A Deep Learning Framework for Recognition of Various Skin Lesions due to Diabetes

Gupta, Tanu; Saini, Satyam; Saini, Aradhya; Aggarwal, Shrutika; Mittal, Ankush

doi:10.1109/icacci.2018.8554897

Cited by 4 publications

(2 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…[24] Machine Learning Repository Dermatology Data Set 93.70% (ANN) [25] Department of Skin in KEM Hospital Mumbai 90% (ANN & RF) [28], [48] From surveys and websites and From Clinic images CNN provided accurate and efficient result as compared to SVM Classifier [28] , AUROC 95.8% and 100% sensitivity with specificity of 64.8%(ANN) [48] [29] Four Sunyani Municipality, Ghana, medical centers 88%-dermatitis, 85%-arcane vulgaris and 84.7%-scabies (CNN) [34], [37], [47] Dermoscopic images collected from Internet 81%(CNN) [34],…”

Section: %(Ann)mentioning

confidence: 99%

See 1 more Smart Citation

Skin Cancer Detection and Classification using Deep learning methods

Kumari,

Rattan

2023

IJEER

View full text Add to dashboard Cite

Skin cancer is a very dangerous disease that needs to be found early, so that it can be treated effectively. In the past few years, classifiers built on convolutional neural networks (CNNs) have become the best way to find melanoma. According to the review, the CNN-based classifier is as accurate as dermatologist in classifying skin cancer images, allowing for faster and more accurate detection. This article examines the most recent studies on Machine learning and deep learning-based melanoma categorization in depth. We provide a comprehensive description of the machine learning and deep learning classifier, including details on the accuracy of these classifiers. The primary objective of this research is to analyze and collect current research trends, issues, and opportunities for melanoma diagnosis, as well as to investigate the current approach for using deep learning to detect and recognize melanoma. The main finding of this review is that the neural network provides high accuracy as comparison to machine learning methods.

show abstract