Background: LDL-C is a strong risk factor for cardiovascular disorders. The formulas used to calculate LDL-C showed varying performance in different populations. Machine learning models can study complex interactions between the variables and can be used to predict outcomes more accurately. The current study evaluated the predictive performance of three machine learning models—random forests, XGBoost, and support vector Rregression (SVR) to predict LDL-C from total cholesterol, triglyceride, and HDL-C in comparison to linear regression model and some existing formulas for LDL-C calculation, in eastern Indian population. Methods: The lipid profiles performed in the clinical biochemistry laboratory of AIIMS Bhubaneswar during 2019–2021, a total of 13,391 samples were included in the study. Laboratory results were collected from the laboratory database. 70% of data were classified as train set and used to develop the three machine learning models and linear regression formula. These models were tested in the rest 30% of the data (test set) for validation. Performance of models was evaluated in comparison to best six existing LDL-C calculating formulas. Results: LDL-C predicted by XGBoost and random forests models showed a strong correlation with directly estimated LDL-C (r = 0.98). Two machine learning models performed superior to the six existing and commonly used LDL-C calculating formulas like Friedewald in the study population. When compared in different triglycerides strata also, these two models outperformed the other methods used. Conclusion: Machine learning models like XGBoost and random forests can be used to predict LDL-C with more accuracy comparing to conventional linear regression LDL-C formulas.
1 January 2016 in Journal of Clinical and Diagnostic Research
Journal of Clinical and Diagnostic Research aims to publish findings of doctors at grass root level and post graduate students, so that all unique medical experiences are recorded in literature.
Introduction This study was conducted to assess the analytical performance of biochemical tests using Six Sigma methodology and to assess the underlying causes of unsatisfied performance of analytes with a sigma value of less than 4 using quality goal index (QGI) and root cause analysis (RCA). Methodology Daily data for internal quality control (IQC) for both level 1 (L1) and level 2 (L2) and monthly data for external quality assessment for a period of six months were recorded. The coefficient of variation (CV), bias, and total allowable error (TEa) were calculated to analyze the sigma (sigma) values for 19 biochemical analytes. Quality goal index (QGI) analysis was done to analyze impressions and inaccuracies in analyte performance having a sigma value of less than 4. Root cause analysis (RCA) was done to evaluate the possible causes that can improve quality performance. Results Creatinine and high -density lipoprotein (HDL) had sigma metrics of <= 2.0, and chloride, aspartate aminotransferase (AST), and alkaline phosphatase (ALP) had sigma values between 2 and 3. Glucose, total protein (TP), phosphate (Phos), and potassium had sigma values between 4 and 5 in level 1 QC. Sigma grading for level 2 quality control (QC) also gave similar results. For analytes with sigma < 4, QGI analysis exposed inaccuracy or imprecision issues and identified errors such as the reconstitution of IQC, storage temperature, and air bubbles while processing the QC, being common causes of poor performance. Conclusion Six Sigma approach is helpful for quality assurance and identifying areas for improvement. Assessing Six Sigma metrics should be a routine practice to decide the frequency of QC run and to detect errors in analysis.
1 September 2018 in Journal of Clinical and Diagnostic Research
Journal of Clinical and Diagnostic Research aims to publish findings of doctors at grass root level and post graduate students, so that all unique medical experiences are recorded in literature.