Literature DB >> 28332777

Diagnostic performance of the MelaFind device in a real-life clinical setting.

Christine Fink1, Claudia Jaeger2, Katharina Jaeger2, Holger A Haenssle1,3.   

Abstract

BACKGROUND: MelaFind is a multispectral computer vision system intended to -provide additional information on melanocytic lesions suspected of being melanoma by -objectively assessing their three-dimensional morphology.
OBJECTIVES: Analysis of the diagnostic performance of MelaFind in a real-life clinical setting. PATIENTS AND METHODS: In this observational study, 360 pigmented skin lesions (PSL) in 111 patients were assessed by office-based dermatologists using MelaFind. Scores ≥ 2 were considered to be suspicious of malignancy. The decision for surgical excision was left to the discretion of the examining dermatologists.
RESULTS: MelaFind scores ≥ 2 were observed in 147 of 360 PSL (40.8 %). Of the 107 excised lesions with a MelaFind-score ≥ 2, the diagnosis of melanoma was made in three cases; 53 (49.5 %) lesions proved to be dysplastic nevi. Among all lesions biopsied (n = 113), the sensitivity and specificity of MelaFind was 100 % and 5.5 %, respectively. While a higher specificity of 68.5 % may be assumed with respect to the overall data set (n = 360), this assumption is limited by incomplete follow-up data required to confirm that all non-excised lesions with a score < 2 were actually benign.
CONCLUSION: The high sensitivity of MelaFind facilitated the detection of melanoma. The overall specificity and benign-to-malignant ratio of excised lesions were acceptable. These parameters may be improved by using higher cutoff scores for excisional biopsies, and by more vigorously selecting PSL for MelaFind examination.
© 2017 Deutsche Dermatologische Gesellschaft (DDG). Published by John Wiley & Sons Ltd.

Entities:  

Mesh:

Year:  2017        PMID: 28332777     DOI: 10.1111/ddg.13220

Source DB:  PubMed          Journal:  J Dtsch Dermatol Ges        ISSN: 1610-0379            Impact factor:   5.584


  4 in total

Review 1.  Lack of Transparency and Potential Bias in Artificial Intelligence Data Sets and Algorithms: A Scoping Review.

Authors:  Roxana Daneshjou; Mary P Smith; Mary D Sun; Veronica Rotemberg; James Zou
Journal:  JAMA Dermatol       Date:  2021-11-01       Impact factor: 11.816

2.  Hyperspectral and multispectral image processing for gross-level tumor detection in skin lesions: a systematic review.

Authors:  Eleni Aloupogianni; Masahiro Ishikawa; Naoki Kobayashi; Takashi Obi
Journal:  J Biomed Opt       Date:  2022-06       Impact factor: 3.758

3.  Curve-Based Classification Approach for Hyperspectral Dermatologic Data Processing.

Authors:  Stig Uteng; Eduardo Quevedo; Gustavo M Callico; Irene Castaño; Gregorio Carretero; Pablo Almeida; Aday Garcia; Javier A Hernandez; Fred Godtliebsen
Journal:  Sensors (Basel)       Date:  2021-01-20       Impact factor: 3.576

Review 4.  Applications of hyperspectral imaging in the detection and diagnosis of solid tumors.

Authors:  Yating Zhang; Xiaoqian Wu; Li He; Chan Meng; Shunda Du; Jie Bao; Yongchang Zheng
Journal:  Transl Cancer Res       Date:  2020-02       Impact factor: 1.241

  4 in total

北京卡尤迪生物科技股份有限公司 © 2022-2023.