Imaging in Dermatology

From Clinfowiki
Revision as of 20:33, 26 April 2020 by Joneshea (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

Imaging in Dermatology



Importance of Imaging in Dermatology

The specialty of dermatology has a great reliance on visualization in making a diagnosis, monitoring the skin, care follow up, research, teaching, and consultation. To these ends, the advancements in the areas of information technology, modality of image capture, and image utilization, are especially important.

Information Technology, Images

It is important to have collected images available for screening, diagnosis, and treatment. Images may be stored within electronic health records (EHRs), within secure, HIPAA compliant external storage systems, with quick access for viewing, or within health image libraries with patient permission. HIM professionals play a critical role in protecting a patient’s privacy (1).

Modalities of Imaging in Dermatology and Utilization


Digital Photography/Clinical Photographs

Images are captured and digitized to be stored as a computer file for viewing as an adjunct to patient care. In dermatology these images are used for monitoring skin/nail conditions, changes in characteristics of moles or other lesions, are diagnostic aids for consensus, assist in full body skin exam. These images may prevent unnecessary biopsy, lead to early detection of a disease or skin cancer, and may be extracted for consultation, teaching, research, and publication. Digital photography also allows for teledermatology, including virtual visits, e-consultation and e-visits to have photographs available for consideration in patient care. Any release of images for any reason must follow disclosure of personal health information policy and procedure. The Center for Disease Control and Prevention has a library of digital photographic images available to the public through the Public Health Image Library (PHIL) for use by students, researchers, teachers, communities and others, including a large array of dermatologic photos (2).

Dermoscopy

Dermoscopy, also termed epiluminescence microscopy (ELM), is a noninvasive method of imaging, used to allow visualization with magnification and illumination of sub-surface morphological features of the skin. It allows the in vivo evaluation of colors, patterns, and structures of the skin, including the epidermis, the dermoepidermal junction, and the papillary dermis that are not visible to the naked eye. Dermatoscopes are equipped with magnifying lenses and light sources that may be polarized or non-polarized, and toggling between polarizations can aid in seeing structures at difference levels within the skin. Visualization can be done with contact or without contact of the lens and without or without liquid. The structures seen are correlated to histologic features used in diagnosis. Experienced clinicians utilize dermoscopy to aid in the identification of tumors such as melanoma, basal cell carcinoma, benign lesions, and use the tool to identify features in skin and nails folds in diseases such as scleroderma, granulomatous conditions, and sarcoidosis. Dermoscopy for dermatology clinicians has been compared to the stethoscope for a cardiologist. Attachments to cameras or smart-phones allow for dermoscopic pictures or videos (3, 4, 5, 6).

Total body photography/mole mapping

Total body imaging may be 2D or 3D. It may allow for never before appreciated lesions consistent with melanoma, revealing features considered ugly ducklings, compared to the features of other lesions. Photographs can be taken of most all areas of the skin, though it can be difficult to capture areas of groin, axilla, scalp, and body folds. Resolution of the imaging has continued to improve over time with this technology. Though total body imaging photography allows for improved mapping of lesions, including moles, and may allow for early detection of skin cancer, it is not commonly covered by most insurers and the expense may be prohibitive for some patients (7).

Reflectance Confocal Microscopy

Reflectance confocal microscopy (RCM) is an imaging device that allows for non-invasive visualization of morphology at the cellular level, near histopathology resolution, and has been shown to improve diagnostic accuracy of when used in combination with dermoscopy. Interpretation of patterns seen with RCM requires expert training, accumulation of experience, and ideally has a component of quality control with peer review of cases. RCM utilizes interpretation of features based on grayscale, horizontally-oriented visualization, and identification of key criterial features from an extensive list of published RCM features. RCM has the backing of significant literature, demonstrating efficacy in diagnosing melanocytic and nonmelanocytic neoplasms (8,9,10)

Optical Coherence Tomography

Optical coherence tomography (OCT) is a non-invasive, optical, real-time imaging modality. It utilizes infrared radiation to visualize cross-sections of structures within tissues, for a morphological interpretation. It can be viewed as 2 or 3 dimensional and is high resolution to a depth of several millimeters, in some instances. The cross-sections are able to be visualized both horizontally (en-face) and vertically (slices). The structures within the skin can be visualized in the epidermis, the dermoepidermal junction, the dermis and also can appreciate structures such as hair follicles, vessels, and sweat glands. It has the capability to offer the well-trained provider the ability to diagnose and monitor conditions of the skin such as malignancy or inflammatory disease. OCT has better penetration depth compared to RCM, but RCM offers better lateral resolution (11, 12, 13).

