Skin cancer is the most common type of cancer in the world and it is more treatable if diagnosed early. The diagnosis process usually starts with segmenting the skin lesion area and planning a follow-up treatment by the dermatologists. Thus, the segmentation process plays a critical role in the treatment process. In recent years, machine learning methods, especially deep convolutional neural networks are proposed to address the segmentation challenge. The common segmentation methods (e.g., U-Net) deploy a series of encoding blocks to model the local representation and subsequently a series of decoding blocks to capture the semantic relation. However, these structures are usually limited to model multi-scale objects with large variations in texture and shape. To address these limitations, we propose a Multi-Scale Attention U-Net (MSAU-Net) for skin lesion segmentation. In particular, we improve the typical U-net by inserting an attention mechanism at the bottleneck of the network to model the hierarchical representation. The attention module aggregates the multi-level representation in a non-linear fashion to selectively adjust the representative features. Then it deploys a bidirectional convolutional LSTM structure to fetch the common discriminative features and suppress the less informative ones. We incorporate the resulted features in each block of the decoding path to highlight the important regions. We have evaluated our proposed network in three public skin lesion datasets, including ISIC 2017, ISIC 2018, and PH2 datasets. The experimental results demonstrate that the proposed pipeline outperforms the existing alternatives.

Diabetic Retinopathy (DR) is a disease caused by a high level of glucose in retina vessels. This malicious disease put millions of people around the world at risk for vision loss each year. Being a life-threatening disease, early diagnosis can be an effective step in the treatment and prevention of vision loss. To automate the early diagnosis process, computer-aided diagnosis methods are not only useful in detecting the diabetic signatures but also provide information regarding the diabetic grade for the optometrist to determine an appropriate treatment. Several deep classification models are proposed in the literature to solve the diabetic retinopathy classification task, however, these methods usually lack incorporate an attention mechanism to better encode the semantic dependency and highlight the most important region for boosting the model performance. To overcome these limitations, we propose to incorporate a style and content recalibration mechanism inside the deep neural network to adaptively scale the informative regions for diabetic retinopathy classification. In our proposed method, the input image passes through the encoder module to encode both high-level and semantic features. Next, by utilizing a content and style separation mechanism, we decompose the representational space into a style (e.g., texture features) and content (e.g., semantic and contextual features) representation. The texture attention module takes the style representation and applies a high-pass filter to highlight the texture information while the spatial normalization module uses a convolutional operation to determine the more informative region inside the retinopathy image to detect diabetic signs. Once the attention modules are applied to the representational features, the fusion module combines both features to form a normalized representation for the decoding path. The decoder module in our model performs both diabetic grading and healthy, non-healthy classification tasks. Our experiment on APTOS Kaggle dataset (accuracy 0.85) demonstrates a significant improvement compared to the literature work. This fact reveals the applicability of our method in a real-world scenario.

An automatic skin lesion segmentation algorithm not only facilitates the dermatologist’s workload on skin cancer analysis but also provides a platform for early cancer prediction. Over the years, several deep learning methods have been proposed to address the skin lesion segmentation problem. However, training deep models usually requires a large-scale annotated dataset, which is not feasible in the medical domain due to the annotation burden. In addition, the low data regime highly increases the overfitting potential for the neural network. To address these limitations in an end-to-end manner, we propose to incorporate unlabelled samples during the training process. Our network offers a semi-supervised training schema, wherein the first stage performs a supervised training strategy to learn semantic segmentation map while the second step focuses on the unsupervised technique to enrich the encoder module. Specifically, unlike the literature work on skin lesion segmentation, we design a surrogate task on top of the convolutional and Transformer representations to learn data-driven features from the image itself to alleviate the requirement of the large annotated dataset. The effectiveness of the proposed method is demonstrated using three different skin lesion segmentation datasets, namely ISIC 2018 (dice score 0.905), ISIC 2017 (dice score 0.898) and PH2 (dice score 0.940). Particularly we observed that including the unsupervised samples can increase the dice score by 2%.

