Biomedical research is a field of study designed to improve knowledge about health. It is based on the collection of biological and/or medical information from volunteers. The ability of an extraction system to retrieve features using its feature descriptor is the first indication of its effectiveness. The pertinent feature descriptor to obtain features from two types of medical images is the Orthogonal Fourier Triangular Pulse Moments system (OFTPMs), due to its efficient performance compared to OMFMs and ROJMs. In my work, I have attained a percentage of NIRE value for colour medical images compared to ARP, ARR and F_SCORE. These values prove a great practical importance to the field of medical scanners (MRI). This paper uses two databases to retrieve images, which are briefly described as chest X-ray (pneumonia) and LIDC-IDRI datasets. The conducted experiments have proved that my OFTPF approach seems viable enough since it represents a clear destination compared to other techniques such as RZM, RALM and ROJM in terms of results and performance.