Academy Journal of Science and Engineering

A REVIEW ON ARTIFICIAL INTELLIGENCE AND BIOTECHNOLOGY TECHNIQUES APPLIED IN MEDICAL RESEARCH

Rita Maneju Sunday — 2025-12-31

Biotechnology has to do with the use of biological systems to develop useful products by applying current trend in science, technology and artificial intelligence technique. This article therefore reviews the application of artificial intelligence in current biotechnology techniques applied in advanced medical research. Artificial intelligence and biotechnology techniques applied in medical research covers areas such as therapeutics, molecular diagnosis, drug discovery, personalized medicine, and drug development. Biotechnology techniques applied in advanced medical research include flow cytometry, next generation sequencing, enzyme-linked immunosorbent assay, recombinant deoxyribonucleic acid technology, polymerase chain reaction, and electrophoresis. In conclusion, the integration of artificial intelligence and advanced biotechnology techniques in medical research has benefited humans globally in various healthcare areas namely gene therapy, diagnostics, stem cell therapy, and development of drugs such as vaccine, insulin and antibiotics.

A NOVEL DUAL-FRAMEWORK FOR AI SYNTHETIC MEDIA DETECTION BASED ON PHYSIOLOGICAL AND LINGUISTIC INCONSISTENCIES

Mrindoko Mrindoko Nicholaus — 2025-12-31

The rapid advancement of generative artificial intelligence has intensified the challenge of detecting AI synthesized facial media, commonly known as deepfakes. This study introduces a novel dual-framework that fuses physiological and linguistic inconsistency analysis for robust synthetic media detection. The first component, Spatiotemporal Drift Entropy Mapping (SDEM), quantifies micro-temporal irregularities in facial motion using entropy and spectral variance of 468 FaceMesh landmarks. The second component, Inverse Phoneme Reconstruction Modeling (IPRM), predicts phoneme sequences directly from landmark trajectories and aligns them with audio-derived phonemes to reveal cross-modal mismatches. Evaluated on FaceForensics++ and DFDC, the proposed framework achieves a mean AUC of 0.967 and 0.943, respectively, surpassing single-module baselines (SDEM AUC = 0.923, IPRM AUC = 0.887) and competing deep architectures such as EfficientNet (AUC = 0.999) while maintaining interpretability through physiolinguistic cues. Experiments further demonstrate resilience against compression, occlusion, and adversarial perturbations. Limitations include reduced accuracy on extremely low-resolution videos and reliance on precise facial and audio segmentation. This research establishes a reproducible, interpretable pathway toward physiolinguistically grounded deepfake detection, providing both methodological novelty and practical forensic utility.

DEVELOPMENT OF DATA-DRIVEN SYSTEM FOR EARLIER CHILDHOOD MALNUTRITION PREDICTION

Mrindoko Mrindoko Nicholaus — 2025-12-31

Childhood malnutrition remains a critical issue in Tanzania, impacting the health and development of children under five. Early detection and intervention are vital to mitigating its effects, yet they are often hampered by a lack of effective tools. This study addresses this challenge by developing a data-driven system that uses machine learning to predict malnutrition risks early in childhood. Unlike previous studies that focused on limited data types or specific regions, this research presents an original approach that integrates multiple data categories to enhance prediction accuracy and relevance to the Tanzanian context. T he system analyzes a range of factors, including socioeconomic factors (poorest, Urban-Rural), health data such (height, weight, stunted, wasted, underweight, sex and age) and environmental variables (healthy status), to identify at-risk children before they exhibit significant symptoms. By leveraging a Random Forest algorithm, the study achieved a high accuracy of 96%, demonstrating the model's strong predictive performance. The data used for model development were obtained from a publicly available Kaggle dataset, which provides a valuable foundation but also represents limitations, as the secondary and non-Tanzanian data may affect the model’s generalizability to local contexts.

ETHNOBOTANICAL SURVEY AND COMPARATIVE ABUNDANCE OF FICUS BENGHALENSIS AND FICUS RELIGIOSA IN AGAIE LOCAL GOVERNMENT AREA NIGER STATE, NIGERIA

Philip Ogbevire — 2025-12-31

Ficus benghalensis and Ficus religiosa are culturally, ecologically and medicinal significant tree species in the tropics, yet little is known about their distribution and utilization in Niger State, Nigeria. This study investigated their abundance and ethnobotanical significance in Agaie Local Government Area using stratified random sampling, morphological identification and ethnobotanical survey with 54 informants. A total of 2,293 trees were recorded, comprising 1,691 F. benghalensis and 602 F. religiosa. F. benghalensis was more abundant with the highest density (316 trees) recorded in G.R.A Area 3. Both Species were cited for multiple uses including shade provision, goat fodder and fever treatment, typhoid, headaches, and body pain. Quantitative indices revealed higher relative frequency of citation (RFC = 0.74) and use value (UV = 2.09) for F. benghalensis compared to F. religiosa (RFC = 0.48; UV = 0.56). Fidelity levels revealed high cultural dependence with F. benghalensis exclusively cited for fever (100%) and F. religiosa strongly associated with body pain relief (96%). Both species demonstrated strong informant consensus (ICF ≥ 0.96). These results highlight the ecological dominance and cultural preference for F. benghalensis but reaffirm the sustained medicinal importance for both species. Conservation and phytochemical studies are recommended to ensure their sustainable use.

