ABSTRACT: 

This paper aimed at developing Machine learning (ML) model that would predicts the presence of black soot particles in Niger-Delta region of Nigeria in real time. In most parts of Niger Delta, it is observed that burning activities have increased tremendously because of hardship the occupants faced each day. People trying to engage in dubious businesses like kpo fire, burning tires to extract metal wires, etc. in order to survive in the country thereby polluting the air with various particles dangerous to health. Inhaling all these particles have deep health side effect which damage some vital organs in the body and reduces the life span of individuals on earth. The situation has calls for urgent concern and this research work has come to bridge the gap by developing ML model that would cater for it. In order to achieve this, the wearable Internet of things (IoT) device was designed and implemented first, then used to gather dataset of soot particles within the targeted environments for the model’s development. Machine Learning algorithms such as Artificial Neural Network (ANN), linear regression, Support Vector Machine and Gradient boosting were deployed and retrained with the 80% black soot dataset. After the training, the 20% dataset were used for testing and validation.  The Mean Absolute Error (MAE) metric was used to evaluate the performance of the chosen algorithms. During testing, it was observed that the ANN algorithm outperformed the other algorithms, achieved 83.30% percentage accuracy and was selected. The model was stored in the IoT device memory and tested to checkmate the behavior of the system in real time application.  The predicted results gotten were the exact particles of the soot around the tested environments as was confirmed by laboratory test conducted. Hence, the system performed excellently and can now be deployed in such related areas of interest.

Keywords: Machine Learning; IoT; Niger Delta, MAE, Black Soot, Predictive model

ABSTRACT: 

Cutaneous leishmaniasis (CL) is a problematical tropical disease that is caused by Leishmania parasite. This parasite primarily infects macrophages replicating inside and colonizing as non-motile amastigotes to establish infection. The cutaneous form is predominant. Leishmania major and Leishmania tropica are the main causative parasites of this disease, especially in Iraq. Unfortunately, drugs used for leishmaniasis are very toxic and no vaccine is available. Many efforts were made to identify novel immunological targets to produce safe and effective vaccines against leishmaniasis.

Keywords:  Cutaneous, leishmaniosis, Iraq, Immune response.

ABSTRACT: 

Since the unveiling of the blueprint of the first human genome in 2001 and the subsequent advancements in high-throughput genetic analysis technologies, personalized nutrition (PN) has emerged as a novel scientific discipline. This emergence coincided with the initial commercial offerings of genotype-based nutrition advice. This article provides an overview of the existing evidence regarding the influence of genetic and epigenetic factors on the development of obesity, metabolic syndrome, and related conditions such as non-insulin-dependent diabetes mellitus and cardiovascular diseases. Furthermore, it critically evaluates the concepts of PN that revolve around this new genetic paradigm, considering both academic and commercial perspectives, and assesses their efficacy in inducing sustained changes in diet, lifestyle, and health improvement. Despite nearly two decades of research and the proliferation of commercial direct-to-consumer services, conclusive evidence supporting the effectiveness of gene-based dietary recommendations remains largely elusive. This underscores the need for novel approaches in future PN solutions, which should integrate a broader array of phenotypic metrics. Additionally, such solutions should offer a comprehensive suite of tools—including self- and bio-monitoring devices, feedback mechanisms, and AI-driven algorithms—that enhance adherence by aligning with individuals' physical, social, and value-based contexts.

Keywords:  Personalized nutrition, Genotype-based nutrition advice, Epigenetic, Obesity, Metabolic syndrome, Cardiovascular diseases, Diet, Lifestyle .

ABSTRACT: 

The bladder, a crucial organ for urine storage, can be considered as an Overactive Bladder (OAB), a dysfunction causing an urgent and sudden urge to urinate. OAB significantly impacts various aspects of daily life, affecting interpersonal interactions and routine activities such as work, travel, exercise, sleep, and leisure. The updated Global Constipation societal definition of OAB emphasizes its characteristics, encompassing sexual pressure, periodicity, nocturia, and sudden urination without evident urinary tract infection or other pathological signs. Recognizing the profound influence of OAB on physical and social functioning, vitality, and emotional well-being, this research addresses the imperative need for an effective diagnostic model. We aim to develop a supervised Machine Learning Model capable of classifying patients' bladders as either overactive (BAD) or not overactive (GOOD). Bayesian Beliefs Network (BBN) emerges as the preferred classification model due to its unique ability to model causation and correlation. In pursuit of this goal, an artificial intelligence system was developed to detect OAB issues. The employed algorithm estimates the probability of OAB based on input data, assuming mutual independence exclusivity among predictor variables during design and implementation. The results obtained demonstrate the successful classification of bladders into overactive or non-overactive categories, showcasing the potential of Bayesian Networks as a robust tool for addressing OAB diagnostic challenges. Notably, the Bayesian Network achieved a classification accuracy of 68%, underscoring its effectiveness in OAB classification. By recognizing the intricate relationship between fluid choices, age, urinary symptoms, and gender, the way for more targeted interventions is paved, personalized treatment plans and a refined clinical approach to addressing the prevalence of Overactive Bladder as a global health concern are modeled showing major improvement over existing approaches.

Keywords:   Artificial Intelligence, Machine Learning, Clinical Decision Support System, Bladder, Bayesian Network, Hyperactive bladder.