The COVID-19 infection is increasing at a rapid rate, with the availability of limited number of testing kits. Therefore, the development of COVID-19 testing kits is still an open area of research. Recently, many studies have shown that chest Computed Tomography (CT) images can be used for COVID-19 testing, as chest CT images show a bilateral change in COVID-19 infected patients. However, the classification of COVID-19 patients from chest CT images is not an easy task as predicting the bilateral change is defined as an ill-posed problem. Therefore, in this paper, a deep transfer learning technique is used to classify COVID-19 infected patients. Additionally, a top-2 smooth loss function with cost-sensitive attributes is also utilized to handle noisy and imbalanced COVID-19 dataset kind of problems. Experimental results reveal that the proposed deep transfer learning-based COVID-19 classification model provides efficient results as compared to the other supervised learning models.

Deep Transfer Learning Based Classification Model for COVID-19 Disease.

Automated skin lesion segmentation and classification are two most essential and related tasks in the computer-aided diagnosis of skin cancer. Despite their prevalence, deep learning models are usually designed for only one task, ignoring the potential benefits in jointly performing both tasks. In this paper, we propose the mutual bootstrapping deep convolutional neural networks (MB-DCNN) model for simultaneous skin lesion segmentation and classification. This model consists of a coarse segmentation network (coarse-SN), a mask-guided classification network (mask-CN), and an enhanced segmentation network (enhanced-SN). On one hand, the coarse-SN generates coarse lesion masks that provide a prior bootstrapping for mask-CN to help it locate and classify skin lesions accurately. On the other hand, the lesion localization maps produced by mask-CN are then fed into enhanced–SN, aiming to transfer the localization information learned by mask-CN to enhanced-SN for accurate lesion segmentation. In this way, both segmentation and classification networks mutually transfer knowledge between each other and facilitate each other in a bootstrapping way. Meanwhile, we also design a novel rank loss and jointly use it with the Dice loss in segmentation networks to address the issues caused by class imbalance and hard-easy pixel imbalance. We evaluate the proposed MB-DCNN model on the ISIC-2017 and PH2 datasets, and achieve a Jaccard index of 80.4% and 89.4% in skin lesion segmentation and an average AUC of 93.8% and 97.7% in skin lesion classification, which are superior to the performance of representative state-of-the-art skin lesion segmentation and classification methods. Our results suggest that it is possible to boost the performance of skin lesion segmentation and classification simultaneously via training a unified model to perform both tasks in a mutual bootstrapping way.

A Mutual Bootstrapping Model for Automated Skin Lesion Segmentation and Classification

Skin lesion classification with ensembles of deep convolutional neural networks.

In this paper, we present a novel framework for dermoscopy image recognition via both a deep learning method and a local descriptor encoding strategy. Specifically, deep representations of a rescaled dermoscopy image are first extracted via a very deep residual neural network pretrained on a large natural image dataset. Then these local deep descriptors are aggregated by orderless visual statistic features based on Fisher vector (FV) encoding to build a global image representation. Finally, the FV encoded representations are used to classify melanoma images using a support vector machine with a Chi-squared kernel. Our proposed method is capable of generating more discriminative features to deal with large variations within melanoma classes, as well as small variations between melanoma and nonmelanoma classes with limited training data. Extensive experiments are performed to demonstrate the effectiveness of our proposed method. Comparisons with state-of-the-art methods show the superiority of our method using the publicly available ISBI 2016 Skin lesion challenge dataset.

Melanoma Recognition in Dermoscopy Images via Aggregated Deep Convolutional Features

The ABCD (asymmetry, border irregularity, colour and dermoscopic structure) rule of dermoscopy is a scoring method used by dermatologists to quantify dermoscopy findings and effectively separate melanoma from benign lesions. Automatic detection of the ABCD features and separation of benign lesions from melanoma could enable earlier detection of melanoma. In this study, automatic ABCD scoring of dermoscopy lesions is implemented. Pre-processing enables automatic detection of hair using Gabor filters and lesion boundaries using geodesic active contours. Algorithms are implemented to extract the characteristics of ABCD attributes. Methods used here combine existing methods with novel methods to detect colour asymmetry and dermoscopic structures. To classify lesions as melanoma or benign nevus, the total dermoscopy score is calculated. The experimental results, using 200 dermoscopic images, where 80 are malignant melanomas and 120 benign lesions, show that the algorithm achieves 91.25% sensitivity of 91.25 and 95.83% specificity. This is comparable to the 92.8% sensitivity and 90.3% specificity reported for human implementation of the ABCD rule. The experimental results show that the extracted features can be used to build a promising classifier for melanoma detection.

https://ietresearch.onlinelibrary.wiley.com/doi/epdf/10.1049/iet-ipr.2015.0385

Classification of malignant melanoma and benign skin lesions: implementation of automatic ABCD rule

