International Symposium On Biomedical Imaging Research Articles

Hierarchical agglomerative clustering-based skin lesion detection with region based neural networks classification

The skin cancer disease is widely spreading in the world due to abnormal increment of radiation. The segmentation of images is a crucial yet challenging aspect of the image processing process. It has emerged as a central focus in the study of image comprehension in recent years. Skin cancer is one of the critical diseases, which caused millions of deaths over the globe. The computer aided detection (CAD) of skin cancer can save the human lives. Further, the CAD models are resulted in poor segmentation and classification performance due to basic image processing, and machine learning models. Therefore, this work focused on development of skin lesion detection and classification (SLDC) using deep learning models. Initially, Guided image filtering (GIF) based preprocessing method, which performs noise removal, skin lesion enhancements, and basic hair removal operations. Then, skin lesion disease effected region is identified by using hierarchical agglomerative clustering (HAC). Then, multi-level features are extracted from segmented images. Finally, the region based neural network with long short term memory (RNN-LSTM) are used to perform the benign and malignant classes classification using trained features. The simulations conducted on International Symposium on Biomedical Imaging (ISBI) dataset shows that, the proposed RNN-SLDC system resulted in superior subjective and objective performance as compared to other approaches. The proposed method achieved 99.71% of segmentation accuracy, and 99.46% of classification accuracy.

Improved multiple sclerosis diagnosis with advanced deep learning techniques

AbstractWhite matter pathologies analysis is important for segmenting multiple sclerosis (MS) lesions in MRI slices. Myelin damage affects motor function, sensory conflicts, paralysis, and vision. Extracting an optimistic number of hidden characteristic features induces high feature maps and reduces redundancy and dimensionality. A framework based on Hybrid Deep Convolutional Neural Networks (HD‐CNN) is presented to classify MS based on the acquired optimistic feature characteristic. First, Chaotic Leader‐Selective Particle Swarm Optimization uses the pixel intensity ratio of the targeted region to extract the most hidden white matter features. Second, Refined Slime Mold categorizes hidden features and ensures that the necessary features are selected effectively for a better segmentation result. An effective inbuilt Maximum Bidirectional Gradient Classifier accurately classifies MS lesions by detecting white matter spots in the target region. Finally, we will evaluate the proposed HD‐CNN framework using the industry‐standard University Medical Centre Ljubljana (UMCL) and international symposium on biomedical imaging (ISBI) 2015 datasets as benchmarks. It reaches 98.56%, 93.15%, and 98.44% accuracy for FLAIR, T1W, and T2W, respectively. Our HD‐CNN improves MS classification over other state‐of‐the‐art techniques, compared with accuracy, F‐score, Recall, Precision, Dice, and Jaccard indexes values.

Segmentation of liver tumors with abdominal computed tomography using fully convolutional networks.

Dividing liver organs or lesions depicting on computed tomography (CT) images could be applied to help tumor staging and treatment. However, most existing image segmentation technologies use manual or semi-automatic analysis, making the analysis process costly and time-consuming. This research aims to develop and apply a deep learning network architecture to segment liver tumors automatically after fine tuning parameters. The medical imaging is obtained from the International Symposium on Biomedical Imaging (ISBI), which includes 3D abdominal CT scans of 131 patients diagnosed with liver tumors. From these CT scans, there are 7,190 2D CT images along with the labeled binary images. The labeled binary images are regarded as gold standard for evaluation of the segmented results by FCN (Fully Convolutional Network). The backbones of FCN are extracted from Xception, InceptionresNetv2, MobileNetv2, ResNet18, ResNet50 in this study. Meanwhile, the parameters including optimizers (SGDM and ADAM), size of epoch, and size of batch are investigated. CT images are randomly divided into training and testing sets using a ratio of 9:1. Several evaluation indices including Global Accuracy, Mean Accuracy, Mean IoU (Intersection over Union), Weighted IoU and Mean BF Score are applied to evaluate tumor segmentation results in the testing images. The Global Accuracy, Mean Accuracy, Mean IoU, Weighted IoU, and Mean BF Scores are 0.999, 0.969, 0.954, 0.998, 0.962 using ResNet50 in FCN with optimizer SGDM, batch size 12, and epoch 9. It is important to fine tuning the parameters in FCN model. Top 20 FNC models enable to achieve higher tumor segmentation accuracy with Mean IoU over 0.900. The occurred frequency of InceptionresNetv2, MobileNetv2, ResNet18, ResNet50, and Xception are 9, 6, 3, 5, and 2 times. Therefore, the InceptionresNetv2 has higher performance than others. This study develop and test an automated liver tumor segmentation model based on FCN. Study results demonstrate that many deep learning models including InceptionresNetv2, MobileNetv2, ResNet18, ResNet50, and Xception have high potential to segment liver tumors from CT images with accuracy exceeding 90%. However, it is still difficult to accurately segment tiny and small size tumors by FCN models.

