Automated detection of nonmelanoma skin cancer using digital images: a systematic review

Abstract

Background Computer-aided diagnosis of skin lesions is a growing area of research, but its application to nonmelanoma skin cancer (NMSC) is relatively under-studied. The purpose of this review is to synthesize the research that has been conducted on automated detection of NMSC using digital images and to assess the quality of evidence for the diagnostic accuracy of these technologies. Methods Eight databases (PubMed, Google Scholar, Embase, IEEE Xplore, Web of Science, SpringerLink, ScienceDirect, and the ACM Digital Library) were searched to identify diagnostic studies of NMSC using image-based machine learning models. Two reviewers independently screened eligible articles. The level of evidence of each study was evaluated using a five tier rating system, and the applicability and risk of bias of each study was assessed using the Quality Assessment of Diagnostic Accuracy Studies tool. Results Thirty-nine studies were reviewed. Twenty-four models were designed to detect basal cell carcinoma, two were designed to detect squamous cell carcinoma, and thirteen were designed to detect both. All studies were conducted in silico. The overall diagnostic accuracy of the classifiers, defined as concordance with histopathologic diagnosis, was high, with reported accuracies ranging from 72 to 100% and areas under the receiver operating characteristic curve ranging from 0.832 to 1. Most studies had substantial methodological limitations, but several were robustly designed and presented a high level of evidence. Conclusion Most studies of image-based NMSC classifiers report performance greater than or equal to the reported diagnostic accuracy of the average dermatologist, but relatively few studies have presented a high level of evidence. Clinical studies are needed to assess whether these technologies can feasibly be implemented as a real-time aid for clinical diagnosis of NMSC. Electronic supplementary material The online version of this article (10.1186/s12880-019-0307-7) contains supplementary material, which is available to authorized users.

Machine Learning in Dermatology: Current Applications, Opportunities, and Limitations

“…Tschandl and colleagues admit that this algorithm is not ready to be implemented in the clinic, but the results of the study do demonstrate that CNNs are capable of more complex diagnoses and call for the collection of more dermoscopic and clinical images of rare malignant lesions. Marka et al [3] carried out a systematic review of 39 studies on the automated detection of NMSC and found that most studies report model performance greater than or equal to the reported diagnostic accuracy of the average dermatologist, but that relatively few studies have presented a high level of evidence.…”

“…Translational research in dermatology is already abundant, with data from the genome, epigenome, transcriptome, proteome, and microbiome, areas of research that are often referred to by the shortened term ''omics'' [2]. Recent advancements in faster processing and cheaper storage have allowed for the development of machine learning (ML) algorithms with human-like intelligence that have numerous applications in dermatology [3][4][5]. To assess the effectiveness of these emerging technologies, it is imperative that dermatologists have a basic understanding of artificial intelligence and ML.…”

Highly Accurate Melanoma Detection from Skin Images using Multiple Feature Extraction and Support Vector Machine (SVM)

“…Frequently, the inspiration is to accomplish consistency in a powerful range for a lot of information, flag, or pictures to stay away from mental interruption or exhaustion. [14] For instance, a paper will endeavor to make the entirety of the pictures in an issue share a comparative scope of grayscale.…”

“…Gamma correction, or regularly basically gamma, is a nonlinear activity used to encode and interpret luminance or tristimulus values in video or still picture systems. [14] Gamma correction is, in the least complex cases, characterized by the accompanying force law articulation: = (2) where the non-negative genuine info esteem I in is raised to the power and duplicated by the consistent K, to get the yield esteem Iout. In the regular instance of K = 1, information sources and yields are commonly in the range 0-1.…”

Microscopic melanoma detection and classification: A framework of pixel‐based fusion and multilevel features reduction

“…But still, several challenges exist which decreased the accuracies of these automated systems. In general, a simple automated system follows these steps: preprocessing of original data, detection of the lesion through segmentation or featuresbased techniques, and finally classification by extracting patterns information (Jadooki, Mohamad, Saba, Almazyad, & Rehman, 2016;Majumder & Ullah, 2019;Marka, Carter, Toto, & Hassanpour, 2019;Sharif et al, 2017). A lot of techniques are presented for contrast enhancement in this domain to improve the visual quality of lesion in the given images.…”