These researchers trace a few of the modern development in the field of learning imbalanced data. Review approaches were adopted for this problem and it identifies challenges and points out potential directions in this comparatively new field. In medical province, data features frequently contain missing values. This can make grave bias in the logical modeling. Characteristic standard data mining methods often produce poor performance measures. In this paper, the authors proposed a new method to concurrently classify large datasets and decrease the belongings of missing values. The proposed method is based on a multilevel structure of the cost-sensitive ASVM (Adaptive Support Vector Machine) and the probable maximization charge method for missing principles, which learn the breakdown analysis of excessive dataset. Thus the authors developed the PBI²D- (Priority Based Intelligent Imbalanced Data Classification) of HealthCare data with missing values to produce contrast classification results of multilevel ASVM-based algorithms on public benchmark datasets with imbalanced classes and missing values as well as real data in health applications. This method produces fast, more accurate and robust classification results.