Author's Reply to "MoNuSAC2020: A Multi-Organ Nuclei Segmentation and Classification Challenge".

We had released MoNuSAC2020 as one of the largest publicly available, manually annotated, curated, multi-class, and multi-instance medical image segmentation datasets. Based on this dataset, we had organized a challenge at the International Symposium on Biomedical Imaging (ISBI) 2020. Along with the challenge participants, we had published an article summarizing the results and findings of the challenge (Verma et al., 2021). Foucart et al. (2022) in their "Analysis of the MoNuSAC 2020 challenge evaluation and results: metric implementation errors" have pointed ways in which the computation of the segmentation performance metric for the challenge can be corrected or improved. After a careful examination of their analysis, we have found a small bug in our code and an erroneous column-header swap in one of our result tables. Here, we present our response to their analysis, and issue an errata. After fixing the bug the challenge rankings remain largely unaffected. On the other hand, two of Foucart et al.'s other suggestions are good for future consideration, but it is not clear that those should be immediately implemented. We thank Foucart et al. for their detailed analysis to help us fix the two errors.

Diagnosis of Melanoma Lesion Using Neutrosophic and Deep Learning

Melanoma is a kind of skin cancer which occurs due to too much exposure of melanocyte cells to the dangerous UV radiations, that gets damaged and multiplies uncontrollably. This is popularly known as malignant melanoma and is comparatively less heard of than certain other types of skin cancers; however it can be more detrimental as it swiftly spreads if not detected and attended at a primary stage. The differentiation between benign and melanocytic lesions sometimes may be confusing, but the symptoms of the disease can reasonably be discriminated by a profound investigation of its histopathological and clinical characteristics. In the recent past, Deep Convolutional Neural Networks (DCNNs) have advanced in accomplishing far better results. The necessity of the present day is to have faster and computationally efficient mechanisms for diagnosis of the deadly disease. This paper makes an effort to showcase a deep learning-based ‘Keras’ algorithm, which is established on the implementation of DCNNs to investigate melanoma from dermoscopic and digital pictures and provide swifter and more accurate result as contrasted to standard CNNs. The main highlight of this paper, basically stands in its incorporation of certain ambitious notions like the segmentation performed by a culmination of a moving straight line with a sequence of points and the application of the concept of triangular neutrosophic number based on uncertain parameters. The experiment was done on a total of 40,676 images obtained from four commonly available datasets— International Symposium on Biomedical Imaging (ISBI) 2017, International Skin Imaging Collaboration (ISIC) 2018, ISIC 2019 and ISIC 2020 and the end result received was indeed motivating. It attained a Jac score of 86.81% on ISIC 2020 dataset and 95.98%, 95.66% and 94.42% on ISBI 2017, ISIC 2018 and ISIC 2019 datasets, respectively. The present research yielded phenomenal output in most instances in comparison to the pre-defined parameters with the similar types of works in this field.

