<p>Modern computerized systems for diagnosing skin neoplasms are mainly focused on providing recommendations to patients, but the application of such systems in the clinical practice remains limited. This is related to the lack of qualitative studies of such systems and low level of trust in non-transparent mechanisms of their operation by the physicians. This challenge can be solved by creating a&#xa0;clinical decision support system based on the logic of physician’s diagnostic search. An important task of the clinical decision support system is to recognize the globule color of skin neoplasms, however, no methods for addressing this task have been described in scientific publications to date. The application of the method of automated globule color recognition on dermoscopic images of skin neoplasms is analyzed, which enables a&#xa0;globular recognition based on color in accordance with a&#xa0;7-color palette (blue, white-yellow, brown, red, orange, nude, and black). An original set of 9&#xa0;color features has been developed as part of this method. The Random Forest method was applied to classify the images based on this feature (globule color). According to the experimental results obtained by using a&#xa0;sample of 313 images, the classification accuracy was 91%. The developed method can be implemented as a&#xa0;software within the framework of a&#xa0;modified pattern analysis algorithm. In addition, this method can be used as part of a&#xa0;clinical decision support system when diagnosing skin cancer.</p>