This paper is concerned with kernel-based techniques forcategorizing laryngeal disorders based on information extracted fromsequences of laryngeal colour images. The features used tocharacterize a laryngeal image are given by the kernel principalcomponents computed using the $N$-vector of the 3-D colourhistogram. The least squares support vector machine (LS-SVM) isdesigned for categorizing an image sequence into the healthy, nodular and diffuse classes. The kernel functionemployed by the SVM classifier is defined over a pair of matrices, rather than over a pair of vectors. An encouraging classificationperformance was obtained when testing the developed tools on datarecorded during routine laryngeal videostroboscopy.