Search

A 47-year-old female presented with Easy fatigability, abdominal pain for 4 days. Past history of open cholecystectomy with Common bile duct exploration and Choledocho Duodenostomy done 13 years before. She later had multiple episodes of abdomen pain with fever on and off. ERCP revealed stone in common bile duct but not retrieved and patient lost follow-up. During this hospitalisation patient had fever, Jaundice and Sepsis with Acute Kidney Injury. Ultrasound Abdomen showed Hypoechoic lesion in seg IV left lobe liver. Mild pneumobilia. CECT Abdomen showed Pneumobilia, ill-defined hypodense lesions in segment IVA, II, III causing left lobe Intrahepatic biliary radicle dilation. MRCP showed Multiple tiny T2 hyperintense lesions with surrounding edema in SEGMENT IVB? Cholangitic abscess. Patient was taken up for surgery. Intraoperatively Left lateral segment II, III fibrotic, left hepatic duct & branches grossly dilated, stones retrieved from LHD; Left lateral sectionectomy, hepaticojejunostomy, jejunojejunostomy with access loop done. This was a rare case of recurrent pyogenic cholangiohepatitis managed surgically.

There are many solutions for detection and classification and classification of dementia into its corresponding patterns using PET, SPECT, MRI, fMRI addressing to the classification based on the patterns. They are classified based on the corresponding diseased and Non-diseased Syndrome types. Here in this scope, solution anticipated to the inter- related disease with Picks Syndrome (PD) and Alzheimers Disease Syndrome (AD). The image data sets implied her are the functional Magnetic Resonance Images. The images become very contrasting based on many parameters - capturing intensity, noise content, geographical locations, Image acquisition based on the different corporate machines, capability of the Radiologist, the moment during the motion capturing, patient nature, etc., and hence many more features that comes into the account. Ultimately this effects the intensity and the noise level imposed during the motion capturing. A Daubechies Wavelet Transform is implied here. Its properties are: 1) Smoothening filters 2) extraction of frequency (feature) components 3) Median filter - removing the random noises better. A fourth level of extraction of features using the Wavelet Decomposition by Daubechies type was implied for formation of the feature vectors. From the sub-band regions of H-H level order with fourth level of decomposition level which will have high resolution was chosen for this act. Other have low-high combinatory, and fourth band has low - low combination which are very difficult to pick up the resolution points. This obtained features were implied for feature vector formation. The extraction of the features was obtained from the approximated co-efficients got from the Approximated sub-band levels thus chosen to form the input images and extracted co-efficients were fed to design, train and test the network for convergence, network framing and efficiency estimation respectively. A Self Organizing Map technique was chosen for designing up of the network using the Competitive Neural Network. This was used to classify the normal and Dementia based image types into its corresponding types. This enables a good proviso for designing up of the network and also provides efficiency in convergence of the network, classification of the two types of diseases and also in classification of the disease types into its individual inter dependent types such as Demential and Non-Demential types.

This propaganda of research implied in this progress would be arised by introducing the clustering scheme in a very poignant distribute manner. This method is very high responsive enough in order to bring up an efficient Cluster acting as the base for the designing up of the Routing protocol for the Delay Tolerant Mobile Networks. This research proposes with an approach that they are clustered into a group. The design of the kind of networks is as: 1) Distributively grouped into mobile nodes 2) The grouping up of the nodes is based on similarity pattern. 3) The similarity of the pattern is it should have a definite cluster of fixed size. The result of such implications is to - reduce the overheads based on the parameters and its constraints, secondly – balancing of the loads implied. They share the resources and perform the same. This would enhance and aim in bringing out an efficient and a enhanced routing scaled up to its standard and hence the routing is aimed to achieve a routing protocol based on exponentially Weighted moving area shortly know an as the EWMA. This one gives the details about the nodal contact it occurs and mathematical probability amongst the nodes themselves, using the weighted mean with the clusters formed. The clusters what were formed using the functions such as Sync (), Leave () and Join (). The defined number of nodes in a particular cluster need to be defined and must be greater or exceeding the optimum value the of nodes allotted for a particular cluster. Finally, some of the gateway nodes group the network information form the other gateways and determines a routing table which are all required in that particular nodes or stipulated nodes in that cluster, so as to determine the moving weighted average and perform the routing process. The results obtained from the simulations revealed the static clusters for improving the sharing of the resource capacity and decreases the memory requirement of then nodes, thus making the computations of the Routing table gets easier for access and at a faster pace too. The consistency of clusters gets improved by making the moving node to inform to its neighbours about its movement from the cluster. The use of Cluster connectivity modification packet (CCMP) is that it would bring down the traffic of the network and would calculate the computation table for shortest path route easier. From the above, one would be able to obtain greater delivery ratio values at a predominantly lesser overhead and the delay in the nodes when reaches end to end. They are poignant enough in achieving this feat when comparing the results obtained from the non – clustered algorithms.