Image quality issues in teledermatology: A comparative analysis of artificial intelligence solutions

To the Editor: Deep learning artificial intelligence (AI) models capable of classifying skin lesions have become common in recent years. 1 Esteva A. Kuprel B. Novoa R.A. et al. Dermatologist-level classification of skin cancer with deep neural networks. Nature. 2017; 542: 115-118https://doi.org/10.1038/nature21056 Crossref PubMed Scopus (54) Google Scholar , 2 Codella N.C.F. Gutman D. Celebi M.E. et al. Skin lesion analysis toward melanoma detection: a challenge at the 2017 international symposium on biomedical imaging (ISBI), hosted by the international skin imaging collaboration (ISIC). in: 2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018). IEEE, 2018: 168-172 Crossref Scopus (407) Google Scholar , 3 Brinker T.J. Hekler A. Enk A.H. et al. Deep learning outperformed 136 of 157 dermatologists in a head-to-head dermoscopic melanoma image classification task. Eur J Cancer. 2019; 113: 47-54https://doi.org/10.1016/j.ejca.2019.04.001 Abstract Full Text Full Text PDF PubMed Scopus (146) Google Scholar , 4 Lopez A.R. Giro-i-Nieto X. Burdick J. Marques O. Skin lesion classification from dermoscopic images using deep learning techniques. in: 2017 13th IASTED international conference on biomedical engineering (BioMed). IEEE, 2017: 49-54 Google Scholar Many models are built using high-quality professional images from publicly available datasets, introducing biases to the data. 5 Bissoto A. Valle E. Avila S. Debiasing skin lesion datasets and models? Not so fast. in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops. IEEE, 2020: 740-741 Google Scholar However, with the rise in teledermatology, patient-recorded images are often taken in poor, unnatural lighting, increasing the likelihood of inadequate contrast, color, or exposure, which presents a challenge to those models.

An Efficient Deep Learning Based Coarse-to-Fine Cephalometric Landmark Detection Method

Accurate and automatic quantitative cephalometry analysis is of great importance in orthodontics. The fundamental step for cephalometry analysis is to annotate anatomic-interested landmarks on X-ray images. Computer-aided automatic method remains to be an open topic nowadays. In this paper, we propose an efficient deep learning-based coarse-to-fine approach to realize accurate landmark detection. In the coarse detection step, we train a deep learning-based deformable transformation model by using training samples. We register test images to the reference image (one training image) using the trained model to predict coarse landmarks' locations on test images. Thus, regions of interest (ROIs) which include landmarks can be located. In the fine detection step, we utilize trained deep convolutional neural networks (CNNs), to detect landmarks in ROI patches. For each landmark, there is one corresponding neural network, which directly does regression to the landmark's coordinates. The fine step can be considered as a refinement or fine-tuning step based on the coarse detection step. We validated the proposed method on public dataset from 2015 International Symposium on Biomedical Imaging (ISBI) grand challenge. Compared with the state-of-the-art method, we not only achieved the comparable detection accuracy (the mean radial error is about 1.0-1.6mm), but also largely shortened the computation time (4 seconds per image).

Curvelet and fast marching method‐based technique for efficient artifact detection and removal in dermoscopic images

AbstractEffective detection and removal of the artifacts from dermoscopic images require the analysis of structures within lesions for accurate cancer detection. But the dermoscopic images acquired may be affected by image acquisition noise, variation in texture, color, and irrelevant structure such as hairs, air bubbles, and contour blurring around the lesion. Hence preprocessing is very essential for accurate diagnosis. The proposed algorithm includes the preprocessing, to remove acquisition noise using curvelet transform, illumination correction using contrast limited adaptive histogram equalization (CLAHE) algorithm for image enhancement, Frangi vesselness algorithm is utilized for detection and removal of hairs and vessels like structures present in enhanced image, fast marching inpainting method is applied for repairing the information of lesion from removed hair pixels. The performance of the proposed algorithm is analyzed based on commonly used parameters such as diagnostic accuracy (DA), sensitivity (SE), specificity (SP), and precision (P). The performance is evaluated on PH2 and International Symposium on Biomedical Imaging (ISBI) datasets than compared with existing methods. The proposed algorithm segments the skin lesion with the highest diagnostic accuracy of 81.49% and sensitivity of 93.88% for the PH2 dataset and 75.39% and 79.34% respectively for the ISBI dataset. Results are better than existing methods. The proposed algorithm with improved preprocessing for artifact detection and removal is highly accurate and able to retrieve the hair occluded information.

