Computer-Aided Diagnosis of Skin Lesions Using Conventional Digital Photography: A Reliability and Feasibility Study

Background Computer-aided diagnosis (CADx) software that provides a second opinion has been widely used to assist physicians with various tasks. In dermatology, however, CADx has been mostly limited to melanoma or melanocytic skin cancer diagnosis. The frequency of non-melanocytic skin cancers and the accessibility of regular digital macrographs have raised interest in developing CADx for broader applications. Objectives To investigate the feasibility of using CADx to diagnose both melanocytic and non-melanocytic skin lesions based on conventional digital photographic images. Methods This study was approved by an institutional review board, and the requirement to obtain informed consent was waived. In total, 769 conventional photographs of melanocytic and non-melanocytic skin lesions were retrospectively reviewed and used to develop a CADx system. Conventional and new color-related image features were developed to classify the lesions as benign or malignant using support vector machines (SVMs). The performance of CADx was compared with that of dermatologists. Results The clinicians' overall sensitivity, specificity, and accuracy were 83.33%, 85.88%, and 85.31%, respectively. New color correlation and principal component analysis (PCA) features improved the classification ability of the baseline CADx (p = 0.001). The estimated area under the receiver operating characteristic (ROC) curve (Az) of the proposed CADx system was 0.949, with a sensitivity and specificity of 85.63% and 87.65%, respectively, and a maximum accuracy of 90.64%. Conclusions We have developed an effective CADx system to classify both melanocytic and non-melanocytic skin lesions using conventional digital macrographs. The system's performance was similar to that of dermatologists at our institute. Through improved feature extraction and SVM analysis, we found that conventional digital macrographs were feasible for providing useful information for CADx applications. The new color-related features significantly improved CADx applications for skin cancer.

[1]  S. Feldman,et al.  Incidence Estimate of Nonmelanoma Skin Cancer (Keratinocyte Carcinomas) in the U.S. Population, 2012. , 2015, JAMA dermatology.

[2]  Brett M Coldiron,et al.  Analysis of Skin Cancer Treatment and Costs in the United States Medicare Population, 1996–2008 , 2013, Dermatologic surgery : official publication for American Society for Dermatologic Surgery [et al.].

[3]  Rodney Sinclair,et al.  Nonmelanoma skin cancer in Australia , 2013, The British journal of dermatology.

[4]  Cesare Furlanello,et al.  mlpy: Machine Learning Python , 2012, ArXiv.

[5]  Ezzeddine Zagrouba,et al.  A PRELIMARY APPROACH FOR THE AUTOMATED RECOGNITION OF MALIGNANT MELANOMA , 2011 .

[6]  Xavier Robin,et al.  pROC: an open-source package for R and S+ to analyze and compare ROC curves , 2011, BMC Bioinformatics.

[7]  Robert LeAnder,et al.  Differentiation of melanoma from benign mimics using the relative‐color method , 2010, Skin research and technology : official journal of International Society for Bioengineering and the Skin (ISBS) [and] International Society for Digital Imaging of Skin (ISDIS) [and] International Society for Skin Imaging.

[8]  Omar S. Al-Kadi,et al.  Texture measures combination for improved meningioma classification of histopathological images , 2010, Pattern Recognit..

[9]  Zhong Linghui,et al.  Pre‐diagnostic digital imaging prediction model to discriminate between malignant melanoma and benign pigmented skin lesion , 2010, Skin research and technology : official journal of International Society for Bioengineering and the Skin (ISBS) [and] International Society for Digital Imaging of Skin (ISDIS) [and] International Society for Skin Imaging.

[10]  M Rademaker,et al.  Successful triage of patients referred to a skin lesion clinic using teledermoscopy (IMAGE IT trial) , 2009, The British journal of dermatology.

[11]  Hadley Wickham,et al.  ggplot2 - Elegant Graphics for Data Analysis (2nd Edition) , 2017 .

[12]  David Koh,et al.  Skin cancer trends among Asians living in Singapore from 1968 to 2006. , 2009, Journal of the American Academy of Dermatology.

