An intuitive risk factors search algorithm: usage of the Bayesian network technique in personalized medicine

Machine learning for predictive data analytics in medicine: A review illustrated by cardiovascular and nuclear medicine examples

The evidence-based medicine allows the physician to evaluate the risk-benefit ratio of a treatment through setting and data. Risk-based choices can be done by the doctor using different information. With the emergence of new technologies, a large amount of data is recorded offering interesting perspectives with machine learning for predictive data analytics. Machine learning is an ensemble of methods that process data to model a learning problem. Supervised machine learning algorithms consist in using annotated data to construct the model. This category allows to solve prediction data analytics problems. In this paper, we detail the use of supervised machine learning algorithms for predictive data analytics problems in medicine. In the medical field, data can be split into two categories: medical images and other data. For brevity, our review deals with any kind of medical data excluding images. In this article, we offer a discussion around four supervised machine learning approaches: information-based, similarity-based, probability-based and error-based approaches. Each method is illustrated with detailed cardiovascular and nuclear medicine examples. Our review shows that model ensemble (ME) and support vector machine (SVM) methods are the most popular. SVM, ME and artificial neural networks often lead to better results than those given by other algorithms. In the coming years, more studies, more data, more tools and more methods will, for sure, be proposed.

Prediction of features of the course of chronic hepatitis C using Bayesian networks

MATERIALS AND METHODS

253 patients with chronic hepatitis C (CHC) and liver cirrhosis were included in the study. Assessment of gene polymorphisms of genes involved in inflammatory reactions and antiviral immunity (IL-1β-511C/T, IL-10 -1082G/A, IL28B C/T, IL28B T/G, TNF-α -238G/A, TGF-β -915G/C, IL-6 -174G/C), activators of local hepatic fibrosis (AGT G-6A, AGT 235 M/T, ATR1 1166 A/C), hemochromatosis (HFE C282Y, HFE H63D), platelet receptors (ITGA2 807 C/T, ITGB3 1565 T/C), coagulation proteins and endothelial dysfunction (FII 20210 G/A, FV 1691G/A, FVII 10976 G/A, FXIII 103 G/T, eNOS 894 G/T, CYBA 242 C/T, FBG -455 G/A, PAI-675 5G/4G, MTHFR 677 C/T) was carried. Using Bayesian networks we studied the predictor value of clinical and laboratory factors for the following conditions - end points (EP): development of cirrhosis (EP1), fibrosis rate (EP2), presence of portal hypertension (EP3) and cryoglobulins (EP4).

RESULTS AND DISCUSSION

In addition to traditional factors we have shown the contribution of the following mutations. Predicting EP1- liver cirrhosis - HFE H63D, C282Y, CYBA 242 C/T, AGT G-6G, ITGB31565 T/C gene mutations were significant. We also found a link between the rate of progression of liver fibrosis and gene polymorphisms of AGT G-6G, AGT M235T, FV 1691G/A, ITGB31565 T/C. Among the genetic factors associated with portal hypertension there are gene polymorphisms of PAI-I-675 5G/4G, FII 20210 G/A, CYBA 242 C/T, HFE H63D and Il-6 174GC. Cryoglobulins and cryoglobuliemic vasculitis (EP4) are associated with gene mutations MTHFR C677T, ATR A1166C and HFE H63D.

CONCLUSION

The results obtained allow to detect the major pathophysiological and genetic factors which determine the status of the patient and the outcome of the disease, to clarify their contribution, and to reveal the significance of point mutations of genes that control the main routes of HCV course and progression.