PaI‐Net: A modified U‐Net of reducing semantic gap for surgical instrument segmentation

Tracking the instruments in a surgical scene is an essential task in minimally invasive surgery. However, due to the unpredictability of scenes, automatically segmenting the instruments is very challenging. In this paper, a novel method named parallel inception network (PaI-Net) is proposed, in which an attention parallel module (APM) and an output fusion module (OFM) are integrated with U-Net to improve the segmentation ability. Specially, APM utilizes multi-scale convolution kernels and global average pooling operations to extract semantic information and global context information of different scales, while OFM combines the feature maps of the decoder part to aggregate the abundant boundary information of shallow layers and the rich semantic information of deep layers together, which achieve a significant improvement in generating segmentation masks. Finally, the evaluation of proposed method on robotic instruments segmentation task from Medical Image Computing and Computer Assisted Intervention Society (MICCAI) and retinal image segmentation task from International Symposium on Biomedical Imaging (ISBI) show that our model has achieved advanced performance on multi-scale semantic segmentation and is superior to the current state-of-the-art models.

MoNuSAC2020: A Multi-Organ Nuclei Segmentation and Classification Challenge.

Detecting various types of cells in and around the tumor matrix holds a special significance in characterizing the tumor micro-environment for cancer prognostication and research. Automating the tasks of detecting, segmenting, and classifying nuclei can free up the pathologists' time for higher value tasks and reduce errors due to fatigue and subjectivity. To encourage the computer vision research community to develop and test algorithms for these tasks, we prepared a large and diverse dataset of nucleus boundary annotations and class labels. The dataset has over 46,000 nuclei from 37 hospitals, 71 patients, four organs, and four nucleus types. We also organized a challenge around this dataset as a satellite event at the International Symposium on Biomedical Imaging (ISBI) in April 2020. The challenge saw a wide participation from across the world, and the top methods were able to match inter-human concordance for the challenge metric. In this paper, we summarize the dataset and the key findings of the challenge, including the commonalities and differences between the methods developed by various participants. We have released the MoNuSAC2020 dataset to the public.

Deep Learning Approach for Medical Image Analysis

Localization of region of interest (ROI) is paramount to the analysis of medical images to assist in the identification and detection of diseases. In this research, we explore the application of a deep learning approach in the analysis of some medical images. Traditional methods have been restricted due to the coarse and granulated appearance of most of these images. Recently, deep learning techniques have produced promising results in the segmentation of medical images for the diagnosis of diseases. This research experiments on medical images using a robust deep learning architecture based on the Fully Convolutional Network- (FCN-) UNET method for the segmentation of three samples of medical images such as skin lesion, retinal images, and brain Magnetic Resonance Imaging (MRI) images. The proposed method can efficiently identify the ROI on these images to assist in the diagnosis of diseases such as skin cancer, eye defects and diabetes, and brain tumor. This system was evaluated on publicly available databases such as the International Symposium on Biomedical Imaging (ISBI) skin lesion images, retina images, and brain tumor datasets with over 90% accuracy and dice coefficient.

Melanoma Diagnosis Using Deep Learning and Fuzzy Logic.

