Gray Hausdorff distance measure for medical image comparison in dermatology: Evaluation of treatment effectiveness by image similarity Panagiota Spyridonos 1, *, Georgios Gaitanis 2 , Ioannis D. Bassukas 2 and Margaret Tzaphlidou 1 1 Department of Medical Physics, University of Ioannina Medical School, University Campus, Ioannina, Greece and 2 Department of Skin and Venereal Diseases, University of Ioannina Medical School, University Campus, Ioannina, Greece Abstract: Introduction: In clinical dermatology, the stabiliza- tion of the overall skin condition can be in many cases the ear- liest qualitative measure of the effectiveness of the therapeutic intervention. Subjective image comparisons, that offer empiri- cal ‘qualitative’ judgments of degrees of image similarities, are traditionally employed by the involved physicians. Objectives: To quantify, by means of an image similarity met- ric, the degree of stabilization of an expanding skin disease, and to identify the situation of ‘no further change’ of the skin condition of the patient, providing thus the physician with an early, objective measure of the efficacy of the used therapy. Methods: For treatment assessment, a variant of gray Hausdorff distance metric was employed to compare images of lesional skin segments of a patient, taken at different time points during a therapeutic course. Prior to image comparison, an effective preprocessing scheme was adapted to constrain wide pose and light variations. The proposed similarity algo- rithm was tested on raw clinical image data sets of patients diagnosed with toxic epidermal necrolysis, a life-threatening condition with rapid evolution. Fine tuning of algorithm’s parameters was optimized using Precision-Recall curves. Results: Proposed image comparison method resulted in a high-degree of image similarity (about 96%) between pictures taken at second and fifth day of hospitalization. Current similar- ity results substantiate a significant agreement between the computer-treatment assessment, by means of image compari- son, and the corresponding clinical experts’ review of skin con- dition. Conclusion: Objective evidence of ‘no further change’ situa- tion may provide (a) intuitive clinical decision support to derma- tologists in assessing aggressive skin conditions, where the timely evaluation of treatment response is of vital importance and (b) a versatile end-point measure for corresponding thera- peutic clinical trials. Key words: Image similarity – Hausdorff Distance – Toxic Epidermal Necrolysis – Treatment evaluation Ó 2012 John Wiley & Sons A/S Accepted for publication 9 August 2012 I N MEDICINE, the timely assessment of patient’s response to an ongoing therapeutic inter- vention is of vital importance for the clinical evaluation of a therapeutic scheme. Particularly in dermatology, the stabilization of the overall skin condition can be in many cases the earliest qualitative measure of the effectiveness of the therapeutic intervention. Subjective image comparisons, that offer empirical ‘qualitative’ judgments of degrees of image similarities, are traditionally employed by the involved physicians. For medical uses, a quantitative approach to image similarity has been primarily used in image registration problems as a criterion of best matching (1). Currently, with large image databases becoming a reality in many medical domains, image matching is an active research area in content-based image retrieval systems (2–5). Especially, within the framework of image similarity analysis applications to skin dis- eases, to the best of our knowledge, most pre- vious work focused on mole evaluation: Several studies have proposed a computerized comparison of serial skin images aiming to assist nevi diagnosis and melanoma identifica- tion (6–9). In practice, these previous studies applied matching or ‘registering’ single lesions or selected skin areas in a pair of metachro- nous images mainly looking for clinically sig- nificant changes, i.e. evidence for emerging new lesions or certain alterations in already existing ones. 1 Skin Research and Technology 2012; 0:1–9 Printed in Singapore  All rights reserved doi: 10.1111/srt.12001 © 2012 John Wiley & Sons A / S Skin Research and Technology