Model learning analysis of 3D optoacoustic mesoscopy images for the classification of atopic dermatitis.

Atopic dermatitis (AD) is a skin inflammatory disease affecting 10% of the population worldwide. Raster-scanning optoacoustic mesoscopy (RSOM) has recently shown promise in dermatological imaging. We conducted a comprehensive analysis using three machine-learning models, random forest (RF), support vector machine (SVM), and convolutional neural network (CNN) for classifying healthy versus AD conditions, and sub-classifying different AD severities using RSOM images and clinical information. CNN model successfully differentiates healthy from AD patients with 97% accuracy. With limited data, RF achieved 65% accuracy in sub-classifying AD patients into mild versus moderate-severe cases. Identification of disease severities is vital in managing AD treatment.

[1]  D. Norbäck,et al.  Exposure to outdoor air pollution during trimesters of pregnancy and childhood asthma, allergic rhinitis, and eczema. , 2016, Environmental research.

[2]  H. Koga,et al.  Dermoscopic diagnostic performance of Japanese dermatologists for skin tumors differs by patient origin: A deep learning convolutional neural network closes the gap , 2020, The Journal of dermatology.

[3]  A. Reich,et al.  Assessment of Intra- and Inter-Rater Reliability of Three Methods for Measuring Atopic Dermatitis Severity: EASI, Objective SCORAD, and IGA , 2017, Dermatology.

[4]  Habib Safigholi,et al.  Dermatologist Level Dermoscopy Skin Cancer Classification Using Different Deep Learning Convolutional Neural Networks Algorithms , 2018, ArXiv.

[5]  H. Williams,et al.  Measuring atopic eczema severity visually: which variables are most important to patients? , 2005, Archives of dermatology.

[6]  L. Llamado,et al.  Epidemiology, Diagnosis, and Treatment of Atopic Dermatitis in the Developing Countries of Asia, Africa, Latin America, and the Middle East: A Review , 2019, Dermatology and Therapy.

[7]  H. Williams,et al.  Systematic review of treatments for atopic eczema. , 2000, Health technology assessment.

[8]  V. Ntziachristos,et al.  Precision assessment of label-free psoriasis biomarkers with ultra-broadband optoacoustic mesoscopy , 2017, Nature Biomedical Engineering.

[9]  顏素美 Suh-May Yen,et al.  The American Journal of Managed Care , 2015 .

[10]  Y. Ohe,et al.  The Development of a Skin Cancer Classification System for Pigmented Skin Lesions Using Deep Learning , 2020, Biomolecules.

[11]  M. Olivieri,et al.  Adult eczema in Italy: prevalence and associations with environmental factors , 2015, Journal of the European Academy of Dermatology and Venereology : JEADV.

[12]  V. Ntziachristos,et al.  Optoacoustic mesoscopy shows potential to increase accuracy of allergy patch testing. , 2020, Contact dermatitis.

[13]  S. Han,et al.  Classification of the Clinical Images for Benign and Malignant Cutaneous Tumors Using a Deep Learning Algorithm. , 2018, The Journal of investigative dermatology.

[14]  J. Silverberg,et al.  Validation of patient‐reported global severity of atopic dermatitis in adults , 2018, Allergy.

[15]  Amalina Binte Ebrahim Attia,et al.  Raster-scanning optoacoustic mesoscopy (RSOM) imaging as an objective disease severity tool in atopic dermatitis patients. , 2020, Journal of the American Academy of Dermatology.

[16]  M. Tarbox,et al.  Atopic Dermatitis: Pathophysiology. , 2017, Advances in experimental medicine and biology.

[17]  K. H. Kim.  Overview of atopic dermatitis , 2013, Asia Pacific allergy.

[18]  J. Woodfolk,et al.  Skin Barrier Defects in Atopic Dermatitis , 2014, Current Allergy and Asthma Reports.

[19]  A. Irvine,et al.  Harper's Textbook of Pediatric Dermatology , 2011 .

[20]  Rainer Hofmann-Wellenhof,et al.  A deep learning system for differential diagnosis of skin diseases , 2019, Nature Medicine.

[21]  Cui Tao,et al.  Deep learning-based, computer-aided classifier developed with dermoscopic images shows comparable performance to 164 dermatologists in cutaneous disease diagnosis in the Chinese population , 2020, Chinese medical journal.

[22]  Sebastian Thrun,et al.  Dermatologist-level classification of skin cancer with deep neural networks , 2017, Nature.

[23]  Yuan Yu,et al.  TensorFlow: A system for large-scale machine learning , 2016, OSDI.

[24]  Vasilis Ntziachristos,et al.  Raster-scan optoacoustic mesoscopy in the 25-125 MHz range. , 2013, Optics letters.

[25]  Gaël Varoquaux,et al.  Scikit-learn: Machine Learning in Python , 2011, J. Mach. Learn. Res..

[26]  Mohammad H. Jafari,et al.  Melanoma detection by analysis of clinical images using convolutional neural network , 2016, 2016 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC).

[27]  E. Simpson,et al.  What the Eczema Area and Severity Index score tells us about the severity of atopic dermatitis: an interpretability study , 2015, The British journal of dermatology.

[28]  S. Kežić,et al.  Skin barrier in atopic dermatitis. , 2014, Frontiers in bioscience.

[29]  V. Ntziachristos,et al.  Quantification of skin sensitivity to ultraviolet radiation using ultrawideband optoacoustic mesoscopy , 2020, The British journal of dermatology.

[30]  Giuseppe Jurman,et al.  Not again! Data Leakage in Digital Pathology , 2019, 1909.06539.

[31]  Daniel C. Castro,et al.  Causality matters in medical imaging , 2019, Nature Communications.

[32]  J. Silverberg,et al.  Adult eczema prevalence and associations with asthma and other health and demographic factors: a US population-based study. , 2013, The Journal of allergy and clinical immunology.

[33]  K. S. Ravichandran,et al.  Novel Approaches for Diagnosing Melanoma Skin Lesions Through Supervised and Deep Learning Algorithms , 2016, Journal of Medical Systems.

[34]  Vasilis Ntziachristos,et al.  Optoacoustic mesoscopy analysis and quantitative estimation of specific imaging metrics in Fitzpatrick skin phototypes II to V , 2019, Journal of biophotonics.

[35]  Melvin Kai Weng Wong,et al.  Investigation of morphological, vascular and biochemical changes in the skin of an atopic dermatitis (AD) patient in response to dupilumab using raster scanning optoacoustic mesoscopy (RSOM) and handheld confocal Raman spectroscopy (CRS). , 2019, Journal of dermatological science.

[36]  Ching-Hsien Hsu,et al.  Deep learning based feature representation for automated skin histopathological image annotation , 2017, Multimedia Tools and Applications.

[37]  Nima Tajbakhsh,et al.  Embracing Imperfect Datasets: A Review of Deep Learning Solutions for Medical Image Segmentation , 2019, Medical Image Anal..

[38]  N. Holm,et al.  Atopic dermatitis. A genetic-epidemiologic study in a population-based twin sample. , 1993, Journal of the American Academy of Dermatology.

[39]  Y. Fujisawa,et al.  Deep‐learning‐based, computer‐aided classifier developed with a small dataset of clinical images surpasses board‐certified dermatologists in skin tumour diagnosis , 2018, The British journal of dermatology.

[40]  J. Silverberg,et al.  Severity strata for Eczema Area and Severity Index (EASI), modified EASI, Scoring Atopic Dermatitis (SCORAD), objective SCORAD, Atopic Dermatitis Severity Index and body surface area in adolescents and adults with atopic dermatitis , 2017, The British journal of dermatology.