Journal of Current Science and Technology

Optimizing Diabetes Prediction: An Evaluation of Machine Learning Models Through Strategic Feature Selection

2024-03-15T18:48:27+07:00

Diabetes, a widespread chronic ailment in the United States, imposes significant economic and health burdens, impacting quality of life and life expectancy. This study analyzes a clinical dataset of 253,680 patients from the Behavioral Risk Factor Surveillance System (BRFSS). The dataset encompasses 21 predictors, including high blood pressure, cholesterol, body mass index (BMI), smoking, stroke, heart disease, physical activity, fruit consumption, vegetable consumption, alcohol consumption, insurance coverage, lack of medical visits due to financial constraints, general health, days with mental health issues, days with physical injuries in the past 30 days, difficulties in walking, gender, age, income, and education level. The objective is to balance the training dataset, compare different supervised machine learning models, and identify critical clinical features contributing to diabetes using unsupervised feature selection methods. A total of 34 machine learning models in MATLAB2023a were trained and compared. Quadratic Support Vector Machine (SVM), Coarse Gaussian SVM, and Narrow Neural Networks achieved the highest training accuracy (76.3%), while the Bilayered Neural Network attained 74.7% on an unseen test dataset. Among all, Quadratic SVM demonstrated the best overall performance based on average accuracy, precision, recall, and F1 score. Feature selection highlighted nine key predictors: high blood pressure, high cholesterol, BMI, heart disease, physical activity, general health, recent bodily injuries, mobility issues, and age. A model trained on these features achieved a commendable accuracy of 75.4%, demonstrating the feasibility of a simplified, efficient diagnostic tool with a diagnostic efficacy of 0.7.

This study underscores the potential of streamlined models to predict diabetes with fewer parameters while maintaining high accuracy, offering a valuable tool for healthcare diagnostics.

Optimizing Chronic Kidney Disease Prediction: A Machine Learning Approach with Minimal Diagnostic Predictors

2024-04-01T11:38:49+07:00

Chronic kidney disease (CKD) is a major public health issue that necessitates accurate diagnostic methods for effective management. This study involved training an open-source clinical dataset of 200 patients from Enam Medical College, comprising 28 clinical features, obtained from the UCI machine learning repository. After preprocessing to ensure a balanced dataset for objectivity, the data was split into training and testing sets in an 80:20 ratio. The research trained 22 machine learning models, including Naïve Bayes, decision trees, support vector machines (SVM), logistic regression, ensemble methods, kernel models, and neural networks. These models were evaluated using several metrics-accuracy, precision, recall, F1-score, and the area under the receiver operating characteristic (ROC) curve-computed through 5-fold cross-validation to assess their performance and ensure they were not overfitting or underfitting. The best-performing model was the Kernel Naïve Bayes, achieving a 96.55% accuracy, 95% precision, 98.28% recall, and 96.61% F1-score on the training dataset. For the test dataset, it showed a slight performance drop but remained robust with 92.86% accuracy, 87.50% precision, 100% recall, and 93.33% F1-score. Furthermore, feature selection techniques such as minimum-redundancy-maximum-relevance, Chi2, ANOVA, and Kruskal-Wallis tests were used to determine the most significant predictors. It was found that only four features-packed cell value, stages of glomerular filtration rate, specific gravity of urine, and albumin content in urine-were necessary for maintaining similar model performance. This systematic approach not only highlighted critical clinical features but also helped in simplifying the model complexity, which could benefit broader medical applications like lung cancer screening by reducing screen time, resources, and medical costs.

In Vitro Biological Activity of Tiliacora triandra (Colebr.) Diels Root Extract

2024-04-18T13:03:35+07:00

In this study, we explore the phytochemical properties of Tiliacora triandra (Colebr.) Diels, commonly known as Yanang, a herb traditionally used for treating fever and malaria. Our objective is to isolate and analyze bioactive compounds from T. triandra's root extract, with a specific focus on tiliacorinine (T1), assessing its antioxidant, and anticancer capabilities. The antioxidant activity of T. triandra extract (TTE) was evident, with IC₅₀ values of 196.32 ± 10.51 µg/mL and 88.22 ± 1.95 µg/mL in the DPPH and ABTS assays, respectively. The results also demonstrated that TTE and T1 had a significant cytotoxic effect, with IC₅₀ values varying in a concentration-dependent manner against A549 with the IC₅₀ of 48.25 μg/mL and 26.38 μg/mL, respectively. Additionally, both TTE and T1 displayed anti-inflammatory effects, with the most effective inhibition of nitric oxide production observed at a concentration of 40 μg/mL. This study underlines the potential of T. triandra, and its compound tiliacorinine in developing treatments with antioxidant, anti-inflammatory, and anticancer properties.