Melanoma or malignant melanoma is a type of skin cancer that develops when melanocyte cells, damaged by excessive exposure to harmful UV radiations, start to grow out of control. Though less common than some other kinds of skin cancers, it is more dangerous because it rapidly metastasizes if not diagnosed and treated at an early stage. The distinction between benign and melanocytic lesions could at times be perplexing, but the manifestations of the disease could fairly be distinguished by a skilled study of its histopathological and clinical features. In recent years, deep convolutional neural networks (DCNNs) have succeeded in achieving more encouraging results yet faster and computationally effective systems for detection of the fatal disease are the need of the hour. This paper presents a deep learning-based ‘You Only Look Once (YOLO)’ algorithm, which is based on the application of DCNNs to detect melanoma from dermoscopic and digital images and offer faster and more precise output as compared to conventional CNNs. In terms with the location of the identified object in the cell, this network predicts the bounding box of the detected object and the class confidence score. The highlight of the paper, however, lies in its infusion of certain resourceful concepts like two phase segmentation done by a combination of the graph theory using minimal spanning tree concept and L-type fuzzy number based approximations and mathematical extraction of the actual affected area of the lesion region during feature extraction process. Experimented on a total of 20250 images from three publicly accessible datasets—PH2, International Symposium on Biomedical Imaging (ISBI) 2017 and The International Skin Imaging Collaboration (ISIC) 2019, encouraging results have been obtained. It achieved a Jac score of 79.84% on ISIC 2019 dataset and 86.99% and 88.64% on ISBI 2017 and PH2 datasets, respectively. Upon comparison of the pre-defined parameters with recent works in this area yielded comparatively superior output in most cases.

MMTDNN: Multi-view Massive Training Deep Neural Network for Segmentation and Detection of Abnormal Tissues in Medical Images

Purpose: Automated segmentation of abnormal tissues in medical images is considered as an essential part of those computer-aided detection and diagnosis systems which analyze medical images. However, automated segmentation of abnormalities is a challenging task due to the limitations of imaging technologies and complex structure of abnormalities, including low contrast between normal and abnormal tissues, shape diversity, appearance inhomogeneity, and the vague boundaries of abnormalities. Therefore, more intelligent segmentation techniques are required to tackle these challenges.
 Materials and Methods: In this study, a method, which is called MMTDNN, is proposed to segment and detect medical image abnormalities. MMTDNN, as a multi-view learning machine, utilizes convolutional neural networks in a massive training strategy. Moreover, the proposed method has four phases of preprocessing, view generation, pixel-level segmentation, and post-processing. The International Symposium on Biomedical Imaging (ISBI)-2016 dataset is used for the evaluation of the proposed method.
 Results: The performance of the proposed method has been evaluated on the task of skin lesion segmentation as one of the challenging applications of abnormal tissue segmentation. Both qualitative and quantitative results demonstrate outstanding performance. Meanwhile, the accuracy of 0.973, the Jaccard index of 0.876, and the Dice similarity coefficient of 0.931 have been achieved.
 Conclusion: In conclusion, the experimental result demonstrates that the proposed method outperforms stateof-the-art methods of skin lesion segmentation.

Convolutional descriptors aggregation via cross-net for skin lesion recognition

Malignant melanoma is one of the rare but deadliest types of skin cancers. Clinically, the early diagnosis of this disease is based on human visual inspection with dermoscopy imaging. However, human observations are subjective and prone to errors due to huge variations within dermoscopy images. To address it, we propose a framework for automatic skin lesion recognition using cross-net based aggregation of multiple convolutional networks. The output activation maps of each network are extracted as indicator maps to select the local deep convolutional descriptors (i.e., local patterns and color) in dermoscopy images. Also, this map of a convolutional layer captures the semantic regions of the input image and localizes the target object. These selected features are aggregated into an informative feature map, which are potentially better preserved in the convolutional feature maps. Finally, we use Fisher vector (FV) to encode the selected features. Extensive experiments demonstrate the effectiveness of our proposed method. Comparing with aggregation strategy using pooling approaches, the proposed method learns more robust and discriminative representations based on the publicly available skin lesion challenge datasets from the International Symposium on Biomedical Imaging (ISBI) 2016 and 2017.