EXPLORATORY EVALUATION OF MACHINE LEARNING ALGORITHMS IN SICKLE CELL GENOTYPE DETECTION FROM HAEMORHEOLOGICAL PARAMETERS DATASET OBTAINED IN NIGERIA

Rahman Abiodun Olalekan — 2025-12-31

Sickle cell disease (SCD) patients have characteristic abnormal haemoglobins that cause red blood cells to become sickle-like in shape, leading to various complications. Early detection is desirous, yet existing diagnostic methods require high cost and deep learning curves. This study evaluated the potential of three machine learning (ML) algorithms—Random Forest, Support Vector Machine (SVM), and a Neural Network—in detecting sickle cell genotypes (SS, AS, AA) from a Nigerian dataset of 54 participants using haemorheological parameters. We employed a stratified 5-fold cross-validation methodology to ensure reliable performance evaluation. The Random Forest and SVM models achieved the highest mean accuracy at 90.9% ± 5.8%. Feature importance analysis confirmed Packed Cell Volume (PCV) as the most discriminative parameter, followed by Plasma Viscosity (PV) and Age. While all models demonstrated high sensitivity in identifying sickle cell anaemia (SS), they consistently failed to correctly classify the sickle cell trait (AS), a critical limitation highlighted by the validation. Our findings suggest that ML leveraging routine lab parameters is a promising screening tool for sickle cell disease, but is not yet viable for comprehensive genotype classification due to challenges with small dataset size and class imbalance. Future works need to focus on acquiring larger, more balanced datasets to improve the detection of the AS trait.

PERFORMANCE ANALYSIS OF A SOLAR BATTERY CHARGE CONTROLLER FOR TYPICAL RESIDENTIAL PV SYSTEMS IN NIGERIA: A SIMULATION APPROACH

Gerard Nonso Obiora — 2025-12-31

The performance analysis of a Perturb and Observe (P&O) Maximum Power Point Tracking (MPPT) based solar battery charge controller suitable for typical residential photovoltaic (PV) systems in Nigeria was carried out in this study. The system was modelled using MATLAB/Simulink environment, with specific attention given to modelling and observing a lead-acid battery behaviour in terms of State of Charge (SOC) and Depth of Discharge (DoD). The system comprises of a PV array, MPPT controller, buck converter, lead-acid battery, inverter, and Alternating Current (AC) load. The simulation results demonstrated a gradual charging and discharging pattern. A +0.97 and +0.69 SOC change was observed for SOC of 50.97% and 90.69% under 33 minutes for both cases. The system also performed satisfactorily under varying irradiance levels – a stable 32V output PV voltage was obtained under high solar irradiance (1000

ANTI-TRYPANOSOMAL EFFECTS O7F SOME SELECTED NIGERIAN MEDICINAL PLANTS

Enimie Endurance Oaikhena — 2025-12-31

Treatments of African trypanosomiasis have been dependent on the use of synthetic drugs which have several drawbacks. This is compounded by the emergence of drug-resistant parasites. These limitations have craved the urgent need for novel drugs from non-synthetic sources such as plants. The study, therefore, was aimed at evaluating in vitro antitrypanosomal activity of crude leaf extracts (CLE) of Acalypha wilkesiana, Annona muricata, Calotropis procera, Hyptis sualovens, Momordica charantia, Artemisia annua, and Siphonochilus aethiopicus using Hexane, ethyl acetate and methanol as solvents. Extraction was carried out by Soxhlet extraction method and different concentrations (100 µg/ml to 0.049 µg/ml) of the selected plants were evaluated for antitrypanosomal activity against trypomastigotes stage of Trypanosoma brucei brucei S427 in 96 well plates. Bioassay of CLE was assessed using Alamar blue TM assay. Hyptis sualovens and M. charantia, displayed moderate activities against T. b. brucei S427 with an EC50 range of 11- 14 µg/ml. Poor activities (EC50 ranging between 19 - 53 µg/ml) were exhibited by all extracts of A. annua, A. wilkesiana, A. muricata, C. procera and H. suaveolens methanol extract when tested against the parasite. This finding confirms the efficacy of some of the plants that are used by local herdsmen in the treatment of animal trypanosomiasis.