Interdisciplinary Research for Predictive Maintenance of MRI Machines Using Machine Learning

2024-08-13T15:24:40+07:00

Predictive maintenance is crucial for ensuring the reliability and availability of medical equipment, particularly MRI machines in healthcare facilities. This study presents a comprehensive approach to predictive maintenance of MRI machines using machine learning techniques. The objective of this research is to develop and evaluate predictive models capable of identifying patterns and indicators of impending equipment failures, thereby improving the operational efficiency and reliability of MRI machines. We utilized a dataset comprising historical maintenance records, sensor readings, and environmental conditions collected from three 1.5 T Siemens MRI machines at MGM Hospital, Warangal, Telangana, India. The dataset, initially consisting of 96 records and expanded to 1000 through computer-generated data, encompasses various operational aspects, including temperature, humidity, vibration, power consumption, and coolant flow rate. This study investigated the efficacy of multiple machine learning algorithms for predicting equipment failures, including Random Forest, Gradient Boosting, Long Short-Term Memory (LSTM) networks, and Support Vector Machines (SVM). Model performance was evaluated using standard metrics such as F1-score, accuracy, recall, and precision. Results indicate that LSTM networks achieved the highest accuracy at 89%, while SVM displayed the lowest at 82%. These findings validate the potential of machine learning in anticipating equipment breakdowns and enabling proactive maintenance strategies for MRI machines. The outcomes of this research have significant implications for enhancing the reliability and operational efficiency of medical imaging equipment in healthcare settings.

Effect of Thai Yoga Exercise for Reducing Pain and Improve Functional Disability in Patients with Text Neck Syndrome: A Randomized Controlled Trial

2024-07-31T13:33:49+07:00

Text neck syndrome (TNS) is prevalent among mobile users, and exercise is a simple, non-invasive approach, for treating chronic neck pain. Research on Thai yoga (TY) exercise to relieve or reduce neck and shoulder pain in people with TNS is still limited. This study aimed to investigate the effects of TY exercise on health, physical function and quality of life in patients with TNS. A randomized clinical trial was conducted from April to July 2023 (TCTR20230920002, 10/2023). Seventy-eight volunteers with TNS participated in the study, performing TY exercises, while continuing their routine daily life (CON) for 6 weeks, with week 8 as a follow-up appointment. Clinical evaluation was conducted using a numerical pain-rating scale, functional disability scored by the neck disability index questionnaire (NDI), range of motion (ROM) and cervical vertebral angle, rounded shoulder angle by goniometry, and health-related quality of life (QoL) assessed by the 36-Item Short Form Survey. The results of the numerical pain-rating scale for TY group in the follow-up (6.71±0.52 weeks) were significantly different (p-value < 0.001) from the baseline (4.18±0.95) while the CON group showed no change. The NDI also decreased from the baseline (5.03±1.38) in the TY group, while the CON group remained at 12.03±1.98. The data of craniovertebral angle of TY group show an increase from baseline (62.44±2.32) while CON group was 39.37±4.73. The results showed that the rounded shoulder angle of TY group decreased to the normal range. The TY group shows an increased degree from baseline. The results for the TY group demonstrated a significant increase in QoL. These findings suggest that TY exercise is effective for relieving TNS when compared with the CON group. These positive outcomes provide strong support for the use of TY as an effective intervention for treating TNS symptoms.

