During the evolution of object-oriented legacy systems, improving the design quality is most often a highly demanded objective. For such systems which have a large number of classes and are subject to frequent modifications, detection and correction of design defects is a complex task. The use of automatic detection and correction tools can be helpful for this task. Various research approaches have proposed transformations that improve the quality of an object-oriented systems while preserving its behavior. This paper proposes a framework where a catalogue of object-oriented metrics can be used as indicators for automatically detecting situations where a particular transformation can be applied to improve the quality of an object-oriented legacy system. The correction process is based on analyzing the impact of various meta-pattern transformations on these object-oriented metrics.