Browsing by Author "Villota Ojeda, Angie Vanessa"
Now showing 1 - 2 of 2
Results Per Page
Sort Options
Item Adquisición de información relativa a los niveles de colesterol en sangre mediante el análisis morfológico de la señal PPG: Estudio de factibilidad(Universidad Santiago de Cali, 2024) Villota Ojeda, María Yoselin; Villota Ojeda, Angie Vanessa; Argüello Prada, Erick JavierThe detection of cholesterol is carried out through an invasive procedure that involves the extraction of blood through venipuncture, which allows the analysis and measurement of lipids through enzymatic methods, for its part, this technique involves pain, emotional distress and risk of infection. Therefore, this study investigated whether the information provided by the morphology of the PPG signal can predict existing cholesterol levels. 160 features of the morphological parameters of the PPG signal were extracted from forty-six records of people with high (n=12), medium-high (n=18) and optimal (n=16) cholesterol levels. In addition, the correlation analysis, Relief, mRMR and F-test were used to reduce dimensionality and redundancy in the characteristics, implementing the more optimal ones in the different families of models such as regression trees, support vector machines and Gaussian processes. The combination of the selection method and model, which showed the best performance was the Relieff and Rational Quadratic GPR respectively, achieving a root mean square error (RSME) of 2,710 mg/dL and a mean absolute error (MAE) of 1,372 mg/dL. Likewise, a large part of the implemented models obtained better performance with the previous method. Acquiring information on cholesterol levels is feasible using the study of the morphology of the PPG signal, in turn, these results could be useful due to the great contribution of PPG characteristics that were significantly correlated with lipid concentration in bloodItem Non-invasive prediction of cholesterol levels from photoplethysmogram (PPG)-based features using machine learning techniques: a proof-of-concept study(Cogent OA, 2025-02-18) Argüello Prada, Erick Javier; Villota Ojeda, Angie Vanessa; Villota Ojeda, María YoselinRegular monitoring of cholesterol levels is crucial to reducing the risk of vascular blockage and preventing atherosclerotic cardiovascular diseases. However, standardized cholesterol measurement tests involve 8–12 hours of strict fasting and blood extraction via finger pricking, which may cause pain and discomfort. This study explores the usefulness of fiducial-based features extracted from the photoplethysmogram (PPG) in estimating blood cholesterol by combining feature selection methods and machine learning techniques. We extracted 150 features from forty-six 2-minute PPG recordings and included participants’ age as a feature. Several variations of linear regressions (LR), regression trees (RT), support vector regressions (SVR), and Gaussian process regressions (GPR) were trained with the most relevant features. The rational quadratic GPR model achieved the lowest errors (MAE = 11.70, MSE = 281.57, and RMSE = 16.78 mg/dL) and the highest coefficient of determination (r2 = 0.832) when combined with ReliefF. The proposed method holds promise for developing lightweight and non-invasive approaches for blood cholesterol estimation, although it may require further validation due to the limited sample size.