Deep Learning-Based System for Automatic Melanoma Detection

Melanoma is the deadliest form of skin cancer. Distinguishing melanoma lesions from non-melanoma lesions has however been a challenging task. Many Computer Aided Diagnosis and Detection Systems have been developed in the past for this task. They have been limited in performance due to the complex visual characteristics of the skin lesion images which consists of inhomogeneous features and fuzzy boundaries. In this paper, we propose a deep learning-based method that overcomes these limitations for automatic melanoma lesion detection and segmentation. An enhanced encoder-decoder network with encoder and decoder sub-networks connected through a series of skip pathways which brings the semantic level of the encoder feature maps closer to that of the decoder feature maps is proposed for efficient learning and feature extraction. The system employs multi-stage and multi-scale approach and utilizes softmax classifier for pixel-wise classification of melanoma lesions. We devise a new method called Lesion-classifier that performs the classification of skin lesions into melanoma and non-melanoma based on results derived from pixel-wise classification. Our experiments on two well-established public benchmark skin lesion datasets, International Symposium on Biomedical Imaging(ISBI)2017 and Hospital Pedro Hispano (PH2), demonstrate that our method is more effective than some state-of-the-art methods. We achieved accuracy and dice coefficient of 95% and 92% on ISIC 2017 dataset and accuracy and dice coefficient of 95% and 93% on PH2 datasets.

Enhanced classifier training to improve precision of a convolutional neural network to identify images of skin lesions.

BackgroundIn recent months, multiple publications have demonstrated the use of convolutional neural networks (CNN) to classify images of skin cancer as precisely as dermatologists. However, these CNNs failed to outperform the International Symposium on Biomedical Imaging (ISBI) 2016 challenge which ranked the average precision for classification of dermoscopic melanoma images. Accordingly, the technical progress represented by these studies is limited. In addition, the available reports are impossible to reproduce, due to incomplete descriptions of training procedures and the use of proprietary image databases or non-disclosure of used images. These factors prevent the comparison of various CNN classifiers in equal terms.ObjectiveTo demonstrate the training of an image-classifier CNN that outperforms the winner of the ISBI 2016 CNNs challenge by using open source images exclusively.MethodsA detailed description of the training procedure is reported while the used images and test sets are disclosed fully, to insure the reproducibility of our work.ResultsOur CNN classifier outperforms all recent attempts to classify the original ISBI 2016 challenge test data (full set of 379 test images), with an average precision of 0.709 (vs. 0.637 of the ISBI winner) and with an area under the receiver operating curve of 0.85.ConclusionThis work illustrates the potential for improving skin cancer classification with enhanced training procedures for CNNs, while avoiding the use of costly equipment or proprietary image data.

Pathological Myopia Image Analysis Using Deep Learning

Abstract Pathological Myopia (PM) is one among the main reason of visual defect in the world. Pathological myopia is associated with decaying changes in the retina. If it remains untreated it may lead to vision loss that can’t be recoverable. The correct diagnosis of pathological myopia will facilitate proper treatment and improve disease management which reduce the growth of the disease. However, it is nearly impossible to scan the whole population. Computer-aided diagnosing tools in eye image study will build the method scalable and economical. This paper focuses on the problems of classification of Pathological myopia images and non-pathological myopia images and optical disc, fovea detection, localization, lesions (atrophy and detachment) segmentation with 400 samples provided by International Symposium on Biomedical Imaging (ISBI). In this paper, a deep learning method with Convolutional Neural Networks (CNN) is used for classification and U-net model is used for Image segmentation which shows that it achieves highly competitive results.

Segmentation of skin lesions in dermoscopy images using fuzzy classification of pixels and histogram thresholding

Background and objective: To ensure proper functioning of a Computer Aided Diagnosis (CAD) system for melanoma detection in dermoscopy images, it is important to accurately detect the border of the lesion. This paper proposes a method developed by the authors to address this problem.Methods: The algorithm for segmentation of skin lesions in dermoscopy images is based on fuzzy classification of pixels and subsequent histogram thresholding.Results: This method participated in the 2016 and 2017 ISBI (International Symposium on Biomedical Imaging) Challenges, hosted by the ISIC (International Skin Imaging Collaboration). It was tested against two public databases containing 379 and 600 images respectively, and compared using the same defined metrics (Accuracy, Dice Coefficient, Jaccard Index, Sensitivity and Specificity) with the rest of participating state-of-the-art work, obtaining good results: (0.934, 0.869, 0.791, 0.870 and 0.978) and (0.884, 0.760, 0.665, 0.869 and 0.923) respectively, ranking 9th and 15th out of a total of 21 and 28 participants respectively using the Jaccard Index (which was the indicator used as a basis for ranking) and the 1st in the 2017 Challenge using the Sensitivity.Conclusion: The method has been proven to be robust and reliable. It’s main contribution is the very design of the algorithm, highly innovative, which could also be used to deal with other segmentation problems of a similar nature.

