This paper reviews the Case-based reasoning (CBR) approach and its usability in the medicine and presents a new concept on how to improve its adaptation phase. We use the CBR as a supporting method for decision support like diseases diagnostics or therapy identification. We investigated existing approaches, studies, and research works to solve one of the most critical problems in the CBR cycle - adaptation, which is often done manually by the experts in the relevant field. Based on the findings and our experiences with medical diagnostics through suitable data analytical methods, we proposed a new solution to solve this challenge. This approach is based on a comparison of the stored decision rules with the new one related to the current case. This comparison can result in three alternative states: (1) case base contains a similar case, and relevant rule can be applied. (2) The new case is very different from the stored ones, so the input from participated experts is needed, and a new rule will be stored. (3) The new case is partially similar satisfying adaptability conditions, in such a situation we adopt related decision rule to the new conditions under the supervision of the expert. We plan to experimentally test and verify this concept within available medical samples from our previous experiments.