Deep learning plays a vital role in precise grapevine disease detection, yet practical applications for farmer assistance are scarce despite promising results. The objective of this research is to develop an intelligent approach, supported by user-friendly, open-source software named AI GrapeCare (Version 1, created by Osama Elsherbiny). This approach utilizes RGB imagery and hybrid deep networks for the detection and prevention of grapevine diseases. Exploring the optimal deep learning architecture involved combining convolutional neural networks (CNNs), long short-term memory (LSTM), deep neural networks (DNNs), and transfer learning networks (including VGG16, VGG19, ResNet50, and ResNet101V2). A gray level co-occurrence matrix (GLCM) was employed to measure the textural characteristics. The plant disease detection platform (PDD) created a dataset of real-life grape leaf images from vineyards to improve plant disease identification. A data augmentation technique was applied to address the issue of limited images. Subsequently, the augmented dataset was used to train the models and enhance their capability to accurately identify and classify plant diseases in real-world scenarios. The analyzed outcomes indicated that the combined CNN_RGB-LSTM_GLCM deep network, based on the VGG16 pretrained network and data augmentation, outperformed the separate deep network and nonaugmented version features. Its validation accuracy, classification precision, recall, and F-measure are all 96.6%, with a 93.4% intersection over union and a loss of 0.123. Furthermore, the software developed through the proposed approach holds great promise as a rapid tool for diagnosing grapevine diseases in less than one minute. The framework of the study shows potential for future expansion to include various types of trees. This capability can assist farmers in early detection of tree diseases, enabling them to implement preventive measures.

Automatic medical image segmentation is an essential step toward accurate diseases diagnosis and designing a follow-up treatment. This assistive method facilitates the cancer detection process and provides a benchmark to highlight the affected area. The U-Net model has become the standard design choice. Although the symmetrical structure of the U-Net model enables this network to encode rich semantic representation, the intrinsic locality of the CNN layers limits this network’s capability in modeling long-range contextual dependency. On the other hand, sequence to sequence Transformer models with a multi-head attention mechanism can enable them to effectively model global contextual dependency. However, the lack of low-level information stemming from the Transformer architecture limits its performance for capturing local representation. In this paper, we propose a two parallel encoder model, where in the first path the CNN module captures the local semantic representation whereas the second path deploys a Transformer module to extract the long-range contextual representation. Next, by adaptively fusing these two feature maps, we encode both representations into a single representative tensor to be further processed by the decoder block. An experimental study demonstrates that our design can provide rich and generic representation features which are highly efficient for a fine-grained semantic segmentation task.

The image of the tolerant religion of Islam has been distorted by extremists in the last two decades in many ways, such as luring teenagers into terrorist acts. Nowadays, millions of users socialize and share ideas using social media platforms such as Twitter. Typically, the ideas shared on Twitter (tweets) reach and influence many people who could simply retweet them and make them even spread faster. Unfortunately, some of these ideas are posted by extremists who share hateful Arabic content. Thus, it is very important to automate the process of controlling and monitoring hateful Arabic tweets, given that Arabic is the most widely used language in the Islamic world. In this paper, we provide a manually labeled and curated dataset of 3,000 Arabic tweets that contain hateful and non-hateful tweets. To automate the process of detecting hateful tweets, we utilize advanced Machine Learning (ML) techniques and perform sentiment analysis to capture the meaning of the Arabic words in a proper word embedding (Word2Vec). Also, we used the proposed model to classify and analyze 100,000 tweets of the last decade. The outcome of this work promotes future research on analyzing Arabic hateful speech by providing a manually labeled Arabic dataset, and the trained model (achieved 92% accuracy) which can be used as an underlying tool by governments, Internet service providers, and social media applications to detect any inflammatory tweets before they spread to a wider audience.