Unimaged Melanomas in Store-and-Forward Teledermatology

Gendreau, James L; Gemelas, Jordan; Wang, Melinda; Capulong, Dana; Lau, Clayton; Bratten, Dustin M; Dougall, Brittany; Markham, Craig; Raugi, Gregory J.

doi:10.1089/tmj.2016.0170

Cited by 20 publications

(11 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…General practitioners and patients are likely to omit suspect lesions [ 31 , 33 , 39 ]. This limitation, which has been described in general practice [ 13 , 65 ], exists in teledermatology, that is, general practitioners could miss up to 30% of melanomas [ 66 ]. Another difficulty is the omission of specific areas that are either difficult to access, such as hair, wounds, or the ear [ 51 , 54 ], or those considered sensitive (genitalia) [ 67 , 68 ].…”

Section: Discussionmentioning

confidence: 99%

Use of Smartphones for Early Detection of Melanoma: Systematic Review

Rat¹,

Hild²,

Sérandour³

et al. 2018

J Med Internet Res

View full text Add to dashboard Cite

BackgroundThe early diagnosis of melanoma is associated with decreased mortality. The smartphone, with its apps and the possibility of sending photographs to a dermatologist, could improve the early diagnosis of melanoma.ObjectiveThe aim of our review was to report the evidence on (1) the diagnostic performance of automated smartphone apps and store-and-forward teledermatology via a smartphone in the early detection of melanoma, (2) the impact on the patient’s medical-care course, and (3) the feasibility criteria (focusing on the modalities of picture taking, transfer of data, and time to get a reply).MethodsWe conducted a systematic search of PubMed for the period from January 1, 2007 (launch of the first smartphone) to November 1, 2017.ResultsThe results of the 25 studies included 13 concentrated on store-and-forward teledermatology, and 12 analyzed automated smartphone apps. Store-and-forward teledermatology opens several new perspectives, such as it accelerates the care course (less than 10 days vs 80 days), and the related procedures were assessed in primary care populations. However, the concordance between the conclusion of a teledermatologist and the conclusion of a dermatologist who conducts a face-to-face examination depended on the study (the kappa coefficient range was .20 to .84, median κ=.60). The use of a dermoscope may improve the concordance (the kappa coefficient range was .29 to .87, median κ=.74). Regarding automated smartphone apps, the major concerns are the lack of assessment in clinical practice conditions, the lack of assessment in primary care populations, and their low sensitivity, ranging from 7% to 87% (median 69%). In this literature review, up to 20% of the photographs transmitted were of insufficient quality. The modalities of picture taking and encryption of the data were only partially reported.ConclusionsThe use of store-and-forward teledermatology could improve access to a dermatology consultation by optimizing the care course. Our review confirmed the absence of evidence of the safety and efficacy of automated smartphone medical apps. Further research is required to determine quality criteria, as there was major variability among the studies.

show abstract

Section: Discussionmentioning

confidence: 99%

Use of Smartphones for Early Detection of Melanoma: Systematic Review

Rat¹,

Hild²,

Sérandour³

et al. 2018

J Med Internet Res

View full text Add to dashboard Cite

show abstract

“…Mobile teledermatology is the use of a smartphone to take and send images and information to a teleconsultant; mobile as an incidental part of the FTF examination. With reduced FTF appointments, there is a real risk of these lesions going unnoticed [62,63]. However, the reverse can also be true:…”

Section: Mobile Teledermatology and Teledermoscopymentioning

confidence: 99%

Recent trends in teledermatology and teledermoscopy

Lee¹,

Finnane²,

Soyer³

2018

Dermatol Pract Concept

100

View full text Add to dashboard Cite

Recent studies continue to examine teledermatology for a wide variety of disorders (Table 1); teledermoscopy has been mostly studied for assessing melanocytic and keratinocytic lesions [11][12][13][14][15][16][17][18][19][20][21][22][23][24][25]. Accuracy and Interobserver ConcordanceMost studies show comparable diagnostic accuracy between teledermatology and face-to-face (FTF) care [55,64,66,71,72], although 3 earlier studies reviewed by Lee and English [71] found teledermatology either significantly superior [9] or inferior [73,74]. More specifically, Finnane et al [75] reviewed accuracy of diagnosis for skin cancer; most studies showed that FTF consultations were more accurate than teledermatology (67%-85% vs 51%-85%); however, some studies found teledermatology was more accurate.

show abstract

“…These results further emphasize the high sensitivity of our System B, which identified cancerous lesions smaller than 6 mm diameter. An accurate operative telemedicine prototype as a tool for cloud-directed diagnosis is a field which might be further improved, rendering this system as a candidate for use in the detection of unimaged skin cancer [30].…”

Section: Discussionmentioning

confidence: 99%

Dermoscopy diagnosis of cancerous lesions utilizing dual deep learning algorithms via visual and audio (sonification) outputs: Laboratory and prospective observational studies

Walker

Rehg

Kalra³

et al. 2019

EBioMedicine

View full text Add to dashboard Cite

Background: Early diagnosis of skin cancer lesions by dermoscopy, the gold standard in dermatological imaging, calls for a diagnostic upscale. The aim of the study was to improve the accuracy of dermoscopic skin cancer diagnosis through use of novel deep learning (DL) algorithms. An additional sonification-derived diagnostic layer was added to the visual classification to increase sensitivity. Methods: Two parallel studies were conducted: a laboratory retrospective study (LABS, n = 482 biopsies) and a non-interventional prospective observational study (OBS, n = 63 biopsies). A training data set of biopsy-verified reports, normal and cancerous skin lesions (n = 3954), were used to develop a DL classifier exploring visual features (System A). The outputs of the classifier were sonified, i.e. data conversion into sound (System B). Derived sound files were analyzed by a second machine learning classifier, either as raw audio (LABS, OBS) or following conversion into spectrograms (LABS) and by image analysis and human heuristics (OBS). The OBS criteria outcomes were System A specificity and System B sensitivity as raw sounds, spectrogram areas or heuristics. Findings: LABS employed dermoscopies, half benign half malignant, and compared the accuracy of Systems A and B. System A algorithm resulted in a ROC AUC of 0.976 (95% CI, 0.965-0.987). Secondary machine learning analysis of raw sound, FFT and Spectrogram ROC curves resulted in AUC's of 0.931 (95% CI 0.881-0.981), 0.90 (95% CI 0.838-0.963) and 0.988 (CI 95% 0.973-1.001), respectively. OBS analysis of raw sound dermoscopies by the secondary machine learning resulted in a ROC AUC of 0.819 (95% CI, 0.7956 to 0.8406). OBS image analysis of AUC for spectrograms displayed a ROC AUC of 0.808 (CI 95% 0.6945 To 0.9208). By applying a heuristic analysis of Systems A and B a sensitivity of 86% and specificity of 91% were derived in the clinical study. Interpretation: Adding a second stage of processing, which includes a deep learning algorithm of sonification and heuristic inspection with machine learning, significantly improves diagnostic accuracy. A combined two-stage system is expected to assist clinical decisions and de-escalate the current trend of over-diagnosis of skin cancer lesions as pathological.

show abstract

Unimaged Melanomas in Store-and-Forward Teledermatology

Cited by 20 publications

References 16 publications

Use of Smartphones for Early Detection of Melanoma: Systematic Review

Use of Smartphones for Early Detection of Melanoma: Systematic Review

Recent trends in teledermatology and teledermoscopy

Dermoscopy diagnosis of cancerous lesions utilizing dual deep learning algorithms via visual and audio (sonification) outputs: Laboratory and prospective observational studies

Contact Info

Product

Resources

About