Literature DB >> 30207594

Diagnostic accuracy of content-based dermatoscopic image retrieval with deep classification features.

P Tschandl1,2, G Argenziano3, M Razmara4, J Yap4.   

Abstract

BACKGROUND: Automated classification of medical images through neural networks can reach high accuracy rates but lacks interpretability.
OBJECTIVES: To compare the diagnostic accuracy obtained by using content-based image retrieval (CBIR) to retrieve visually similar dermatoscopic images with corresponding disease labels against predictions made by a neural network.
METHODS: A neural network was trained to predict disease classes on dermatoscopic images from three retrospectively collected image datasets containing 888, 2750 and 16 691 images, respectively. Diagnosis predictions were made based on the most commonly occurring diagnosis in visually similar images, or based on the top-1 class prediction of the softmax output from the network. Outcome measures were area under the receiver operating characteristic curve (AUC) for predicting a malignant lesion, multiclass-accuracy and mean average precision (mAP), measured on unseen test images of the corresponding dataset.
RESULTS: In all three datasets the skin cancer predictions from CBIR (evaluating the 16 most similar images) showed AUC values similar to softmax predictions (0·842, 0·806 and 0·852 vs. 0·830, 0·810 and 0·847, respectively; P > 0·99 for all). Similarly, the multiclass-accuracy of CBIR was comparable with softmax predictions. Compared with softmax predictions, networks trained for detecting only three classes performed better on a dataset with eight classes when using CBIR (mAP 0·184 vs. 0·368 and 0·198 vs. 0·403, respectively).
CONCLUSIONS: Presenting visually similar images based on features from a neural network shows comparable accuracy with the softmax probability-based diagnoses of convolutional neural networks. CBIR may be more helpful than a softmax classifier in improving diagnostic accuracy of clinicians in a routine clinical setting.
© 2018 The Authors. British Journal of Dermatology published by John Wiley & Sons Ltd on behalf of British Association of Dermatologists.

Entities:  

Year:  2018        PMID: 30207594     DOI: 10.1111/bjd.17189

Source DB:  PubMed          Journal:  Br J Dermatol        ISSN: 0007-0963            Impact factor:   9.302


  5 in total

1.  A Deep Learning Approach for Histopathological Diagnosis of Onychomycosis: Not Inferior to Analogue Diagnosis by Histopathologists.

Authors:  Florence Decroos; Sebastian Springenberg; Tobias Lang; Marc Päpper; Antonia Zapf; Dieter Metze; Volker Steinkraus; Almut Böer-Auer
Journal:  Acta Derm Venereol       Date:  2021-08-31       Impact factor: 3.875

Review 2.  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

Review 3.  A Survey on Human Cancer Categorization Based on Deep Learning.

Authors:  Ahmad Ibrahim; Hoda K Mohamed; Ali Maher; Baochang Zhang
Journal:  Front Artif Intell       Date:  2022-06-27

Review 4.  Machine learning for precision dermatology: Advances, opportunities, and outlook.

Authors:  Ernest Y Lee; Nolan J Maloney; Kyle Cheng; Daniel Q Bach
Journal:  J Am Acad Dermatol       Date:  2020-07-06       Impact factor: 11.527

5.  A patient-centric dataset of images and metadata for identifying melanomas using clinical context.

Authors:  Veronica Rotemberg; Nicholas Kurtansky; Brigid Betz-Stablein; Liam Caffery; Emmanouil Chousakos; Noel Codella; Marc Combalia; Stephen Dusza; Pascale Guitera; David Gutman; Allan Halpern; Brian Helba; Harald Kittler; Kivanc Kose; Steve Langer; Konstantinos Lioprys; Josep Malvehy; Shenara Musthaq; Jabpani Nanda; Ofer Reiter; George Shih; Alexander Stratigos; Philipp Tschandl; Jochen Weber; H Peter Soyer
Journal:  Sci Data       Date:  2021-01-28       Impact factor: 6.444

  5 in total

北京卡尤迪生物科技股份有限公司 © 2022-2023.