This paper presents an approach that combines conventional image processing with deep learning by fusing the features from the individual techniques. We hypothesize that the two techniques, with different error profiles, are synergistic. The conventional image processing arm uses three handcrafted biologically inspired image processing modules and one clinical information module. The image processing modules detect lesion features comparable to clinical dermoscopy information—atypical pigment network, color distribution, and blood vessels. The clinical module includes information submitted to the pathologist—patient age, gender, lesion location, size, and patient history. The deep learning arm utilizes knowledge transfer via a ResNet-50 network that is repurposed to predict the probability of melanoma classification. The classification scores of each individual module from both processing arms are then ensembled utilizing logistic regression to predict an overall melanoma probability. Using cross-validated results of melanoma classification measured by area under the receiver operator characteristic curve (AUC), classification accuracy of 0.94 was obtained for the fusion technique. In comparison, the ResNet-50 deep learning based classifier alone yields an AUC of 0.87 and conventional image processing based classifier yields an AUC of 0.90. Further study of fusion of conventional image processing techniques and deep learning is warranted.

Deep Learning and Handcrafted Method Fusion: Higher Diagnostic Accuracy for Melanoma Dermoscopy Images

BACKGROUND/PURPOSE Computer-aided diagnosis of skin cancer requires accurate lesion segmentation, which must overcome noise such as hair, skin color variations, and ambient light variability. METHODS A biologically inspired geodesic active contour (GAC) technique is used for lesion segmentation. The algorithm presented here employs automatic contour initialization close to the actual lesion boundary, overcoming the 'sticking' at minimum local energy spots caused by noise artifacts such as hair. The border is significantly smoothed to mimic natural lesions. In addition, features that mimic biological parameters include spectral image subtraction and removal of peninsulas and inlets. Multiple boundary choices borders are created by parameter options used at different steps. These choices can allow future improvement over the basic default border. RESULTS The basic GAC algorithm was tested on 100 images (30 melanomas and 70 benign lesions), yielding a median XOR border error of 6.7%, comparable to the median inter-dermatologist XOR border error (7.4%), and lower than the gradient vector flow snake median XOR error of 14.2% on the same image set. On a difficult low-contrast border set of 1238 images, which included 350 non-melanocytic lesions, a median XOR error of 23.9% is obtained. CONCLUSION GAC techniques show promise in attaining the goal of automatic skin lesion segmentation.

Biologically inspired skin lesion segmentation using a geodesic active contour technique.

Purpose Algorithms employed for pigmented lesion segmentation perform poorly on dermoscopy images of basal cell carcinoma (BCC), the most common skin cancer. The main objective was to develop better methods for BCC segmentation. Methods Fifteen thresholding methods were implemented for BCC lesion segmentation. We propose two error metrics that better measure the type II error: Relative XOR Error and Lesion Capture Ratio. Results On training/test sets of 305 and 34 BCC images, respectively, five new techniques outperform two state-of-the-art methods used in segmentation of melanomas, based on the new error metrics. Conclusion The proposed algorithms, which include solutions for image vignetting correction and border expansion to achieve dermatologist-like borders, provide more inclusive and feature-preserving border detection, favoring better BCC classification accuracy, in future work.

Thresholding methods for lesion segmentation of basal cell carcinoma in dermoscopy images.

Background

Pink blush is a common feature in basal cell carcinoma (BCC). A related feature, semitranslucency, appears as smooth pink or orange regions resembling skin color. We introduce an automatic method for detection of these features based on smoothness and brightness. We also introduce a neighborhood correction method for texture area correction.



Methods

Smoothness and brightness were analyzed over four bands: luminance, red, green, and blue, then merged using variance-based dynamic thresholding. Dermoscopic images of 100 biopsy-proven BCCs and 254 competitive benign mimics were used to train the algorithm. Sixteen color and texture features were extracted from the automatically detected areas. The confusion matrix for the algorithm showed 15 classification errors in the training set for the 354 images: three errors in the BCC set and 12 errors in the benign set.



Results

Logistic regression analysis on a separate 1024-image test set was able to achieve good separation of BCC from benign lesions with an area under the receiver operating characteristic curve (ROC) of 0.878 and 0.877 using manually-created and automatically-generated BCC border masks, respectively.



Conclusion

This pilot study indicates that automatic detection of semitranslucent and pink blush areas in BCC is feasible using colors and first-order texture statistics.

Adaptable texture-based segmentation by variance and intensity for automatic detection of semitranslucent and pink blush areas in basal cell carcinoma.

Reda Kasmi

Papers

Classification of malignant melanoma and benign skin lesions: implementation of automatic ABCD rule

Deep Learning and Handcrafted Method Fusion: Higher Diagnostic Accuracy for Melanoma Dermoscopy Images

Biologically inspired skin lesion segmentation using a geodesic active contour technique.

Thresholding methods for lesion segmentation of basal cell carcinoma in dermoscopy images.

Adaptable texture-based segmentation by variance and intensity for automatic detection of semitranslucent and pink blush areas in basal cell carcinoma.