Artificial Intelligence, Cybersight Detection of Diabetic Retinopathy in the Elderly in Vietnam

2024-08-09T13:40:59+07:00

Diabetic retinopathy (DR) is a highly prevalent cause of vision loss worldwide. Detection of DR requires substantial human resources and high medical costs. Therefore, the use of diagnostic software has been recently explored. The study aimed to assess the results of DR diagnoses by Cybersight, an artificial intelligence software. A total of 1,012 patients with type 2 diabetes mellitus (1,943 eyes) with a mean age of 74.61 ± 6.73 years were included. Comprehensive demographic and clinical data were gathered, and all patients underwent color fundus photography following Cybersight's standardized protocols. The study compared Cybersight's accuracy with that of ophthalmologists in identifying key DR lesions, including retinal microvascular changes, exudates, hemorrhages, the diagnosis and staging of DR, using sensitivity, specificity, and weighted Kappa metrics. The prevalence of DR was 16.2%. A high level of agreement was found between Cybersight and ophthalmologists in DR diagnosis, with a sensitivity of 85.0%, specificity of 95.8%, and a weighted Kappa of 0.78. The presence of cataracts and the degree of pupil dilation notably impacted on the accuracy of DR diagnosis. The results have important implications for the potential application of Cybersight as a low-cost and effective tool for diabetic eye screening.

Study of MCC, Mannitol and SiO2 Based Co-processed Excipient for Improving the Direct Compression Properties of Paracetamol using SeDeM/SeDeM-ODT Expert System

2024-09-03T16:40:10+07:00

Co-processed excipients have enhanced functionality attributes required for developing tablet dosage forms by direct compression technology. In the present study, mannitol, microcrystalline cellulose, and silicon dioxide were developed by dry granulation and proposed as a viable solution for correcting the flowability and compressibility issues commonly encountered in developing paracetamol tablets. SeDeM (12 parameter based) and SeDeM ODT (15 parameter based) expert systems were employed as tools for developing orally disintegrating tablets of paracetamol. SeDeM diagram of paracetamol, mannitol, microcrystalline cellulose and co-processed excipients were prepared and parametric index values of 3.9 and 2.01 indicated poor flowability and compressibility properties of paracetamol. The percentage of corrective excipient required to correct the flowability and compressibility of paracetamol was then calculated followed by preparation of SeDeM ODT expert system diagrams. An increase in concentration of silicon dioxide from 1% to 5% in the co-processed excipient resulted in a decrease in dissolution rate due to increased apparent viscosity/ gel-like structure at higher concentrations of silicon dioxide. The study revealed that co-processed excipient sample with 2.5% of silicon dioxide showed the highest IGCB value of 6.39, implying its suitability for the direct compression of tablets.

The Effect of Foam-mat Drying Parameters on the Quality of Instant Porridge Fortified with Kale as a Calcium source

2024-09-18T20:09:20+07:00

The objective of this research was to develop instant porridge fortified with calcium for the elderly. This study aimed to investigate the effects of foaming and drying processes on the quality parameters of Leuang Pratew brown rice instant porridge. The results showed that using a 20% concentration of foaming agent and a whipping time of 10 minutes was suitable for producing the instant porridge. The foam porridge exhibited low density, high stability, and significant overrun. The optimized process for foam-mat drying involved a drying temperature of 65ºC and a drying time of 2 hours. The properties of the instant porridge were as follows: yield of 32.43%, moisture content of 4.03%, water activity of 0.356, water absorption index of 3.47 g/g, water solubility index of 36.89%, and glycemic index of 62.48. The sensory test using the 9-point hedonic scale indicated that the overall acceptance scores of the developed product were 7.46 (liked moderately). Nutrient analysis of the calcium-fortified instant porridge revealed a total energy of 365.26 kcal, total fat of 3.38 g, total protein of 22.36 g, total carbohydrate of 61.35 g, ash content of 9.81 g, moisture content of 3.30 g, calcium content of 3,604 mg/kg, and sodium content of 704 mg/kg. The microbiological qualities were within safe levels.

