Classification Of Dermoscopic Images Research Articles

Design of inception architecture for skin melanoma classification

The number of people diagnosed with skin cancer is increasing sharply. Both invasive and non-invasive methods of examination may be used to investigate it. However, the invasive method is more difficult for the patient because samples must be taken from the lesion itself, or the whole lesion must be cut out. It also requires more time and cost. To avoid invasive procedures, computer-based analysis and diagnosis have the potential to increase diagnostic accuracy and turnaround time. This study develops a unique discriminative deep learning architecture (DDLA) for dermoscopic image classification (DIC), called DDLA-DIC, which uses the concept of inception. Using this concept, the proposed DDLA-DIC system is designed wider and deeper and the network learns from various spatial patterns. The proposed DDLA-DIC system can extract image characteristics from dermoscopic images for skin cancer diagnosis in an effective and efficient way. The proposed DDLA-DIC system is evaluated by utilizing the dermoscopic images from the PH2 database, and the obtained classification results are based on a random split approach. The simulation results indicate that the framework has a great deal of potential with 99.79% accuracy.

Efficient Feature Based Melanoma Skin Image Classification Using Machine Learning Approaches

Skin cancer is the most prevalent and deadliest kind of cancer. Melanoma is the most dangerous type of skin cancer, but it can be detected earlier and successfully treated. The dermoscopic image classification using Machine Learning (ML) approaches in identifying melanoma is increased over the last two decades. The proposed classification system involves three stages. Initially, the pre-processing employs the median filter and thresholding approach aids to remove the hairs and unwanted noise. Then, the shape components, Asymmetry, Border Irregularity, Colour and Dermoscopic structure (ABCD) rule, and Grey Level Co-occurrence Matrix (GLCM) features are utilized to extract the skin lesion region. After that, the K-Nearest Neighbor (KNN), Random Forest (RF), and Support Vector Machine (SVM) classifiers are employed to perform melanoma skin classification from the skin lesion. The dermoscopic skin images used in this study are obtained from the PH2 database. Finally, the SVM classifier outperformed the other classifiers in identifying melanoma skin cancer, providing 94.81% efficiency.

Dermoscopic Image Classification Method Using an Ensemble of Fine-Tuned Convolutional Neural Networks.

Aiming at the problems of large intra-class differences, small inter-class differences, low contrast, and small and unbalanced datasets in dermoscopic images, this paper proposes a dermoscopic image classification method based on an ensemble of fine-tuned convolutional neural networks. By reconstructing the fully connected layers of the three pretrained models of Xception, ResNet50, and Vgg-16 and then performing transfer learning and fine-tuning the three pretrained models with the ISIC 2016 Challenge official skin dataset, we integrated the outputs of the three base models using a weighted fusion ensemble strategy in order to obtain a final prediction result able to distinguish whether a dermoscopic image indicates malignancy. The experimental results show that the accuracy of the ensemble model is 86.91%, the precision is 85.67%, the recall is 84.03%, and the F1-score is 84.84%, with these four evaluation metrics being better than those of the three basic models and better than some classical methods, proving the effectiveness and feasibility of the proposed method.

Dermoscopic Image Classification of Pigmented Nevus under Deep Learning and the Correlation with Pathological Features.

The objective of this study was to explore the image classification and case characteristics of pigmented nevus (PN) diagnosed by dermoscopy under deep learning. 268 patients were included as the research objects and they were randomly divided into observation group (n = 134) and control group (n = 134). Image recognition algorithm was used for feature extraction, segmentation, and classification of dermoscopic images, and the image recognition and classification algorithm were studied as the performance and accuracy of fusion classifier were compared. The results showed that the classifier was optimized, and the linear kernel accuracy was 85.82%. The PN studied mainly included mixed nevus, junctional nevus, intradermal nevus, and acral nevus. The sensitivity under collaborative training was higher than that under feature training and fusion feature training, and the differences among three trainings were significant (P < 0.05). The sensitivity of the observation group was 88.65%, and the specificity was 90.26%, while the sensitivity and the specificity of the control group were 85.65% and 84.03%, respectively; there were significant differences between the two groups (P < 0.05). In conclusion, dermoscopy under deep learning could be applied as a diagnostic way of PN, which helped improve the accuracy of diagnosis. The dermoscopic manifestations of PN showed a certain corresponding relationship with the type of cases and could provide auxiliary diagnosis in clinical practice. It could be applied clinically.

