Unsupervised learning to cluster the disease stages in parkinson's disease

Parkinson's disease (PD) is the second most common neurodegenerative disorder (after Alzheimer's disease) and directly affects upto 5 million people worldwide. The stages (Hoehn and Yaar) of disease has been predicted by many methods which will be helpful for the doctors to give the dosage according to it. So these methods were brought up based on the data set which includes about seventy patients at nine clinics in Sweden. The purpose of the work is to analyze unsupervised technique with supervised neural network techniques in order to make sure the collected data sets are reliable to make decisions. The data which is available was preprocessed before calculating the features of it. One of the complex and efficient feature called wavelets has been calculated to present the data set to the network. The dimension of the final feature set has been reduced using principle component analysis. For unsupervised learning k-means gives the closer result around 76% while comparing with supervised techniques. Back propagation and J4 has been used as supervised model to classify the stages of Parkinson's disease where back propagation gives the variance percentage of 76-82%. The results of both these models have been analyzed. This proves that the data which are collected are reliable to predict the disease stages in Parkinson's disease.

The kidney in different stages of the cardiovascular continuum

Patients with chronic kidney disease are at higher risk of developing cardiovascular disease. The complex, interaction between the kidney and the cardiovascular system is incompletely understood, particularly at the early stages of the cardiovascular continuum. The overall aim of this thesis was to clarify novel aspects of the interplay between the kidney and the cardiovascular system at different stages of the cardiovascular continuum; from risk factors such as insulin resistance, inflammation and oxidative stress, via sub-clinical cardiovascular damage such as endothelial dysfunction and left ventricular dysfunction, to overt cardiovascular death. This thesis is based on two community-based cohorts of elderly, Uppsala Longitudinal Study of Adult Men (ULSAM) and Prospective Investigation of the Vasculature in Uppsala Seniors (PIVUS). The first study, show that higher insulin sensitivity, measured with euglycemic-hyperinsulinemic clamp technique was associated to improve estimated glomerular filtration rate (eGFR) in participants with normal fasting plasma glucose, normal glucose tolerance and normal eGFR. In longitudinal analyses, higher insulin sensitivity at baseline was associated with lower risk of impaired renal function during follow-up. In the second study, eGFR was inversely associated with different inflammatory markers (C-reactive protein, interleukin-6, serum amyloid A) and positively associated with a marker of oxidative stress (urinary F2-isoprostanes). In line with this, the urinary albumin/creatinine ratio was positively associated with these inflammatory markers, and negatively associated with oxidative stress. In study three, higher eGFR was associated with better endothelial function as assessed by the invasive forearm model. Further, in study four, higher eGFR was significantly associated with higher left ventricular systolic function (ejection fraction). The 5th study of the thesis shows that higher urinary albumin excretion rate (UAER) and lower eGFR was independently associated with an increased risk for cardiovascular mortality. Analyses of global model fit, discrimination, calibration, and reclassification suggest that UAER and eGFR add relevant prognostic information beyond established cardiovascular risk factors in participants without prevalent cardiovascular disease. Conclusion: this thesis show that the interaction between the kidney and the cardiovascular system plays an important role in the development of cardiovascular disease and that this interplay begins at an early asymptomatic stage of the disease process.