Safety of health care processes is inherently limited by human factors and has contributed to increased medication errors. This has raised the necessity of developing computerized solutions that can play a supportive role in medical decision making process by minimizing the occurrence of human error. In this study, we experiment on the potential of Artificial Neural Networks (ANNs) and Decision Tree (DT) algorithms for analyzing a time-series, high-dimensional clinical data set and study the extent to which those techniques can be utilized to capture the medical expert knowledge integrated within the dataset. Two major empirical studies were conducted on the available clinical data set which includes the medical history of patients with chronic disorders who are getting clinical care on a periodic basis. As the first study, we studied the potential of supervised learning classifiers for correctly classifying a patient instance into the corresponding disease category. In the second experiment, potential of above mentioned supervised learning algorithms in detecting anomalous trends in a pattern of readings for a particular medical parameter was studied. Under each experiment, different learning classifier systems were built using multilayered perceptron (MLP) neural networks and C4.5 DT algorithm and the performance of each classifier was validated using confusion matrix analysis and receiver operating characteristic (ROC) curve analysis. Results obtained for the critical evaluation revealed that these kind of computational approaches have a substantial potential for playing an assistive role in medical decision making process.