The procedure to diagnose anemia is time-consuming and resource-intensive due to the existence of a multitude of symptoms that can be felt physically or seen visually. Anemia also has several forms, which can be distinguished based on several characteristics. It is possible to diagnose anemia through a quick, affordable, and easily accessible laboratory test known as the complete blood count (CBC), but the method cannot directly identify different kinds of anemia. Therefore, further tests are required to establish a gold standard for the type of anemia in a patient. These tests are uncommon in settings that offer healthcare on a smaller scale because they require expensive equipment. Moreover, it is also difficult to discern between beta thalassemia trait (BTT), iron deficiency anemia (IDA), hemoglobin E (HbE), and combination anemias despite the presence of multiple red blood cell (RBC) formulas and indices with differing optimal cutoff values. This is due to the existence of several varieties of anemia in individuals, making it difficult to distinguish between BTT, IDA, HbE, and combinations. Therefore, a more precise and automated prediction model is proposed to distinguish these four types to accelerate the identification process for doctors. Historical data were retrieved from the Laboratory of the Department of Clinical Pathology and Laboratory Medicine, Faculty of Medicine, Public Health, and Nursing, Universitas Gadjah Mada, Yogyakarta, Indonesia for this purpose. Furthermore, the model was developed using the algorithm for the extreme learning machine (ELM). This was followed by the measurement of the performance using the confusion matrix and 190 data representing the four classes, and the results showed 99.21% accuracy, 98.44% sensitivity, 99.30% precision, and an F1 score of 98.84%.

Anemia is a widespread worldwide health problem that has a substantial effect on groups who are particularly susceptible. The objective of this work is to improve the diagnosis of anemia by creating a hybrid machine learning model called SMOTE-MRS. This model combines SMOTE, K-Means Clustering, and Random Over Sampling techniques. The model aims to enhance diagnosis accuracy and ultimately improve healthcare outcomes, specifically in Indonesia, by resolving the imbalance in the dataset. The SMOTE-MRS model mitigates the issue of imbalanced datasets by combining the techniques of SMOTE, K-Means clustering, and random oversampling. K-Means clustering first divides the dataset into K groups. SMOTE then produces artificial instances of the underrepresented class inside every cluster. Random oversampling is a technique that replicates instances of the minority class to make the dataset balanced and enhance the training of machine learning models. The research evaluated the performance of Random Forest, Naive Bayes, and Support Vector Machine models in predicting anemia, using the SMOTE-MRS technique. The SMOTE-MRS algorithm exhibited outstanding performance, achieving scores of 0.973 for Accuracy, 0.990 for Recall, 0.968 for Precision, 0.979 for F1-Score, and 0.994 for AUC. These findings demonstrate its exceptional capacity to effectively handle class imbalance and enhance prediction accuracy. The SMOTE-MRS algorithm has outstanding performance on a wide range of datasets, such as those related to anemia, diabetes, breast cancer, and kidney failure datasets. This showcases its robustness and versatility in many predictive modeling scenarios. The findings demonstrate that SMOTE-MRS outperforms standard approaches such as SMOTE, SMOTE-ENC, and ROS in effectively addressing class imbalance. The research affirms the better performance of SMOTE-MRS in addressing class imbalance for the prediction of anemia. SMOTE-MRS surpasses traditional approaches, attaining exceptional results in all measures. Future studies should prioritize the optimization of SMOTE-MRS to minimize small performance declines, conduct a comparative analysis with other approaches, and validate its adaptability across various domains. Furthermore, it is advisable to include SMOTE-MRS in deep learning models and real-time applications.

This study addresses the necessity for more advanced diagnostic tools in managing diabetes, a chronic metabolic disorder that leads to disruptions in glucose, lipid, and protein metabolism caused by insufficient insulin activity. The research investigates the innovative application of machine learning models, specifically Stacked Multi-Kernel Support Vector Machines Random Forest (SMKSVM-RF), to determine their effectiveness in identifying complex patterns in medical data. The innovative ensemble learning method SMKSVM-RF combines the strengths of Support Vector Machines (SVMs) and Random Forests (RFs) to leverage their diversity and complementary features. The SVM component implements multiple kernels to identify unique data patterns, while the RF component consists of an ensemble of decision trees to ensure reliable predictions. Integrating these models into a stacked architecture allows SMKSVM-RF to enhance the overall predictive performance for classification or regression tasks by optimizing their strengths. A significant finding of this study is the introduction of SMKSVM-RF, which displays an impressive 73.37% accuracy rate in the confusion matrix. Additionally, its recall is 71.62%, its precision is 70.13%, and it has a noteworthy F1-Score of 71.34%. This innovative technique shows potential for enhancing current methods and developing into an ideal healthcare system, signifying a noteworthy step forward in diabetes detection. The results emphasize the importance of sophisticated machine learning methods, highlighting how SMKSVM-RF can improve diagnostic precision and aid in the continual advancement of healthcare systems for more effective diabetes management.

