The most common problem in the literature is the clustering of incomplete data sets containing missing features. Several imputations and non-imputations are used to solve this problem. It is found that the kernel-based cluster strategies are more accurate than the conventional ones. This research compares kernel-based cluster algorithms for incomplete data with traditional cluster algorithms. To create kernel-based clusters, the kernel functions are essential. Selecting a kernel function is not an easy task. In the various kinds of kernel-based cluster approaches examined in the literature, Gaussian kernel functions are thought to be more significant. The Gaussian kernel function has been explored for the management of incomplete medical data in this study. This work offers a thorough examination of kernel fuzzy c-mean clustering for incomplete medical data and is conducted on the Thyroid and Wisconsin breast cancer dataset from the UCI repository, which naturally contains missing values.