[13]  Jun Kong,et al.  Computer-aided prognosis of neuroblastoma on whole-slide images: Classification of stromal development , 2009, Pattern Recognit..

[14]  D. Weedon,et al.  Accuracy of clinical diagnosis of skin lesions , 2008, The British journal of dermatology.

[15]  Chih-Jen Lin,et al.  LIBLINEAR: A Library for Large Linear Classification , 2008, J. Mach. Learn. Res..

[16]  Giuseppe Argenziano,et al.  Digital image analysis for diagnosis of skin tumors. , 2008, Seminars in cutaneous medicine and surgery.

[17]  Tim W. Nattkemper,et al.  A method for linking computed image features to histological semantics in neuropathology , 2007, J. Biomed. Informatics.

[18]  A. Marghoob,et al.  Can automated dermoscopy image analysis instruments provide added benefit for the dermatologist? A study comparing the results of three systems , 2007, The British journal of dermatology.

[19]  J F Aitken,et al.  The effect of an educational programme to improve the skills of general practitioners in diagnosing melanocytic/pigmented lesions , 2007, Clinical and experimental dermatology.

[20]  C. D'Orsi,et al.  Influence of computer-aided detection on performance of screening mammography. , 2007, The New England journal of medicine.

[21]  W. Qian,et al.  Computerized analysis of cellular features and biomarkers for cytologic diagnosis of early lung cancer. , 2007, Analytical and quantitative cytology and histology.

[22]  K. Doi,et al.  Development of a computer-aided diagnostic scheme for detection of interval changes in successive whole-body bone scans. , 2006, Medical Physics (Lancaster).

[23]  D. Whiteman,et al.  Diagnostic accuracy in skin cancer clinics: the Australian experience , 2006, International journal of dermatology.

[24]  Ronald M. Summers,et al.  Wavelet method for CT colonography computer-aided polyp detection , 2006, 3rd IEEE International Symposium on Biomedical Imaging: Nano to Macro, 2006..

[25]  J. Marsden,et al.  More skin cancer detected from nonurgent referrals , 2005, The British journal of dermatology.

[26]  K. Bae,et al.  Automated detection of pulmonary nodules on CT images: effect of section thickness and reconstruction interval--initial results. , 2005, Radiology.

[27]  G Rassner,et al.  Clinical and Laboratory Investigations Digital image analysis for diagnosis of cutaneous melanoma. Development of a highly effective computer algorithm based on analysis of 837 melanocytic lesions , 2004 .

[28]  J. Marioni,et al.  A comparison of dermatologists', surgeons' and general practitioners' surgical management of cutaneous melanoma , 2004, The British journal of dermatology.

[29]  Riccardo Bono,et al.  Melanoma Computer-Aided Diagnosis , 2004, Clinical Cancer Research.

[30]  K. Awai,et al.  Pulmonary nodules at chest CT: effect of computer-aided diagnosis on radiologists' detection performance. , 2004, Radiology.

[31]  P Altmeyer,et al.  Diagnostic and neural analysis of skin cancer (DANAOS). A multicentre study for collection and computer‐aided analysis of data from pigmented skin lesions using digital dermoscopy , 2003, The British journal of dermatology.

[32]  Steven R Feldman,et al.  Skin cancer is among the most costly of all cancers to treat for the Medicare population. , 2003, Journal of the American Academy of Dermatology.

[33]  Guillod Joel,et al.  Validation of segmentation techniques for digital dermoscopy , 2002, Skin research and technology : official journal of International Society for Bioengineering and the Skin (ISBS) [and] International Society for Digital Imaging of Skin (ISDIS) [and] International Society for Skin Imaging.

[34]  D. Piccolo,et al.  Clinical and Laboratory Investigations Dermoscopic diagnosis by a trained clinician vs. a clinician with minimal dermoscopy training vs. computer-aided diagnosis of 341 pigmented skin lesions: a comparative study , 2002 .

[35]  Martin Fiebich,et al.  Automatic detection of pulmonary nodules at spiral CT: clinical application of a computer-aided diagnosis system , 2002, European Radiology.

