully used in terms of classification. However, using the K-NN algorithm has stages in terms of data analysis. The data used must also be processed in such a way that it becomes data that is easier to analyse and that the results obtained are also more accurate. Data pre-processing involves transforming raw data into a format that is appropriate for the model. The normalization technique is one of the techniques contained in pre-processing. This research uses two normalization techniques, namely the simple feature scale and min-max. The purpose of this study is to compare the performance of the KNN model to obtain an optimal prediction model. This study contributes to producing a heart failure prediction model based on the K-Nearest Neighbors (KNN) algorithm that can be optimized to improve the accuracy of early detection, so that it can help medical personnel in making more appropriate clinical decisions. The results obtained from this research show that the dataset that uses the min-max normalization method is better than data that is not normalized and data that uses simple feature scale normalization. The highest level of accuracy was achieved by employing the min-max normalisation technique, with a value of K=9, resulting in an accuracy rate of 85.05%.