Man against machine: diagnostic performance of a deep learning convolutional neural network for dermoscopic melanoma recognition in comparison to 58 dermatologists

BackgroundDeep learning convolutional neural networks (CNN) may facilitate melanoma detection, but data comparing a CNN’s diagnostic performance to larger groups of dermatologists are lacking. MethodsGoogle’s Inception v4 CNN architecture was trained and validated using dermoscopic images and corresponding diagnoses. In a comparative cross-sectional reader study a 100-image test-set was used (level-I: dermoscopy only; level-II: dermoscopy plus clinical information and images). Main outcome measures were sensitivity, specificity and area under the curve (AUC) of receiver operating characteristics (ROC) for diagnostic classification (dichotomous) of lesions by the CNN versus an international group of 58 dermatologists during level-I or -II of the reader study. Secondary end points included the dermatologists’ diagnostic performance in their management decisions and differences in the diagnostic performance of dermatologists during level-I and -II of the reader study. Additionally, the CNN’s performance was compared with the top-five algorithms of the 2016 International Symposium on Biomedical Imaging (ISBI) challenge. ResultsIn level-I dermatologists achieved a mean (±standard deviation) sensitivity and specificity for lesion classification of 86.6% (±9.3%) and 71.3% (±11.2%), respectively. More clinical information (level-II) improved the sensitivity to 88.9% (±9.6%, P=0.19) and specificity to 75.7% (±11.7%, P<0.05). The CNN ROC curve revealed a higher specificity of 82.5% when compared with dermatologists in level-I (71.3%, P<0.01) and level-II (75.7%, P<0.01) at their sensitivities of 86.6% and 88.9%, respectively. The CNN ROC AUC was greater than the mean ROC area of dermatologists (0.86 versus 0.79, P<0.01). The CNN scored results close to the top three algorithms of the ISBI 2016 challenge. ConclusionsFor the first time we compared a CNN’s diagnostic performance with a large international group of 58 dermatologists, including 30 experts. Most dermatologists were outperformed by the CNN. Irrespective of any physicians’ experience, they may benefit from assistance by a CNN’s image classification. Clinical trial numberThis study was registered at the German Clinical Trial Register (DRKS-Study-ID: DRKS00013570; https://www.drks.de/drks_web/).

Convolutional neural network classifier for distinguishing Barrett's esophagus and neoplasia endomicroscopy images.

Barrett's esophagus is a diseased condition with abnormal changes of the cells in the esophagus. Intestinal metaplasia (IM) and gastric metaplasia (GM) are two sub-classes of Barrett's esophagus. As IM can progress to the esophageal cancer, the neoplasia (NPL), developing methods for classifying between IM and GM are important issues in clinical practice. We adopted a deep learning (DL) algorithm to classify three conditions of IM, GM, and NPL based on endimicroscopy images. We constructed a convolutional neural network (CNN) architecture to distinguish among three classes. A total of 262 endomicroscopy imaging data of Barrett's esophagus were obtained from the international symposium on biomedical imaging (ISBI) 2016 challenge. 155 IM, 26 GM and 55 NPL cases were used to train the architecture. We implemented image distortion to augment the sample size of the training data. We tested our proposed architecture using the 26 test images that include 17 IM, 4 GM and 5 NPL cases. The classification accuracy was 80.77%. Our results suggest that CNN architecture could be used as a good classifier for distinguishing endomicroscopy imaging data of Barrett's esophagus.