Faculty of Medicine
Permanent URI for this communityhttps://repository.ukim.mk/handle/20.500.12188/14
Browse
8 results
Search Results
- Some of the metrics are blocked by yourconsent settings
Item type:Publication, Perspectives from the Balkans(IOP Publishing, 2022-06) - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Comparing Time and Frequency Domain Heart Rate Variability for Deep Learning-Based Glucose Detection(Springer International Publishing, 2022-04-12); ; ; International Conference on ICT Innovations ICT Innovations 2021: ICT Innovations 2021. Digital Transformation pp 188–197Cite as Comparing Time and Frequency Domain Heart Rate Variability for Deep Learning-Based Glucose Detection Ervin Shaqiri, Marjan Gusev, Lidija Poposka, Marija Vavlukis & Irfan Ahmeti Conference paper First Online: 12 April 2022 165 Accesses Part of the Communications in Computer and Information Science book series (CCIS,volume 1521) Abstract Many researchers have been challenged by the usage of devices related to the Internet of Things in conjunction with machine learning to anticipate and diagnose health issues. Diabetes has always been a problem that society has struggled with. Due to the simplicity with which electrocardiograms can be captured and analysed, deep learning can be used to forecast a patient’s instantaneous glucose levels. Our solution is based on a unique method for calculating heart rate variability that involves segment identification, averaging, and concatenating the data to reveal better feature engineering results. Immediate plasma glucose levels are detected using short-term heart rate variability and applied a deep learning method based on Autokeras. In this paper. we address a research question to compare the predictive capability of time and frequency domain features for instantaneous glucose values. The neural architectural search for the time domain approach provided the best results for the 15-min electrocardiogram measurements. Similarly, the Frequency Domain approach showed better results on the same time frame. Regarding the time domain the best results are as follows: RMSE (0.368), MSE (0.193), R2 (0.513), and R2 loss (0.541). The best results for the Frequency Domain approach the best results are as follows: RMSE (0.301), MSE (0.346), R2 (0.45578), and R2 loss (0.482). - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Public Health Aspects of Non-ionizing Radiation in Health determinants in the scope of new public health(Jacobs Verlag, Lage, Germany, 2005)Exposure to electromagnetic fields is not a new phenomenon. However, dur ing the 20th century, numerous man-made electromagnetic field sources used for individual, industrial and commercial purposes and in medicine have become the focus of the public health interest. All those new and advancing technologies, including power lines, microwave ovens, computer and TV screens, security de vices, radars, mobile cellular phones and their base stations, have made our life richer and easier. At the same time, they have brought with them concerns about possible health risks associated with their use, such as cancer, reduced fertility, memory loss, changes in the behaviour and development of children etc. In response to growing public health concerns over possible health effects from exposure to the electromagnetic field sources, in 1996 the World Health Organization (WHO) launched a large, multidisciplinary International Electro magnetic Field (EMF) Project. This project brings together current knowledge and available resources of key international and national agencies and scientific institutions (1). Despite the feeling of some people that more research needs to be done, the WHO and many other experts concluded that the current evidence does not con firm the existence of any health consequences from exposure to low level electro magnetic fields. There are some gaps in knowledge about the biological effects from long-time exposure and exposure to high levels, so more research is needed in these areas (2). The conflict between concerns about possible health effects from exposure to EMF and the development of electricity supply and telecommunications facilities has led to considerable economic consequences. But the lack of knowledge about the health consequences of technological advances may not be the sole reason for social opposition to innovations and further progress. - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Strengthening the Early-Warning Function of the Surveillance System: The Macedonian Experience(Springer Netherlands, 2010); ; Karadzovski, Zarko - Some of the metrics are blocked by yourconsent settings
Item type:Publication, - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Correlating Glucose Regulation with Lipid Profile(Springer International Publishing, 2020); ; Objectives: The goal of this research was to detect the glucose regulation class by evaluating the correlation between the lipid profile of patients and their glucose regulation class. Methodology: The methods used in this research are: i) Point Biserial Correlation, ii) Univariate Logistic Regression iii) Multivariate Logistic Regression iv) Pearson Correlation and v) Spearman Rank correlation. Data: The dataset consists of the following features: age, BMI, gender, weight, height, total cholesterol (Chol), HDL cholesterol (HDL-C), LDL cholesterol (LDL-C), triglycerides (TG), glycated hemoglobin (HbA1C), glucose regulation and diabetes classes, history of diabetes, heart and other chronic illnesses, habitual behaviors (smoking, alcohol consumption, physical activity), and medications intake (calcium channel blockers, BETA blockers, anti-arrhythmic, AKE/ARB inhibitors, diuretics, statins anti-aggregation medication and anticoagulants). Conclusion: The methodologies that were worked through with our data in search for correlations of the lipid profile with HbA1c or the glucose regulation classes gave some significant correlations. Regarding the glucose regulation classes W and B the methods showed statistically significant negative correlations with Chol, HDL-C and LDL-C. When it comes to the correlations of the lipid profile with HbA1c, for all patients there were significant negative correlations with Chol (corr = −0.264, p = 0.002), LDL-C (corr = −0.297, p < 0.001) and HDL-C (corr = −0.28, p = 0.001) and a significant positive correlation with TG (corr = 0.178, p = 0.03). The correlations mentioned are the stronger ones that were found for linear relationships. For non-diabetic patients there was a stronger positive non-linear correlation for HbA1c and HDL-C (corr = 0.511, p = 0.006), and a slightly weaker linear correlation (corr = 0.393, p = 0.043). For prediabetic patients there were no significant correlations. For type 2 diabetes stronger significant negative non-linear correlations were found for HbA1c with LDL-C (corr = −0.299, p = 0.023) and HDL-C (corr = −0.438, p = 0.001). The linear relationships were again, slightly weaker with LDL-C (corr = −0.273, p = 0.038) and with HDL-C (corr = −0.391, p = 0.002). - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Tachycardia: Risk Factors, Causes and Treatment Options (Cardiology Research and Clinical Developments) 1st Edition(Nova Science Publishers, Inc, 2015) - Some of the metrics are blocked by yourconsent settings
Item type:Publication,