Shape Analysis Approach towards Assessment of Cleft Lip Repair Outcome ⋆ Paul Bakaki 1,4[0000-0001-8277-2554] , Bruce Richard 2[0000-0002-4712-2860] , Ella Pereira 1[0000-0002-6013-3935] , Aristides Tagalakis 1[0000-0002-4610-0803] , Andy Ness 3[0000-0003-3548-9523] , and Yonghuai Liu 1[0000-0002-3774-2134] 1 Faculty of Arts and Science, Edge Hill University, Lancashire L39 4QP, UK {bakakip, pereirae, Aristides.Tagalakis, yonghuai.liu}@edgehill.ac.uk 2 Birmingham Children’s Hospital, Steelhouse Lane Birmingham B4 6NH, UK brucerichard@blueyonder.co.uk 3 British Dental School, University of Bristol, Bristol BS1 2LY, UK Andy.Ness@bristol.ac.uk 4 Department of Computer Science, Makerere University, P.O. Box 7062, Kampala, Uganda Abstract. Current methods of assessing the quality of a surgically re- paired cleft lip rely on humans scoring photographs. This is only practical for research purposes due to the resources necessary and is not used in routine audit. It has poor valid- ity due to human subjectivity and thus low inter-rater reliability. An automatic method for aesthetic outcome assessment of cleft lip repair is required. The appearance and shape of the lips constitute the region of interest for analysis. The mouth borderline and corner points are de- tected using a bilateral semantic network for real-time segmentation. The bisector of the line linking the mouth corners is estimated as the vertical symmetric axis. By splitting the mouth blob into two parts, they are analyzed for similarity and a numeric score ranging from 1 to 5 is then generated. Pearson correlation coefficient between automatically gener- ated scores and human-assigned ones serves as a validation metric. A correlation of about 40% indicates a good agreement between human and computer-based assessments. However, better automatic scoring correla- tion of 95.9% exists between the automatically detected mouth regions and those manually drawn by human experts, the third ground truth set in scenario two. Our method has the potential to automate an outcome estimation of the aesthetics of cleft lip repair with human bias reduced, easy implementation and computational efficiency. Keywords: Cleft Lip, Aesthetic Assessment, Segmentation, Symmetry, Structural Similarity, Correlation coefficient. 1 Introduction Cleft lip (CL) is one of the most common maxillofacial congenital deformities affecting about 1 in 500 Asians, 1 in 1000 Caucasians and 1 in 2500 Africans [1]. ⋆ Supported by Graduate Teaching Assistantship, Edge Hill University.