Effectiveness of a Web-based Application on Health Literacy, Self-Efficacy, Self-Care, and Glycemic Control in Elderly Patients with Type 2 Diabetes: A Randomized Controlled Trial

2024-08-28T19:23:22+07:00

The web-based application serves as a technology platform that enhances health literacy by providing diabetes-related content, activities, and quizzes. It boosts self-efficacy by enabling users to set health goals, promote self-care through practical activities, and use data collection and reminders to help regulate blood sugar and adopt healthy habits for elderly patients. This research study is experimental and includes a randomized controlled trial. The objectives of this study were to compare the effectiveness of using web-based applications in developing health literacy, self-efficacy, self-care behaviors, and blood sugar levels. The sample group consisted of 60 elderly individuals with type 2 diabetes and uncontrolled blood sugar levels, made up of 30 people in the experimental group and 30 people in the control group. In the experiment, the web-based application was utilized over a period of 12 weeks to carry out a health literacy assessment. Self-efficacy, self-care, and blood sugar levels (FBS and HbA1C) were assessed before the experiment (Week 0, T1), immediately after the experiment (Week 12, T2), and during the follow-up period (Week 24, T3). The data were analyzed using two-way repeated measures ANOVA statistics. The results indicated that in T2 and T3, the experimental group achieved an average health literacy score, accompanied by an increase in self-efficacy, enhanced self-care, and a reduction in FBS and HbA1C that was greater than in T1, with statistical significance (p<.05). Moreover, the experimental group showed a greater increase in average health literacy scores, self-efficacy, and self-care compared to the control group, with fasting blood sugar (FBS) and HbA1C levels decreasing more significantly than in the control group, at a statistically significant level (p<.05). In conclusion, the web-based application was able to promote health literacy among patients as it improved self-efficacy, self-care behavior, and the ability to regulate glycemic control in T2 and T3.

Evaluation of Phytoconstituents, Nutritional Quality, and In Vitro Biological Activities of Red Rice Ethanolic Extract from Different Regions of Chiang Rai and Phayao

2024-09-16T19:32:13+07:00

Red rice, a pigmented rice, is a staple food in Thailand that has beneficial biological properties for the consumer. The objective of this research was to assess the phytochemical composition, macro- and micro-nutritional quality, as well as the anti-glycation and in vitro antioxidant properties of three planting sites located in Chiang Rai and three planting sites located in Phayao. Raw red rice was extracted with 70% ethanol, and the phytoconstituents were evaluated using colorimetric analysis and HPLC techniques. The antioxidant activity, ROS production, lipid peroxidation, and in vitro anti-glycation properties were examined. According to the results, the greatest levels of total phenolic and total flavonoid contents were detected in CRR1 and PYR1. The amounts of fat, protein, fiber, and carbohydrates were comparable in red rice extracts. The PYR3 had the greatest quantities of iron and zinc, whereas the CRR3 had the highest levels of magnesium and potassium. The CRR1 and PYR1 had the highest amounts of vitamin E compounds and γ-oryzanol. Additionally, CRR1 and PYR1 had greater antioxidant capacities in comparison to the other red rice extracts. In the RAW264.7 macrophage cell, it was demonstrated that CRR1 and PYR1 prevented the generation of AGE at higher concentrations and had the strongest inhibitory effects on linoleic acid peroxidation and ROS production. The study's findings offered additional valuable sources of red rice from Chiang Rai and Phayao, giving consumers the option to choose red rice as beneficial for their health.

In Vitro Assessment of Antimicrobial Activity and Synergistic Effects of Ethanolic Extracts from Six Medicinal Plants

2024-09-02T17:24:12+07:00

This in vitro experimental study examines and evaluates the antimicrobial and synergistic effects of the ethanolic extract of six plants: Biancaea sappan (L.) Tod, Bauhinia malabarica Roxb, Carthamus tinctorius L., Derris scandens (Roxb.) Benth, Hibiscus sabdariffa L., and Piper nigrum L. against common microbial species representing gram-positive, gram-negative bacteria, and fungi, consisting of Staphylococcus aureus, Escherichia coli, Staphylococcus epidermidis, Pseudomonas aeruginosa, and Candida albicans. The plants were extracted using 90% ethanol. According to the standard method of agar diffusion assay, the micro-dilution method for minimal inhibitory concentration (MIC) and minimum bactericidal concentration (MBC) were determined. This study found that among the six plants, only B. sappan and B. malabarica exhibited moderate inhibitory effects against S. aureus and S. epidermidis. B. sappan had MIC values of 250 µg/ mL and 125 µg/ mL, respectively, and B. malabarica showed MIC values of 62.50 µg/ mL and 31.25 µg/ mL, respectively. The synergistic effects of a combination of B. sappan and B. malabarica extracts at a ratio of 25:75 were analyzed, and it was found that the combination inhibited S. aureus and S. epidermidis with MIC values of 250 µg/ mL and 125 µg/ mL, respectively. The fractional inhibitory concentration index (FICI) and the fractional bactericidal concentration index (FBCI) indicated antagonistic or synergistic effects of the combination, with FICI and FBCI values of 2.5–5.0 for both B. sappan and B. malabarica extracts in the 25:75 mixture. In conclusion, single plant ethanolic extracts of B. sappan and B. malabarica possess potent antimicrobial activity to varying degrees. However, the antimicrobial potency of the 25:75 ratio mixture of these extracts was shown to decrease against the same organisms, with in vitro antimicrobial activity and antagonistic effects observed only against the tested gram-positive bacteria.

