Search results

Diabetes is one of the life-threatening chronic diseases, which is already affecting 422m people globally based on (World Health Organization) WHO report as at 2018. This costs individuals, government and groups a whole lot; right from its diagnosis stage to the treatment stage. The reason for this cost, among others, is that it is a long-term treatment disease. This disease is likely to continue to affect more people because of its long asymptotic phase, which makes its early detection not feasible.

In this study, the authors have presented machine learning models with feature selection, which can detect diabetes disease at its early stage. Also, the models presented are not costly and available to everyone, including those in the remote areas.

The study result shows that feature selection helps in getting better model, as it prevents overfitting and removes redundant data. Hence, the study result when compared with previous research shows the better result has been achieved, after it was evaluated based on metrics such as F-measure, Precision-Recall curve and Receiver Operating Characteristic Area Under Curve. This discovery has the potential to impact on clinical practice, when health workers aim at diagnosing diabetes disease at its early stage.

This study has not been published anywhere else.

This study aims to investigate some physicochemical characteristics of Hildegardia barteri seed oils obtained by cold-pressing and solvent extraction procedures.

Crude oil samples were obtained from the kernels by cold pressing and solvent extraction. The physicochemical properties of the oil samples were investigated according to the standard procedures in published works of literature.

The oil yield was 55.7 and 97 per cent for cold-pressed kariya seed oil (CPKSO) and solvent-extracted kariya seed oil (SEKSO), respectively. Specific gravities, refractive indices, viscosities, iodine value, saponification value, peroxide value and acid value were 0.8742 and 0.9036; 1.4629 and 1.4584; 75.93 and 74.90 mPa.s; 55.78 and 53.56 g of I2/100g of oil; 249.76 and 253.90 mg KOH/g; 4.86 and 5.02 meq KOH/g; 2.12 and 2.09 mg KOH/g of oil for CPKSO and SEKSO, respectively. The physicochemical characteristics of kariya seed oil were not significantly affected by extraction method. The fatty acid profiles of CPKSO and SEKSO showed that the two oil samples contain 24.2 and 23.7, 31.3 and 29.3, 23.2 and 23.7 and 19.6 and 21.3 per cent of myristic, palmitic, stearic and linolenic acids, respectively. Lauric and oleic acids were present in very little proportions of 0.3 and 0.41; and 0.01 and 0.03 per cent, respectively, whereas linoleic acid was 1.4 per cent for the two oil samples. Significant differences in fatty acid profiles were observed for lauric, palmitic and linolenic acids (p = 0.05). Saturated and unsaturated fatty acids were about 79.0 and 77.11 per cent and 21.01 and 22.73 per cent for CPKSO and SEKSO, respectively.

This work promotes H. barteri tree beyond its use as a mere ornamental plant. The non-conventional seed oil it produces may find relevance in the food or biofuels industry subject to further investigation.

This study is the first to document the extraction and physicochemical properties of kariya seed oils.