Artificial Intelligence in Dermatology

Deep learning convolutional neural networks (CNNs) are demonstrating the potential to diagnose skin malignancy, visually. An automated CNN allows for classification of skin lesions utilizing saved pixels and disease labels from datasets of images, differentiating from benign lesions and skin malignancy. Deep learning CNN has achieved success in appropriate diagnosis on par with dermatologists in classifying skin cancer. Testing of the methods include providing biopsy proven images of photographic and dermoscopic lesions to dermatologist, who are asked if they would biopsy or if they would reassure their patient regarding the lesion. In this test method, deep learning CNN outperformed the average of the dermatologist at skin cancer classification, including keratinocyte carcinomas and melanomas(14).


[[ http://www.clinfowiki.org/wiki/index.php/Imaging_Informatics_-_Special_Concerns_for_Dermatology]]


References


1. Wiedemann, L. Using Clinical Photos in EHRs. J AHIMA. 2010; 81(4): 44-45. Retrieved online April20, 2020 from https://bok.ahima.org/doc?oid=99593#.XqTiFaqWzIV

2. Public Health Image Library. Center for Disease Control and Prevention. https://phil.cdc.gov/

3. Lalla, A, Apalla, Z, Lazaridou, E, Ionnides, D . Dermoscopy; Imaging in Dermatology. Hamblin, MR, Avci, P, Gupta, GK, Eds., ed: Academic Press, 2016, pp. 13-18. Retrieved online April 20, 2020, from https://www.sciencedirect.com/science/article/pii/B9780128028384000030

4. Sonthalia S, Kaliyadan F. Dermoscopy Overview and Extradiagnostic Applications. [Updated 2020 Feb 7]. In: StatPearls [Internet]. Treasure Island (FL): StatPearls Publishing; 2020 Jan. Retrieved online April 20,2020 from https://www.ncbi.nlm.nih.gov/books/NBK537131/

5. Kittler, H, Pehamberger, K, Wolff, M, Binder,M. Diagnostic accuracy of dermoscopy. Lanc Onc. 2002; 3(3):159-165. Retrieved online April 20, 2020 from https://www.sciencedirect.com/science/article/abs/pii/S1470204502006794

6. Tanaka, M. Dermoscopy. J Dermatol. 2006; 33: 513-517. Retrieved online April 20, 2020 from https://www.ncbi.nlm.nih.gov/pubmed/16923131

7. Caffery, L. The role of DICOM in digital imaging in dermatology. Centre for Health Online. Retrieved online April 20, 2020 from https://www.dicomstandard.org/wp-content/uploads/2018/10/Day2_S6_Digital-Imaging-in-Dermatology_L.Caffery.pdf

8. Witkowski AM, Łudzik J, Arginelli F, Bassoli S, Benati E, Casari A, et al. (2017) Improving diagnostic sensitivity of combined dermoscopy and reflectance confocal microscopy imaging through double reader concordance evaluation in telemedicine settings: A retrospective study of 1000 equivocal cases. PLoS ONE 12(11): e0187748. https://doi.org/10.1371/journal.pone.0187748

9. Pellacani G, Scope A, Gonzalez S, et al. Reflectance confocal microscopy made easy: The 4 must-know key features for the diagnosis of melanoma and nonmelanoma skin cancers. J Am Acad Dermatol. 2019; 81:520–526. Retrieved online April 20, 2020 from https://www.sciencedirect.com/science/article/pii/S0190962219305365?via%3Dihub

10. Cinotti, E., Labeille, B., Debarbieux, S., et al. Dermoscopy vs. reflectance confocal microscopy for the diagnosis of lentigo maligna. J Eur Acad Dermatol Venereol. 2018; 32: 1284-1291. doi:10.1111/jdv.14791

11. Boone, M, Jemec, GBE, Del Marmol, V. High‐definition optical coherence tomography enables visualization of individual cells in healthy skin: comparison to reflectance confocal microscopy. Exp Dermatol. 2012; 21: 740-744. doi:10.1111/j.1600-0625.2012.01569.x

12. Gambichler, T, Jaedicke, V, Terras, S. Optical coherence tomography in dermatology: technical and clinical aspects. Arch Dermatol Res. 2011; 303: 457–473

13. Garcia-Hernandez, A, Roldan-Marin, R, Iglesias-Garcia, P, Malvehy, J. Healthy lower lip mucosa. A correlation study between high definition optical coherence tomography, relectance confocal microscopy, and histology. Dermatology Research and Practice. 2013; Article: ID 205256 https://doi.org/10.1155/2013/205256

14. Esteva, A, Kuprel, B, Novoa, RA, et al. Dermatologist-level classification of skin cancer with deep neural networks. Nature. 2017; 542: 115-125. https://doi.org/10.1038/nature21056


Submitted by Heather Onoday, FNP