Machine learning based assessment of preclinical health questionnaires.

BACKGROUND: Within modern health systems, the possibility of accessing a large amount and a variety of data related to patients' health has increased significantly over the years. The source of this data could be mobile and wearable electronic systems used in everyday life, and specialized medical devices. In this study we aim to investigate the use of modern Machine Learning (ML) techniques for preclinical health assessment based on data collected from questionnaires filled out by patients. METHOD: To identify the health conditions of pregnant women, we developed a questionnaire that was distributed in three maternity hospitals in the Mureș County, Romania. In this work we proposed and developed an ML model for pattern detection in common risk assessment based on data extracted from questionnaires. RESULTS: Out of the 1278 women who answered the questionnaire, 381 smoked before pregnancy and only 216 quit smoking during the period in which they became pregnant. The performance of the model indicates the feasibility of the solution, with an accuracy of 98 % confirmed for the considered case study. CONCLUSION: The proposed solution offers a simple and efficient way to digitize questionnaire data and to analyze the data through a reduced computational effort, both in terms of memory and computing power used.

Platelet-to-Albumin Ratio: The Prognostic Utility in the Prediction of 2-Month Postoperative Heart Transplant Complications.

BACKGROUND: The platelet-to-albumin ratio (PAR), leucocyte-to-albumin ratio (LAR), neutrophil percentage-to-albumin ratio (NPAR), and monocyte-to-albumin ratio (MAR) represent easily reproducible markers, which may predict the outcomes in various diseases. Early postoperative complications might appear after heart transplantation, such as infections, diabetes mellitus type 2 (DM2), acute graft rejection, and atrial fibrillation (AFib). OBJECTIVE: The aim of our study was to investigate the PAR, LAR, NPAR, and MAR values before and after heart transplantation, and the associations of the preoperative levels of these markers with the presence of postoperative complications in first two months after surgery. METHODS: Our retrospective research was directed from May 2014 to January 2021, with a total number of 38 patients being included. We used cut-off values for the ratios from previously published studies, as well as our own determination of these levels by using a receiver operating characteristic (ROC) curve. RESULTS: By ROC analysis, the optimal preoperative PAR cut-off value was 38.84 (AUC: 0.771, p = 0.0039), with 83.3% sensitivity, and 75.0% specificity. Applying a Chi square (χ2) test, PAR > 38.84 represented an independent risk factor for complications, regardless of cause, and postoperative infections. CONCLUSIONS: Preoperative PAR > 38.84 was a risk factor of developing complications of any cause, and postoperative infections in the first two months after heart transplantation.

Elevated Levels of Neutrophil-to Monocyte Ratio Are Associated with the Initiation of Paroxysmal Documented Atrial Fibrillation in the First Two Months after Heart Transplantation: A Uni-Institutional Retrospective Study.

BACKGROUND: Heart transplantation represents the treatment for patients with end-stage heart failure (HF) being symptomatic despite optimal medical therapy. We investigated the role of NMR (neutrophil-to-monocyte ratio), NLR (neutrophil-to-lymphocyte ratio), NPR (neutrophil-to-platelet ratio), NWR (neutrophil-to-white cells ratio), MLR (monocyte-to-lymphocyte ratio), PLR (platelet-to-lymphocyte ratio), MWR (neutrophil-to-white cells ratio), and LWR (lymphocyte-to-white cells ratio) at the same cut-off values previously studied, to predict complications after heart transplant within 2 months after surgery. METHODS: From May 2014 to January 2021, was included 38 patients in our study from the Cardiovascular and Transplant Emergency Institute of Târgu Mureș. RESULTS: Preoperative NMR > 8.9 (OR: 70.71, 95% CI: 3.39-1473.64; p = 0.006) was a risk factor for the apparition of post-operative paroxysmal atrial fibrillation (Afib). In contrast, preoperative MWR > 0.09 (OR: 0.04, 95% CI: 0.003-0.58; p = 0.0182) represented a protective factor against AFib, but being the risk of complications of any cause (OR: 14.74, 95% CI: 1.05-206.59, p = 0.0458). CONCLUSION: Preoperative elevated levels of NMR were associated with the apparition of documented AFib, with high levels of MWR as a protective factor. High MWR was a risk factor in developing complications of any cause in the first 2 months after heart transplantation.

Active Smoking and Associated Behavioural Risk Factors before and during Pregnancy - Prevalence and Attitudes among Newborns' Mothers in Mures County, Romania.

BACKGROUND AND AIM: Smoking before, during and after pregnancy leads to detrimental outcomes on maternal and foetal health and represents an important public health issue. This study aims to evaluate the prevalence and correlates of smoking before and during pregnancy in a sample of Romanian women. METHODS: A cross-sectional survey was conducted among mothers (N=1,278) in three maternity hospitals in Tirgu-Mures, Romania, immediately after childbirth, in 2014. We evaluated the prevalence of smoking before and during pregnancy and used binary logistic regression to assess the influence of socio-demographics and other health behaviour factors in three groups of women: non-smoking pregnant women, women who continued smoking during pregnancy, and smokers who quit during pregnancy. RESULTS: 30% of the interviewed mothers were smokers prior to pregnancy, of whom 43.3% continued smoking during pregnancy. Women with a family income of less than 100 Euro/month (OR=3.01, 95% CI: 1.02-8.83) and those who were unemployed (OR=13.2, 95% CI: 3.90-44.79) had increased odds of continued smoking versus quitting during pregancy in multivariable analyses. Women who continued smoking during pregnancy were also more likley to be of lower socioeconomic status than never smokers (OR=14.1, 95% CI: 4.97-39.6). CONCLUSIONS: A high percentage of women of reproductive age smoke and continue to smoke despite their knowledge about risks of smoking during pregnancy. Smoking prior to and during pregnancy is predominantly associated with lower socioeconomic status. Women with limited economic means should be a high priority target group for smoking cessation interventions.