Melanoma, most threatening type of skin cancer, is on the rise. In this paper an implementation of a deep-learning system on a computer server, equipped with graphic processing unit (GPU), is proposed for detection of melanoma lesions. Clinical (non-dermoscopic) images are used in the proposed system, which could assist a dermatologist in early diagnosis of this type of skin cancer. In the proposed system, input clinical images, which could contain illumination and noise effects, are preprocessed in order to reduce such artifacts. Afterward, the enhanced images are fed to a pre-trained convolutional neural network (CNN) which is a member of deep learning models. The CNN classifier, which is trained by large number of training samples, distinguishes between melanoma and benign cases. Experimental results show that the proposed method is superior in terms of diagnostic accuracy in comparison with the state-of-the-art methods.

No abstract

No abstract

Abstract-Melanoma is amongst most aggressive types of cancer. However, it is highly curable if detected in its early stages. Prescreening of suspicious moles and lesions for malignancy is of great importance. Detection can be done by images captured by standard cameras, which are more preferable due to low cost and availability. One important step in computerized evaluation of skin lesions is accurate detection of lesion's region, i.e. segmentation of an image into two regions as lesion and normal skin. Accurate segmentation can be challenging due to burdens such as illumination variation and low contrast between lesion and healthy skin. In this paper, a method based on deep neural networks is proposed for accurate extraction of a lesion region. The input image is preprocessed and then its patches are fed to a convolutional neural network (CNN). Local texture and global structure of the patches are processed in order to assign pixels to lesion or normal classes. A method for effective selection of training patches is used for more accurate detection of a lesion's border. The output segmentation mask is refined by some post processing operations. The experimental results of qualitative and quantitative evaluations demonstrate that our method can outperform other state-of-the-art algorithms exist in the literature.Index Terms-Convolutional neural network, deep learning, medical image segmentation, melanoma, skin cancer.

Coronary artery disease (CAD) is the most common type of heart disease which is the leading cause of death all over the world. X-ray angiography is currently the gold standard imaging technique for CAD diagnosis. These images usually suffer from low quality and presence of noise. Therefore, vessel enhancement and vessel segmentation play important roles in CAD diagnosis. In this paper a deep learning approach using convolutional neural networks (CNN) is proposed for detecting vessel regions in angiography images. Initially, an input angiogram is preprocessed to enhance its contrast. Afterward, the image is evaluated using patches of pixels and the network determines the vessel and background regions. A set of 1,040,000 patches is used in order to train the deep CNN. Experimental results on angiography images of a dataset show that our proposed method has a superior performance in extraction of vessel regions.

No abstract

Medical image analysis, especially segmenting a specific organ, has an important role in developing clinical decision support systems. In cardiac magnetic resonance (MR) imaging, segmenting the left and right ventricles helps physicians diagnose different heart abnormalities. There are challenges for this task, including the intensity and shape similarity between left ventricle and other organs, inaccurate boundaries and presence of noise in most of the images. In this paper we propose an automated method for segmenting the left ventricle in cardiac MR images. We first automatically extract the region of interest, and then employ it as an input of a fully convolutional network. We train the network accurately despite the small number of left ventricle pixels in comparison with the whole image. Thresholding on the output map of the fully convolutional network and selection of regions based on their roundness are performed in our proposed post-processing phase. The Dice score of our method reaches 87.24% by applying this algorithm on the York dataset of heart images.

Lesion segmentation in skin images is an important step in computerized detection of skin cancer. Melanoma is known as one of the most life threatening types of this cancer. Existing methods often fall short of accurately segmenting lesions with fuzzy boarders. In this paper, a new class of fully convolutional network is proposed, with new dense pooling layers for segmentation of lesion regions in non-dermoscopic images. Unlike other existing convolutional networks, this proposed network is designed to produce dense feature maps. This network leads to highly accurate segmentation of lesions. The produced dice score here is 91.6% which outperforms state-of-the-art algorithms in segmentation of skin lesions based on the Dermquest dataset.Index Terms-skin cancer, melanoma, deep neural networks, dense pooling layer. IntroductionComputerized diagnosis of skin cancer is of great necessity and interest [1]. About 5.4 million new cases of skin cancer are detected in the USA every year. Most fatal types of skin cancer are melanoma, where 75% of deaths are related to [2]. Recently the incidence pattern of melanoma has shown a rapid increase. The rate of melanoma occurrence has tripled in the past 30 years [3]. In the USA, an estimated 87,110 newly detected cases and an estimated 9,730 melanoma-related deaths occurred only in 2017. A key point in the survival of patients is early detection of malignant skin lesions [4][5]. Patients, with melanoma detected in the early stages, have a 98% of five-year relative chance of survival. The survival rate is shown to be only 18% when melanoma is spread to the other parts of the body, which results in a life expectancy with a median of less than one year [2][3].The importance and variety of computerized methods for melanoma early detection are reviewed in [6][7]. There exist two main categories of images applied for computerized diagnosis of melanoma: Dermoscopic images, also known as microscopic images, which are captured by a special instrument named dermoscope and non-dermoscopic images, which are captured by conventional digital cameras and smartphones. Dermoscopic images contain more detailed information, while the nondermoscopic images have the advantage of ease of access. Dermoscopic images are not easily accessible. Authors in [8] indicates that less than 50% of the dermatologists in the United States use dermatoscopy. Jafari et al. [9] have used deep learning to detect melanoma in non-dermoscopic images.One of the most critical stages in the computerized study of melanoma is the accurate segmentation of skin lesions. Captured images usually contain a lesion, which is surrounded by healthy skin. Proper extraction of the lesion region is critical for assessment of lesion's features. Lesion features include area, border irregularity, shape symmetry, and variation of color. There exist various methods for skin lesions segmentation, based on active contours, region merging [10], and thresholding [11]. Nondermoscopic images exhibit what is seen by the naked eye. There are some chal...