Optimization of Ultrasound-assisted Anthocyanin Extraction from Black Rice Bran for Simultaneous Coloring, UV Protection, and Antioxidant Silk Finishes

2024-09-05T22:15:20+07:00

This study examined the use of ultrasound to extract anthocyanin from agricultural byproduct black rice bran as a natural colorant and a multifunctional finishing agent for silk fabrics, taking into account the various health advantages linked to anthocyanins. The study employed response surface methodology to identify the optimal extraction process that would yield the highest extraction efficiency for total anthocyanin content. The optimal conditions were 30 Hz ultrasonic power, a liquor-to-material ratio of 21, 60°C an ultrasound temperature, and 30 minutes of an ultrasound time, yielding 173.25 mg/L anthocyanin. Silk fabrics were dyed in a reddish purple tone, and the mordant dyeing method produced a 34% higher color strength value than direct dyeing while displaying good colorfastness (grade ≥ 4) to washing and crocking. The dyed silks also provided excellent UV protection (UPF > 40+) and antioxidant activity, with a DPPH scavenging rate over 80%. Anthocyanins derived from black rice bran could thus be used as a bioactive functional colorant in medical and health-related textiles.

Confidence Intervals for the Zeghdoudi Distribution Parameter: Applications in Precipitation and COVID-19 Data Analysis

2024-10-07T11:36:27+07:00

This paper proposes four confidence intervals (CIs) for estimating the parameters of the Zeghdoudi distribution, which is commonly used in the analysis of lifetime data. We introduce and evaluate the likelihood-based, Wald-type, bootstrap-t, and bias-corrected and accelerated (BCa) bootstrap CIs using Monte Carlo simulation studies and apply them to two real datasets. We assess the effectiveness of these CIs by evaluating their empirical coverage probability (CP) and average length (AL), which offer valuable insights into their performance in various scenarios. Furthermore, we have developed an explicit formulation of the Wald-type’s CI formula, simplifying its computation. The results demonstrate that the CPs of likelihood-based and Wald-type CIs converge towards the nominal confidence level of 0.95 for all cases. However, when the sample size is small, the bootstrap-t and BCa bootstrap CIs have CPs less than 0.95. On the other hand, as sample sizes increase, the CPs of all CIs tend to approach 0.95. However, when the sample sizes are small, the CPs of the bootstrap-t and BCa bootstrap CIs tend to decrease. We verified the efficacy of CIs by applying them to precipitation data and COVID-19 mortality rate data, and the results matched those from the simulation study.

Influence of Hydraulic Properties on the Slope Stability of Unsaturated Soil from the Garinono Formation in Sabah

