Data mining is an emanating area not only in the field of computer science but also contributes more to clinical data research. This analysis mainly focuses on medical diagnosis of both patients and drugs through classification and prediction techniques. Nowadays, liver disease is increasing dramatically among alcoholic consumer, smoker and person infected by hepatitis A, hepatitis C and Cirrhosis. This research work focused more on diagnosing patient with liver disorders. In this work, two different datasets namely, Bupa liver dataset and Indian liver patient disorder dataset are taken from UCI machine learning repository which contains 345 records and 745 records respectively for experimental study. Both these liver patient datasets is subjected to ReliefF filtering and stepwise discriminate feature relevance ranking algorithms before applying various classification algorithms. The comparative study on all these datasets is done to evolve the best classifier for diagnosing liver disease. The experimental results showed that Random tree algorithm, C4.5 and k-NN delivered accurate classification rules for identifying liver disorders on all these datasets The performance evaluation is done based on accuracy, precision, sensitivity and specificity for the above three selected classifiers revealed that Random tree classifier is the best classifier for diagnoses liver diseases.

This paper gives a brief introduction to a novel visual attention detection method, which combines features of luminance, color contrast and structural dissimilarity to detect salient regions in images. The method simultaneously considers regional contrast and spatial coherence information and outputs full resolution saliency maps. The main purpose of this paper is to improve robustness of salient object detection for complex texture images, highlight uniformly salient object and boost the overall precision and recall for saliency detection. The main contributions of this paper include: 1) Identifying structural dissimilarity based on internal regions of image to measure structure contrast, which is conducive to recognize salient objects from texture complex images. 2) A nonlinear fusion model, which effectively combines structural dissimilarity and color contrast to distinguish saliency of each image regions. 3) A salient objects segmentation method which can automatically extract salient objects from images based on saliency maps. The proposed method is compared with nine state-of-the-art salient region detection methods on fixed threshold segmentation experiments and on adaptive salient object segmentation experiments. The evaluated results demonstrate that our method has outstanding performances both on acquiring higher precision and better recall, especially improves saliency detection effect for complex texture image.

The performance of a person with DCD (Developmental Coordination Disorder) is lower than the person’s chronological age and measured intelligence regarding the daily activities. It may be manifested by delays in achieving motor milestones (such as walking, crawling and sitting), dropping things, clumsiness, poor performance in sports, and poor handwriting. The main goal of research is to assess the developmental coordination disorder. Descriptive and analytical methods are used in research through library studies to get the results. It was mainly resulted of the research that the pre-school years (ages 3 to 6) are the most ideal time to start rehabilitation to overcome the problems and disorders.

Glaucoma is the eye disorder in which the optic never suffers damage, leading to vision defect in the affected eye(s) and leading to complete blindness if untreated. It is frequently associated with increased pressure in the aqueous humour of the eye. Glaucoma often goes undetected until significant damage to the subject’s visual field has occurred. As glaucoma progresses, neural tissues die, the nerve fiber layer thins, and the cup-to-disc ratio increases. The conventional techniques typically used for this measurement are unreliable and creates intricacies while measuring considerably small changes in the nerve head geometry. In this paper spatially weighted fuzzy c-mean clustering is used to find the CDR from the colour fundus image and to determine pathological process of glaucoma. The blood vessels in the optic disc regions are segmented by using local entropy thresholding approach.The proposed method can be used to automatically segment the neuroretinal rim.We have extracted features such as cup to disc ratio,ratio of the blood vessels area in inferior-superior side to area blood vessels in the nasal-temporal side and the ratio of neuroretinal rim area in imferior-superior side to area of neuroretinal rim in the nasal-temporal side.These features are validated by classifying the normal and glaucoma images using Support vector machine, and neural network classifier.A batch of 300 retinal is used to assess the performance of proposed system and a classification rate of 96% is achieved