The clustering, one of statistical analysis, can be used for understanding population patterns and as a basis for more targeted policy making. In this ecological study, we explored the population dynamics across 23 districts/cities in Aceh Province. The study used the Aceh Population Development Profile Year 2022 data, focusing on the total population, in-migrants, out-migrants, fertility, and maternal mortality as variables. The study employed descriptive statistics to ascertain the data distribution, followed by the Shapiro-Wilk test to evaluate normality, which is crucial for selecting the appropriate statistical methods. The Spearman test was used to determine correlations between the total population and the variable as indicators. Probabilistic Fuzzy C-Means (PFCM) method is used for clustering. To optimize clustering, the silhouette coefficient was calculated using the Euclidean Distance and the elbow method, with the results analyzed using R-4.3.2 software. This study's design and methods aim to provide a nuanced understanding of demographic patterns for targeted policy-making and regional development in Aceh, Indonesia. Based on the data normality test results, only fertility (p-value = 0.45), while the other variables are not normally distributed. Spearman test was used, and the results showed that only in-migrants (p-value = 1.78 x 10-6) and out-migrants (p-value = 2.30 x 10-6) correlated to the Aceh Province population. Using the population variable and the two variables associated with it, it was found that 4 is the best optimum number of clusters, where clusters 1, 2, 3, and 4 consist of three districts/city, nine districts/city, four districts/city and seven districts/city respectively.

Autism spectrum disorder (ASD) is a developmental disability resulting from neurological disparities. People with ASD frequently struggle with communication and social interaction, as well as limited or repetitive interests or behaviors. People with ASD may also have unique learning, movement, and attention styles. ASD sufferers can be interpreted as 1 in every 100 individuals in the globe having ASD. Abilities and requirements of autistic individuals vary and may change over time. Some autistic individuals are able to live independently, while others have severe disabilities and require lifelong care and support. Autism frequently interferes with educational and employment opportunities. Additionally, the demands placed on families providing care and assistance can be substantial. Important determinants of the quality of life for persons with autism are the attitudes of the community and the level of support provided by local and national authorities. Autism is frequently not diagnosed until adolescence, despite the fact that autistic traits are detectable in early infancy. This study will discuss the identification of Autism Spectrum Disorders using Magnetic Resonance Imaging (MRI). MRI images of ASD patients and MRI images of patients without ASD were compared. By employing multiple machine learning and deep learning techniques, such as random forests, support vector machines, and convolutional neural networks, the random forest method achieves the utmost accuracy with 100% using confusion matrix. Therefore, this technique is able to optimally identify ASD through MRI.

The Broad Learning System Forest (BLSF) model proved to be the preeminent classifier across all assessed datasets, demonstrating outstanding performance and efficiency. In the dataset, BLSF attained an accuracy of 94.53%, markedly exceeding standard BLS at 84.32%, Fuzzy BLS at 79.88%, and Intuitionistic Fuzzy BLS, which obtained just 15.55%. BLSF's efficacy is exemplified by its recall of 0.9452, accuracy of 0.9524, and F1-Score of 0.9481, emphasizing its ability to accurately categorize both positive and negative samples. Despite Tree BLS demonstrating a superior accuracy of 97.19%, its substantial memory use (789,862 MB) constrains its usability. In the Breast cancer dataset, BLSF attained a notable accuracy of 98.24%, trailing only Tree BLS at 98.75%, and markedly surpassing BLS (94.95%) and Fuzzy BLS (78.68%). The AUC-ROC score of BLSF at 0.9817 further substantiates its exceptional categorization capability. Although Incremental Weighted Ensemble BLS achieved a commendable performance of 98.57%, BLSF exhibited enhanced memory efficiency and a somewhat improved F1-Score of 0.9827. The Cardiovascular disease dataset revealed that BLSF achieved an accuracy of 76.61%, surpassing BLS at 71.38% and Fuzzy BLS at 53.23%. Although Tree BLS attained a superior accuracy of 91.75%, its considerable memory demands made BLSF the more efficient option. Incremental Weighted Ensemble BLS exhibited enhanced accuracy (81.99%) but required greater memory utilization. In imbalanced datasets such as PIDD (73.60%), Kidney failure (100%), and Heart disease (91.12%), BLSF consistently surpassed Fuzzy BLS and Intuitionistic Fuzzy BLS. It got a flawless score on the Kidney failure dataset, demonstrating its proficiency in managing difficult classifications with exceptional accuracy. BLSF excelled in accuracy, precision, recall, and efficiency across many datasets. In conclusion, the Broad Learning System Forest (BLSF) is a pioneering model that excels in accuracy and efficiency while exhibiting exceptional adaptability to complex and imbalanced datasets, marking a significant advancement in medical data classification.