2024-08-18T09:16:13+07:00

This study investigates the hydraulic response and stability of silty clays from the Garinono Formation during the rainy season, utilizing peak monthly rainfall data recorded over a decade by the Department of Irrigation and Drainage, Sabah. The research focuses on the impact of rainfall infiltration on slope stability under extreme hydrological conditions, modelled using a 31-day antecedent rainfall event from January 2011. The study incorporates both field and laboratory data, assessing the Soil-Water Characteristic Curve (SWCC) and bulk density for three distinct soil samples: (i) MI (Intermediate Plasticity Silt), (ii) ML (Low Plasticity Silt), and (iii) CI (Intermediate Plasticity Clay). The SWCC and hydraulic properties were characterized using the Fredlund, & Xing (1994) model, with subsequent numerical simulations conducted using GEO-SLOPE International. Results indicated that sample (ii) ML exhibited increased cumulative water volume with rising rainfall intensity, influenced by its saturated and unsaturated hydraulic conductivity, as well as the FX model's fitting parameter . The study revealed that high ratios in samples (i) MI and (iii) CI led to greater runoff and reduced infiltration, characterized by lower water storage and higher discharge rates. Conversely, lower ratios in sample (ii) ML facilitated infiltration, increasing water storage and reducing discharge rates. A notable decrease in matric suction was observed during rainfall, with sample (ii) ML showing the most significant reductions, leading to early and substantial decreases in the Factor of Safety (FS), indicating potential instability. In contrast, samples (i) MI and (iii) CI maintained higher stability due to negative Pore-Water Pressure (PWP) and minimal suction reductions. These findings underscore the critical influence of soil permeability on slope hydrological response and stability during intense rainfall events. This study contributes to the understanding of the hydro-mechanical behaviour of the Garinono Formation, with implications for geotechnical design and slope stability analysis in tropical regions.

Strategy for Energy Savings in a Commercial Building Air-conditioning System with Chilled Water Storage: A Case Study in a Retail Mall in Thailand

2024-09-04T10:25:06+07:00

In this study, a commercial retail mall is used as a case study to integrate a chilled water storage (CHWS) with the existing chilled water system to reduce electrical energy consumption and capitalize on the economic benefits of electrical energy saving cost and the differential between on-peak and off-peak tariffs. This study aims to improve the chiller efficiency in three operating strategies: full storage, partial storage load leveling, and partial storage demand limiting, by operating the chillers at optimal part load conditions. Technical and economic assessments were conducted to determine the necessary storage capacity and appropriate operational strategies. In comparison to the existing operation, which uses two 800 RT chillers continuously during on-peak hours and one chiller operating at 20-30% capacity during off-peak hours, the proposed systems: three 800 RT chillers with a 9,150 m³ tank for full storage, one 800 RT and one 260 RT chiller with a 3,292 m³ tank for partial load leveling, and two 800 RT chillers with a 4,987 m³ tank for partial demand limiting, demonstrate significant potential to reduce electrical energy consumption. The full storage strategy achieves the lowest electrical energy consumption, followed by partial demand limiting and partial load leveling. Economically, partial demand limiting strategy emerges as the most feasible, providing a payback period of 7.42 years and an internal rate of return (IRR) of 14.92%. This is more favorable compared to payback periods of 8.6 and 9.65 years and internal rate of return of 12.08 and 10.33% for partial load leveling and full storage strategies, respectively.

Glutathione Levels after Glutathione Supplementation: A Systematic Review and Meta-analysis

2024-10-02T11:33:20+07:00

Glutathione is a crucial antioxidant and plays a vital role in many biochemical processes within living organisms. Abnormal levels or reductions in glutathione are linked to various health conditions and diseases. While glutathione supplementation might offer health benefits, there are ongoing concerns about its low bioavailability. This research aimed to examine the effects of glutathione supplementation on body glutathione levels through a systematic review and meta-analysis of primary studies. Randomized controlled trials (RCTs) were selected from online databases, including PubMed, Scopus, Google Scholar, and Cochrane Library. The studies compared participants who received glutathione supplementation with those in placebo or control groups by evaluating their glutathione levels. The results showed that five primary studies met the inclusion criteria. The quality assessment revealed that most studies had a low risk of bias or some concerns in various areas. However, there was a high risk of bias related to the selection of reported results, primarily due to multiple measurements or analytical methods. Three articles were included in the meta-analysis, which found no statistically significant difference in glutathione levels in erythrocytes [Standardized Mean Difference: 0.74, 95% CI (-0.44, 1.91); P = 0.22] or plasma [Standardized Mean Difference: 0.44, 95% CI (-0.21, 1.09); P = 0.19] between the intervention and placebo groups. This study concluded that glutathione supplementation does not significantly increase glutathione levels in erythrocytes or plasma. However, higher doses and longer durations of supplementation may potentially lead to increased glutathione levels in the body.