Headspace-solid-phase microextraction-gas chromatography (HS-SPME-GC) was used to identify in total 74 volatile lipid oxidation compounds altogether in thermally stressed conventional and high-oleic sunflower (HOSF) oil samples (in accelerated storage conditions for 14 days at 80°C). Out of the volatile compounds identified, six volatile compounds were selected as marker compounds for the assessment of lipid oxidation of sunflower (SF) and HOSF oils due to their low odour threshold values and fatty-rancid odour impression. Additionally, other oxidation parameters such as fatty acid composition, peroxide value (PV), anisidine value and tocopherol and tocotrienol composition were determined. Multivariate statistical methods (principal component analysis and agglomerative hierarchical cluster analysis) were applied to identify sensitive oxidation marker compounds. Preliminary results revealed that hexanal, E-2-heptenal, E-2-decenal and E,E-2,4-nonadienal were the most suitable in differentiating HOSF and SF oil varieties from each other and SF samples with differing oxidative properties. Differentiation of SF samples according to their volatile compound composition was done in accordance with the results from the well-known oil quality parameters (e.g. PV or fatty acid composition). In conclusion, the combination of volatile compound analysis with HS-SPME-GC and multivariate statistical methods provides a sensitive tool in differentiating conventional SF and HOSF oils by means of volatile lipid oxidation marker compounds.