The early detection of breast cancer makes manywomen survive. In this paper, a CAD system classifying breastcancer thermograms to normal and abnormal is proposed. Thisapproach consists of two main phases: automatic segmentationand classification. For the former phase, an improved segmen-tation approach based on both Neutrosophic sets (NS) andoptimized Fast Fuzzy c-mean (F-FCM) algorithm was proposed.Also, post-segmentation process was suggested to segmentbreast parenchyma (i.e. ROI) from thermogram images. For theclassification, different kernel functions of the Support VectorMachine (SVM) were used to classify breast parenchyma intonormal or abnormal cases. Using benchmark database, theproposed CAD system was evaluated based on precision, recall,and accuracy as well as a comparison with related work. Theexperimental results showed that our system would be a verypromising step toward automatic diagnosis of breast cancerusing thermograms as the accuracy reached 100% Thermogram breast cancer detection approach based on Neutrosophic sets and fuzzy c-means algorithm.. Available from: https://www.researchgate.net/publication/281377678_Thermogram_breast_cancer_detection_approach_based_on_Neutrosophic_sets_and_fuzzy_c-means_algorithm [accessed Jun 2, 2016].