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.

Molecular Mechanisms of Frontotemporal Lobar Degeneration

The aim of this thesis was to identify genetic factors involved in frontotemporal lobar degeneration (FTLD), a neurodegenerative disorder clinically characterised by a progressive change in personality, behaviour and language. FTLD is a genetically complex disorder and a positive family history is found in up to 40% of the cases. In 10-20% of the familial cases the disease can be explained by mutations in the gene encoding the microtubule associated protein tau (MAPT). In the first study we describe the clinical and neuropathological features of a Finnish family with FTLD caused by a mutation in MAPT. We also provide evidence that the pathogenic mechanism of this mutation is through altered splicing of MAPT transcripts. Recently, mutations in the gene encoding progranulin (PGRN) were identified as a major cause of FTLD. In the second study we describe a Swedish family with FTLD caused by a frameshift mutation in PGRN. We provide a clinical and neuropathological description of the family, as well as evidence that the pathogenicity of this mutation is through nonsense-mediated decay of the mutant mRNA transcripts and PGRN haploinsufficiency. In the third study we describe a novel PGRN splice site mutation and a previously described PGRN frameshift mutation, found in a mutation screen of 51 FTLD patients. We describe the clinical and neuropathological characteristics of the mutation carriers and demonstrate that haploinsufficiency is the pathogenic mechanism of the two mutations. In the fourth study we investigate the prevalence of PGRN and MAPT gene dosage alterations in 39 patients with FTLD. No gene dosage alterations were identified, indicating that variations in copy number of the PGRN and MAPT genes are not a common cause of disease, at least not in this FTLD patient collection.