AN OPTIMIZED XGBOOST FOR FALSE POSITIVE REDUCTION IN A NETWORK INTRUSION DETECTION

Adeola O. Kolawole — 2025-12-31

Cybersecurity operations are increasingly challenged by large volume of false alerts produced by Intrusion Detection Systems (IDS) which leads to analyst fatigue and increases the likelihood of missing real threats. This study proposes an optimized eXtreme Gradient Boosting (XGBoost) model designed to reduce false positives and improve operational reliability of IDS using University of New South Wales-Network Intrusion Detection System-15 (UNSW-NB15) dataset for validation of the model. The optimization included systematic hyperparameter tuning of key parameters such as learning rate, maximum tree depth, gamma, subsampling ratio, and L1/L2 regularization to balance model complexity and generalization. The performance of the model was evaluated against reproduced benchmark ensemble classifier under identical conditions. The benchmark achieved False Positive Rate (FPR) of 17.69%, while the proposed XGBoost model reduced it to 5.85%, representing a 66.9% improvement and 2,925 fewer false alerts on the test set. In real world deployment, this substantial deduction would significantly lower alert fatigue and enable timely and effective responses to genuine attacks. The most significant gain was observed in the classification of legitimate “Normal” traffic where the FPR decreased from 9.22% in the benchmark model to 0.12%. The results demonstrate that a single well-tuned XGBoost model can provide high accuracy (94.15%) while substantially improving operational dependability. This study shows that prioritizing false positive reduction offers a practical path toward building deployment-ready IDS solutions. The novelty of this research is in its emphasis on minimizing the false positive rate (FPR) over accuracy as the main performance metric.

A FRAMEWORK FOR E-PAYMENT AND FINES NOTIFICATION FOR TANZANIA TRAFFIC POLICE OFFENCES

Waziri Ketto — 2025-12-31

This research critically investigates the efficiency of the existing e-paying system in use by the traffic police in Tanzania, in an attempt to come up with an enhanced approach. A convergent parallel approach to a mixed-methods design was used to seek the quantitative elements of the research by carrying out an online survey of 100 participants comprising drivers, traffic police, and the Tanzania Revenue Authority. A key part of the research methodology was to seek the opinions of key experts through the use of in-depth interviews. This was done in areas of high traffic in Dar es Salaam, Arusha, Dodoma, Mbeya, and Mwanza. Results have established key weaknesses in the current system, whereby the system fails to be interoperable with the Tanzanian government system, allow real-time responses, or be user-centric. To address the challenges encountered, the research presented an innovative API-based e-payment and fines notification system that incorporates mobile and web interfaces, real-time fines allocation and payment verification, central and secured management of user and system data, and automated reporting and user-friendly interfaces that would be ideal on a global scale. Even though research on e-payment and fines notification systems had previously been conducted by the research work in order to address the discrepancies in the current system and pose an effective solution to the challenges reported by the study, the research faced some limitations. For instance, the research relied on specific urban locations and prototypes to test the system.

EVALUATING ADOPTION AND OPERATION OF AN ON-CAMPUS BIOGAS ENERGY SYSTEM AT DAVID UMAHI FEDERAL UNIVERSITY OF HEALTH SCIENCES, UBURU, NIGERIA

Ikenna Uchechukwu MBABUIKE — 2025-12-31

The growing global focus on sustainable energy has renewed interest in biogas, especially in developing regions facing energy poverty and weak organic waste management. This study examines the feasibility, awareness, and adoption potential of campus-based biogas systems among food vendors, cleaners, and security staff at David Umahi Federal University of Health Sciences, Uburu, and its host community. A mixed-method design combined questionnaire, focus group discussions, and statistical analysis to generate primary data from forty-seven respondents. Findings show that although many participants (70.2%) were aware of biogas technology, only 48.9% demonstrated adequate understanding of its operation. However, a large majority (97.9%) expressed willingness to adopt biogas, provided affordability, safety, and reliability are assured. Key barriers include safety concerns (29.8%), perceived maintenance difficulty (23.4%), and high initial costs (19.1%). Chi-square tests revealed significant relationships between awareness, gender, location, and willingness to participate in co-ownership arrangements (p < 0.05). Overall, the study indicates that biogas is a viable and sustainable energy option for the university setting. It recommends targeted capacity building, technical support mechanisms, and pilot-scale installations to improve acceptance, strengthen user confidence, and promote long-term adoption of campus-based biogas technology. These actions will enhance environmental stewardship and strengthen local energy resilience significantly.