Skin Cancer accounts for one-third of all diagnosed cancers worldwide. The prevalence of skin cancers have been rising over the past decades. In recent years, use of dermoscopy has enhanced the diagnostic capability of skin cancer. The accurate diagnosis of skin cancer is challenging for dermatologists as multiple skin cancer types may appear similar in appearance. The dermatologists have an average accuracy of 62% to 80% in skin cancer diagnosis. The research community has been made significant progress in developing automated tools to assist dermatologists in decision making. In this work, we propose an automated computer-aided diagnosis system for multi-class skin (MCS) cancer classification with an exceptionally high accuracy. The proposed method outperformed both expert dermatologists and contemporary deep learning methods for MCS cancer classification. We performed fine-tuning over seven classes of HAM10000 dataset and conducted a comparative study to analyse the performance of five pre-trained convolutional neural networks (CNNs) and four ensemble models. The maximum accuracy of 93.20% for individual model amongst the set of models whereas maximum accuracy of 92.83% for ensemble model is reported in this paper. We propose use of ResNeXt101 for the MCS cancer classification owing to its optimized architecture and ability to gain higher accuracy.

A multi-class skin Cancer classification using deep convolutional neural networks

Deep learning to find colorectal polyps in colonoscopy: A systematic literature review.

In 2015 we began a sub-challenge at the EndoVis workshop at MICCAI in Munich using endoscope images of ex-vivo tissue with automatically generated annotations from robot forward kinematics and instrument CAD models. However, the limited background variation and simple motion rendered the dataset uninformative in learning about which techniques would be suitable for segmentation in real surgery. In 2017, at the same workshop in Quebec we introduced the robotic instrument segmentation dataset with 10 teams participating in the challenge to perform binary, articulating parts and type segmentation of da Vinci instruments. This challenge included realistic instrument motion and more complex porcine tissue as background and was widely addressed with modifications on U-Nets and other popular CNN architectures. In 2018 we added to the complexity by introducing a set of anatomical objects and medical devices to the segmented classes. To avoid over-complicating the challenge, we continued with porcine data which is dramatically simpler than human tissue due to the lack of fatty tissue occluding many organs.

2018 Robotic Scene Segmentation Challenge

Convolutional neural networks (CNNs) deliver exceptional results for computer vision, including medical image analysis. With the growing number of available architectures, picking one over another is far from obvious. Existing art suggests that, when performing transfer learning, the performance of CNN architectures on ImageNet correlates strongly with their performance on target tasks. We evaluate that claim for melanoma classification, over 9 CNNs architectures, in 5 sets of splits created on the ISIC Challenge 2017 dataset, and 3 repeated measures, resulting in 135 models. The correlations we found were, to begin with, much smaller than those reported by existing art, and disappeared altogether when we considered only the top-performing networks: uncontrolled nuisances (i.e., splits and randomness) overcome any of the analyzed factors. Whenever possible, the best approach for melanoma classification is still to create ensembles of multiple models. We compared two choices for selecting which models to ensemble: picking them at random (among a pool of high-quality ones) vs. using the validation set to determine which ones to pick first. For small ensembles, we found a slight advantage on the second approach but found that random choice was also competitive. Although our aim in this paper was not to maximize performance, we easily reached AUCs comparable to the first place on the ISIC Challenge 2017.

/pdf/solo-or-ensemble-choosing-a-cnn-architecture-for-melanoma-1bg32hiaxw.pdf

Solo or Ensemble? Choosing a CNN Architecture for Melanoma Classification

Background and Objective Although automated Skin Lesion Classification (SLC) is a crucial integral step in computeraided diagnosis, it remains a challenging task due to inconsistency in texture, color, indistinguishable boundaries, and shapes Methods In this article, we propose an automatic and robust framework for the dermoscopic SLC named Dermoscopic Expert (DermoExpert) The DermoExpert consists of a preprocessing, a hybrid Convolutional Neural Network (hybrid-CNN), and transfer learning The proposed hybrid-CNN classifier consists of three distinct feature extractors, with the same input images, which are fused to achieve better-depth feature maps of the corresponding lesion Those distinct and fused feature maps are classified using the different fully connected layers, which are then ensembled to get a final prediction probability In the preprocessing, we use lesion segmentation, augmentation, and class rebalancing For boosting the lesion recognition, we have also employed geometric and intensity-based augmentation as well as the class rebalancing by penalizing the loss of the majority class and adding extra images to the minority classes Additionally, we leverage the knowledge from a pre-trained model, also known as transfer learning, to build a generic classifier, although small datasets are being used In the end, we design and implement a web application by deploying the weights of our DermoExpert for automatic lesion recognition Results We evaluate our DermoExpert on the ISIC-2016, ISIC-2017, and ISIC-2018 datasets, where our DermoExpert has achieved the area under the receiver operating characteristic curve (AUC) of 096, 095, and 097, respectively The experimental results outperform the recent state-of-the-art by a margin of 100 % and 20 % respectively for ISIC-2016 and ISIC-2017 datasets in terms of AUC The DermoExpert also outperforms, in concerning a balanced accuracy, by a margin of 30 % for ISIC-2018 dataset Conclusion Since our framework can provide better-classification on three different test datasets, even with limited training data, it can lead to better-recognition of melanoma to aid dermatologists Our source code, and segmented masks, for ISIC-2018 dataset, will be made publicly available for the research community for further improvements

