• Corpus ID: 9376878

Automatic Skin Lesion Analysis using Large-scale Dermoscopy Images and Deep Residual Networks

  title={Automatic Skin Lesion Analysis using Large-scale Dermoscopy Images and Deep Residual Networks},
  author={Lei Bi and Jinman Kim and Euijoon Ahn and David Dagan Feng},
Malignant melanoma has one of the most rapidly increasing incidences in the world and has a considerable mortality rate. Early diagnosis is particularly important since melanoma can be cured with prompt excision. Dermoscopy images play an important role in the non-invasive early detection of melanoma [1]. However, melanoma detection using human vision alone can be subjective, inaccurate and poorly reproducible even among experienced dermatologists. This is attributed to the challenges in… 

Tables from this paper

Automatic lesion segmentation using atrous convolutional deep neural networks in dermoscopic skin cancer images

A novel design of a convolutional neural network (CNN) framework based on atrous convolutions for automatic lesion segmentation based on the concept of atrous/dilated convolutions which are effective for semantic segmentation is proposed.

Melanoma Segmentation and Classification using Deep Learning

  • R. D. SeejaA. Suresh
  • Computer Science, Medicine
    International Journal of Innovative Technology and Exploring Engineering
  • 2019
The main purpose of this research is to develop an automatic system to improve the classification performance of melanoma and this method would constitute a valuable support for physicians in every day clinical practice.

Skin Lesion Segmentation from Dermoscopic Images Using Convolutional Neural Network

An automated method for segmenting lesion boundaries that combines two architectures, the U-Net and the ResNet, collectively called Res-Unet is proposed, which achieves comparable results to the current available state-of-the-art techniques.

Skin Lesion Classification using Deep Learning and Image Processing

The aim of this research paper is to render non-invasive skin cancer screening a common norm, making it simpler.

Deep learning for skin lesion segmentation

  • R. MishraO. Daescu
  • Computer Science
    2017 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)
  • 2017
A deep learning method for automatic skin lesion segmentation that can outperform the submissions in terms of segmentation accuracy and compare against the benchmark results submitted by other participants.

Training FCNs model with lesion-size-unified dermoscopy images for lesion segmentation

Results show that using the lesion-size-unified dermoscopy images can greatly improve the segmentation performance and is of value on assisting artificial delineation, revealing the relations between performance and various lesion sizes, and guiding significance for preprocessing of input dermos copy images while using FCNs model.

Skin Cancer Diagnosis Using Dermoscopic Images and Patient Information

The goal is to understand if there are any performance improvements when incorporating the patient’s clinical information (age, sex, body region) in the decision system, and different strategies based on Deep Neural Networks that combine these covariates with images are proposed.

Two-Stage Deep Neural Network via Ensemble Learning for Melanoma Classification

An ensemble method that can integrate different types of classification networks for melanoma classification using dermoscopy images is proposed, laying the foundation for early diagnosis and later treatment of melanoma.

Skin Lesion Segmentation in Dermoscopic Images with Combination of YOLO and GrabCut Algorithm

A novel and effective pipeline for skin lesion segmentation in dermoscopic images combining a deep convolutional neural network named as You Only Look Once (YOLO) and the GrabCut algorithm is proposed, outperforming other deep learning-based methods.

Input space augmentation for skin lesion segmentation in dermoscopic images

This thesis studies approaches to improve the segmentation of skin lesions in dermoscopic images, which is often the first and the most important task in the diagnosis of dermatological conditions and presents two methods to improve deep learning-based segmentation by augmenting the input space of convolutional neural network models.



Automatic melanoma detection via multi-scale lesion-biased representation and joint reverse classification

This study proposes a new automatic melanoma detection method for dermoscopy images via multi-scale lesion-biased representation (MLR) and joint reverse classification (JRC), and demonstrates superior classification performance compared to the current state-of-the-art methods.

Automated saliency-based lesion segmentation in dermoscopic images

  • E. AhnLei Bi D. Feng
  • Computer Science
    2015 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)
  • 2015
A new automated saliency-based skin lesion segmentation (SSLS) that is designed to exploit the inherent properties of dermoscopic images, which have a focal central region and subtle contrast discrimination with the surrounding regions is proposed.

Automatic detection of blue-white veil and related structures in dermoscopy images

Skin Lesion Analysis Towards Melanoma Detection

  • S. SreenaA. Lijiya
  • Computer Science
    2019 2nd International Conference on Intelligent Computing, Instrumentation and Control Technologies (ICICICT)
  • 2019
The overall aim of the proposed skin lesion analysis system is to fortify efforts to reduce melanoma cognate deaths and dispensable biopsies by ameliorating the precision and efficiency of early melanoma detection.

Saliency-Based Lesion Segmentation Via Background Detection in Dermoscopic Images

It is suggested that saliency detection using the reconstruction errors derived from a sparse representation model coupled with a novel background detection can more accurately discriminate the lesion from surrounding regions.

An Ensemble of Fine-Tuned Convolutional Neural Networks for Medical Image Classification

A new method for classifying medical images that uses an ensemble of different convolutional neural network (CNN) architectures that achieves a higher accuracy than established CNNs and is only overtaken by those methods that source additional training data.

A methodological approach to the classification of dermoscopy images

Deep Residual Learning for Image Recognition

This work presents a residual learning framework to ease the training of networks that are substantially deeper than those used previously, and provides comprehensive empirical evidence showing that these residual networks are easier to optimize, and can gain accuracy from considerably increased depth.

ImageNet classification with deep convolutional neural networks

A large, deep convolutional neural network was trained to classify the 1.2 million high-resolution images in the ImageNet LSVRC-2010 contest into the 1000 different classes and employed a recently developed regularization method called "dropout" that proved to be very effective.

Wider or Deeper: Revisiting the ResNet Model for Visual Recognition