[36]  T. Freer,et al.  Screening mammography with computer-aided detection: prospective study of 12,860 patients in a community breast center. , 2001, Radiology.

[37]  Frank E. Harrell,et al.  Regression Modeling Strategies: With Applications to Linear Models, Logistic Regression, and Survival Analysis , 2001 .

[38]  K Wolff,et al.  Computer-aided epiluminescence microscopy of pigmented skin lesions: the value of clinical data for the classification process , 2000, Melanoma research.

[39]  L. Tabár,et al.  Potential contribution of computer-aided detection to the sensitivity of screening mammography. , 2000, Radiology.

[40]  R J Ott,et al.  Spectrophotometric assessment of pigmented skin lesions: methods and feature selection for evaluation of diagnostic performance. , 2000, Physics in medicine and biology.

[41]  Mackie,et al.  Clinical accuracy of the diagnosis of cutaneous malignant melanoma , 1998, The British journal of dermatology.

[42]  Andrew P. Bradley,et al.  The use of the area under the ROC curve in the evaluation of machine learning algorithms , 1997, Pattern Recognit..

[43]  Marek Elbaum,et al.  Can early malignant melanoma be differentiated from atypical melanocytic nevus by in vivo techniques? , 1997, Skin research and technology : official journal of International Society for Bioengineering and the Skin (ISBS) [and] International Society for Digital Imaging of Skin (ISDIS) [and] International Society for Skin Imaging.

[44]  R. H. Moss,et al.  Neural network diagnosis of malignant melanoma from color images , 1994, IEEE Transactions on Biomedical Engineering.

[45]  R Marchesini,et al.  In vivo SPECTROPHOTOMETRIC EVALUATION OF NEOPLASTIC AND NON‐NEOPLASTIC SKIN PIGMENTED LESIONS. II: DISCRIMINANT ANALYSIS BETWEEN NEVUS AND MELANOMA , 1992, Photochemistry and photobiology.

[46]  H Fujita,et al.  Application of artificial neural network to computer-aided diagnosis of coronary artery disease in myocardial SPECT bull's-eye images. , 1992, Journal of nuclear medicine : official publication, Society of Nuclear Medicine.

[47]  E. DeLong,et al.  Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. , 1988, Biometrics.

[48]  J. Rodgers,et al.  Thirteen ways to look at the correlation coefficient , 1988 .

[49]  Hideyuki Tamura,et al.  Textural Features Corresponding to Visual Perception , 1978, IEEE Transactions on Systems, Man, and Cybernetics.

[50]  Aixia Guo,et al.  Gene Selection for Cancer Classification using Support Vector Machines , 2014 .

[51]  R Core Team,et al.  R: A language and environment for statistical computing. , 2014 .

[52]  A. Jemal,et al.  Cancer statistics, 2012 , 2012, CA: a cancer journal for clinicians.

[53]  J. L. Smith,et al.  Incidence Estimate of Nonmelanoma Skin Cancer in the United States, 2006 , 2011 .

[54]  J. Q. Rosso,et al.  Skin cancer trends among Asians living in Singapore from 1968 to 2006 , 2010 .

[55]  Heng Tao Shen,et al.  Principal Component Analysis , 2009, Encyclopedia of Biometrics.

[56]  K. Doi,et al.  Current status and future potential of computer-aided diagnosis in medical imaging. , 2005, The British journal of radiology.

[57]  M E Baker,et al.  Computer-aided detection (CAD) for CT colonography: a tool to address a growing need. , 2005, The British journal of radiology.

[58]  Philippe Schmid-Saugeona,et al.  Towards a computer-aided diagnosis system for pigmented skin lesions. , 2003, Computerized medical imaging and graphics : the official journal of the Computerized Medical Imaging Society.

[59]  Su May Hsu,et al.  Multisensor Fusion with Hyperspectral Imaging Data: Detection and Classification , 2003 .

[60]  Vladimir Naumovich Vapni.  The Nature of Statistical Learning Theory , 1995 .

[61]  N. Dubrawsky.  Cancer statistics , 1989, CA: a cancer journal for clinicians.