/pdf/dermoexpert-skin-lesion-classification-using-a-hybrid-103cr56a9u.pdf

DermoExpert: Skin lesion classification using a hybrid convolutional neural network through segmentation, transfer learning, and augmentation

Colorectal polyps are important precursors to colon cancer, the third most common cause of cancer mortality for both men and women. It is a disease where early detection is of crucial importance. Colonoscopy is commonly used for early detection of cancer and precancerous pathology. It is a demanding procedure requiring significant amount of time from specialized physicians and nurses, in addition to a significant miss-rates of polyps by specialists. Automated polyp detection in colonoscopy videos has been demonstrated to be a promising way to handle this problem. {However, polyps detection is a challenging problem due to the availability of limited amount of training data and large appearance variations of polyps. To handle this problem, we propose a novel deep learning method Y-Net that consists of two encoder networks with a decoder network. Our proposed Y-Net method} relies on efficient use of pre-trained and un-trained models with novel sum-skip-concatenation operations. Each of the encoders are trained with encoder specific learning rate along the decoder. Compared with the previous methods employing hand-crafted features or 2-D/3-D convolutional neural network, our approach outperforms state-of-the-art methods for polyp detection with 7.3% F1-score and 13% recall improvement.

Y-Net: A deep Convolutional Neural Network for Polyp Detection

During video-guided minimally invasive surgery, quality of frames may be degraded severely by cauterization-induced smoke and condensation of vapor. This degradation of quality creates discomfort for the operating surgeon, and causes serious problems for automatic follow-up processes such as registration, segmentation and tracking. This paper proposes a novel deep neural network based smoke removal solution that is able to enhance the quality of surgery video frames in real-time. It employs synthetically generated training dataset including smoke embedded and clean reference versions. Results calculated on the test set indicate that our network outperforms previous defogging methods in terms of quantitative and qualitative measures. While eliminating apparent smoke, it also successfully preserves the natural appearance of tissue surface. To the best of our knowledge, the presented method is the first deep neural network based approach for the surgical field smoke removal problem.

/pdf/deep-smoke-removal-from-minimally-invasive-surgery-videos-1q34n8fm80.pdf

Deep Smoke Removal from Minimally Invasive Surgery Videos

In this report, we are presenting our automated prediction system for disease classification within dermoscopic images. The proposed solution is based on deep learning, where we employed transfer learning strategy on VGG16 and GoogLeNet architectures. The key feature of our solution is preprocessing based primarily on image augmentation and colour normalization. The solution was evaluated on Task 3: Lesion Diagnosis of the ISIC 2018: Skin Lesion Analysis Towards Melanoma Detection.

Ensemble of Convolutional Neural Networks for Dermoscopic Images Classification

The enforcement of the General Data Protection Regulation (GDPR) represents a great challenge for online social networks (OSNs). Several OSNs are making significant changes to their systems to achieve compliance with GDPR. OSNs are required to obtain meaningful consent from users to achieve GDPR compliance. GDPR recognizes user's consent as a legitimate ground for personal data processing in the context of online social networks. This article presents a comparative study about the criteria for valid consent under GDPR and existing consent seeking practices of OSNs. In order to simplify the comparative process, Facebook is taken as a case study for online social networks. In conclusion of the comparative study, we argue that existing consent mechanisms in OSNs are not GDPR compliant. To achieve GDPR compliance in online social networks, we advocate a blockchain-based approach for consent management. This paper paves the way for designing a blockchain-based GDPR compliant consent management model for personal data processing in online social networks.

GDPR Compliant Consent Driven Data Protection in Online Social Networks: A Blockchain-Based Approach

Surgical robot technology has revolutionized surgery toward a safer laparoscopic surgery and ideally been suited for surgeries requiring minimal invasiveness. Sematic segmentation from robot-assisted surgery videos is an essential task in many computer-assisted robotic surgical systems. Some of the applications include instrument detection, tracking and pose estimation. Usually, the left and right frames from the stereoscopic surgical instrument are used for semantic segmentation independently from each other. However, this approach is prone to poor segmentation since the stereo frames are not integrated for accurate estimation of the surgical scene. To cope with this problem, we proposed a multi encoder and single decoder convolutional neural network named StreoScenNet which exploits the left and right frames of the stereoscopic surgical system. The proposed architecture consists of multiple ResNet encoder blocks and a stacked convolutional decoder network connected with a novel sum-skip connection. The input to the network is a set of left and right frames and the output is a mask of the segmented regions for the left frame. It is trained end-to-end and the segmentation is achieved without the need of any pre- or post-processing. We compare the proposed architectures against state-of-the-art fully convolutional networks. We validate our methods using existing benchmark datasets that includes robotic instruments as well as anatomical objects and non-robotic surgical instruments. Compared with the previous instrument segmentation methods, our approach achieves a significant improved Dice similarity coefficient.

/pdf/streoscennet-surgical-stereo-robotic-scene-segmentation-4odg302bau.pdf

Sule Yildirim

Papers

Y-Net: A deep Convolutional Neural Network for Polyp Detection

Deep Smoke Removal from Minimally Invasive Surgery Videos

Ensemble of Convolutional Neural Networks for Dermoscopic Images Classification

GDPR Compliant Consent Driven Data Protection in Online Social Networks: A Blockchain-Based Approach

StreoScenNet: surgical stereo robotic scene segmentation