Dermoscopic Image Classification with Neural Style Transfer

Skin cancer, the most commonly found human malignancy, is primarily diagnosed visually via dermoscopic analysis, biopsy and histopathological examination. However, unlike other types of cancer, automated image classification of skin lesions is deemed to be more challenging due to the irregularity and variability in the lesions’ appearances. In this work, we propose an adaptation of the Neural Style Transfer (NST) as a novel image pre-processing step for skin lesion classification problems. We represent each dermoscopic image as a style image and transfer the style of the lesion onto a homogeneous content image. This transfers the main variability of each lesion onto the same localized region, which allows us to integrate the generated images together and extract low-rank latent embeddings via tensor decomposition. We evaluated the performance of our model on competition data sets collected and pre-processed from the International Skin Imaging Collaboration (ISIC) database. We show that the classification performance based on the extracted tensor features using the style-transferred images significantly outperforms that of the raw images by more than 10%, and is also competitive with well-studied, pre-trained CNN models using transfer learning. Additionally, the tensor decomposition also affords clinical interpretations and insights by examining the images which correspond to the largest loadings in the top style embedding features as identified by the common supervised learning models.

Implementation of artificial intelligence algorithms for melanoma screening in a primary care setting.

Skin cancer is currently the most common type of cancer among Caucasians. The increase in life expectancy, along with new diagnostic tools and treatments for skin cancer, has resulted in unprecedented changes in patient care and has generated a great burden on healthcare systems. Early detection of skin tumors is expected to reduce this burden. Artificial intelligence (AI) algorithms that support skin cancer diagnoses have been shown to perform at least as well as dermatologists’ diagnoses. Recognizing the need for clinically and economically efficient means of diagnosing skin cancers at early stages in the primary care attention, we developed an efficient computer-aided diagnosis (CAD) system to be used by primary care physicians (PCP). Additionally, we developed a smartphone application with a protocol for data acquisition (i.e., photographs, demographic data and short clinical histories) and AI algorithms for clinical and dermoscopic image classification. For each lesion analyzed, a report is generated, showing the image of the suspected lesion and its respective Heat Map; the predicted probability of the suspected lesion being melanoma or malignant; the probable diagnosis based on that probability; and a suggestion on how the lesion should be managed. The accuracy of the dermoscopy model for melanoma was 89.3%, and for the clinical model, 84.7% with 0.91 and 0.89 sensitivity and 0.89 and 0.83 specificity, respectively. Both models achieved an area under the curve (AUC) above 0.9. Our CAD system can screen skin cancers to guide lesion management by PCPs, especially in the contexts where the access to the dermatologist can be difficult or time consuming. Its use can enable risk stratification of lesions and/or patients and dramatically improve timely access to specialist care for those requiring urgent attention.

Dermoscopic image classification using CNN with Handcrafted features

Globally skin cancer is one of the main cause of death in humans. Early diagnosis plays a major role in increasing the prevention of death rate caused due to any kind of cancer. Conventional diagnosis of skin cancer is a tedious and time-consuming process. To overcome this an automated skin lesion classification must develop. Automated skin lesion classification is a challenging task due to the fine-grained variability in the visibility of skin lesions. In this work dermoscopic images are obtained from the International Skin Image Collaboration Archive 2016 (ISIC 2016). In the proposed method the analysis and classification of skin lesions is done with the help of a Convolution Neural Network (CNN) along with the hand crafted features of dermoscopic image using Scattered Wavelet Transform as additional input to the fully connected layer of CNN, which leads to an improvement in the accuracy for identifying Melanoma and different skin lesion classification when compared to the other state of the art methods. When raw dermoscopic image is given as an input to the CNN and feature values of segmented dermoscopic image as input to the fully connected layer as an additional information, the proposed method gives a classification accuracy of 98.13% for identification of Melanoma and the accuracy achieved for classification of skin lesions is 93.14% for Melanoma Vs Nevus, 95.4% for Seborrheic Keratosis (SK) vs Squamous Cell Carcinoma (SCC), 96.87% for Melanoma vs Seborrheic Keratosis (SK), 95.65% for Melanoma vs Basal Cell Carcinoma (BCC), 98.5% for Nevus vs Basal Cell Carcinoma(BCC).

A deep analysis on high‐resolution dermoscopic image classification

Convolutional neural networks (CNNs) have been broadly employed in dermoscopic image analysis, mainly as a result of the large amount of data gathered by the International Skin Imaging Collaboration (ISIC). As in many other medical imaging domains, state-of-the-art methods take advantage of architectures developed for other tasks, frequently assuming full transferability between enormous sets of natural images (e.g. ImageNet) and dermoscopic images, which is not always the case. A comprehensive analysis on the effectiveness of state-of-the-art deep learning techniques when applied to dermoscopic image analysis is provided. To achieve this goal, the authors consider several CNNs architectures and analyse how their performance is affected by the size of the network, image resolution, data augmentation process, amount of available data, and model calibration. Moreover, taking advantage of the analysis performed, a novel ensemble method to further increase the classification accuracy is designed. The proposed solution achieved the third best result in the 2019 official ISIC challenge, with an accuracy of 0.593.

Analysis and Classification of Dermoscopic Images Using Spectral Graph Wavelet Transform

The abnormal growth of skin cells leads to skin cancer which occurs due to the unrepaired DNA impairment to the skin cells. Worldwide every year more than 1.23 lakhs of skin cancers are diagnosed, out of which Melanoma is the deadliest one. The aim of this research work is Recognition of Melanoma and skin lesion classification from the Dermoscopic images. Feature extraction of Dermoscopic images can be done by Shape, Color and Texture features. Texture features comprises of Gray Level Co-occurrence Matrix (GLCM) and statistical texture features calculated from the coefficients of the Multiresolution transforms such as Discrete Wavelet Transform (DWT), Curvelet, Tetrolet, and Spectral Graph Wavelet Transform (SGWT). The novelty in this work is using SGWT for extraction of texture features. The superiority of SGWT over conventional wavelet transform is its ability to work on irregular shaped images. Weighted graphs are the base for the SGWT which can be obtained in the form of meshes for irregular shapes. In the present work, skin lesions are obtained from the International Skin Imaging Collaboration (ISIC) 2016 archive. The features obtained from the Dermoscopic images are classified using Naïve Baye's, K-Nearest Neighbor (KNN) and Support Vector Machine (SVM) classifiers. The proposed method using Shape, Color and Texture based features for Melanoma Recognition with SGWT results in an Area Under the Curve (AUC) of 0.951 with Accuracy of 96.79 %, Sensitivity of 88 % and Specificity of 98.26 %. Further, the AUC of skin lesion classification such as Melanoma vs Nevus, Seborrheic Keratosis vs Squamous Cell Carcinoma, Melanoma vs Seborrheic Keratosis, Melanoma vs Basal Cell Carcinoma and Nevus vs Basal Cell Carcinoma using SGWT is 0.895, 0.945, 0.9645, 0.945 and 0.98 respectively.

基于FL-ResNet50的皮肤镜图像分类方法

基于FL-ResNet50的皮肤镜图像分类方法

Computer aided diagnostic support system for skin cancer: a review of techniques and algorithms.

Image-based computer aided diagnosis systems have significant potential for screening and early detection of malignant melanoma. We review the state of the art in these systems and examine current practices, problems, and prospects of image acquisition, pre-processing, segmentation, feature extraction and selection, and classification of dermoscopic images. This paper reports statistics and results from the most important implementations reported to date. We compared the performance of several classifiers specifically developed for skin lesion diagnosis and discussed the corresponding findings. Whenever available, indication of various conditions that affect the technique's performance is reported. We suggest a framework for comparative assessment of skin cancer diagnostic models and review the results based on these models. The deficiencies in some of the existing studies are highlighted and suggestions for future research are provided.