Special Issue "Recent Advances in Deep Learning for Medical Image Analysis"

Submission Deadline: 01 April 2021 (closed)
Guest Editors
Dr. Tallha Akram, COMSATS University Islamabad, Pakistan.
Prof. Dr. Yu-Dong Zhang, University of Leicester, UK.
Prof. Dr. Robertas Damaševičius, Kaunas University of Technology, Lithuania.
Dr. Muhammad Attique Khan, HITEC University Taxila, Pakistan.


Artificial intelligence showed a huge interest, especially in the area of medical imaging from the last three years. Due to the spread of medical imaging modalities such as Magnetic Resonance Imaging (MRI), Ultrasound, and Positron Emission Tomography (PET), Dermoscopic images, X-Ray images, Mammograms, and histological images, enormous amounts of data are being generated related to these medical domains. The data is generated in the form of some images and related to health informatics. However, the amount of this data is too large and difficult to use by employing classical techniques (i.e. hand crafted features). The question is that how we can use this big amount of biomedical data to build the automated system with better accuracy and less computational time. Also, how we can utilize this data to develop an automated system for better diagnosis of cancers such as brain tumor, skin cancer, lung cancer, stomach cancer, COVID19 infected patients, and breast cancer. To handle the large amount of biomedical data, researchers of computer vision used deep learning. However, they facing several issues (i.e. high data dimensionality and imbalanced datasets) and to these issues, the performance of system were degraded. Therefore, the most of existing solutions are based on the balanced datasets which is not a good option for the multiclass classification problem. Therefore, it is essential to develop some advanced deep learning techniques. Also, it is required to develop dimensionality reduction techniques to minimize the prediction time. Also, the less prediction time can be useful for real-time computerized system.

The aim of this special issue is to provide a diverse, but complementary set of solutions using deep learning for medical images. The solutions cover the above mentioned issues. We would also like to accept the new solutions but not limited to the following:
• Deep learning based features extraction for medical images
• Visualization of deep learning features for medical images
• Features selection using heuristic techniques for medical images
• Features selection using met heuristic techniques
• Deep learning features fusion
• Deep learning based biomedical images information fusion
• Transfer learning in medical imaging
• Features reduction techniques
• Theoretical analysis of deep learning for medical images
• Deep learning based segmentation of infected regions
• Semi-Supervised deep learning for medical imaging
• Semantic Segmentation for medical image analysis
• Multitask Learning for medical image analysis

Published Papers
  • Breast Lesions Detection and Classification via YOLO-Based Fusion Models
  • Abstract With recent breakthroughs in artificial intelligence, the use of deep learning models achieved remarkable advances in computer vision, ecommerce, cybersecurity, and healthcare. Particularly, numerous applications provided efficient solutions to assist radiologists for medical imaging analysis. For instance, automatic lesion detection and classification in mammograms is still considered a crucial task that requires more accurate diagnosis and precise analysis of abnormal lesions. In this paper, we propose an end-to-end system, which is based on You-Only-Look-Once (YOLO) model, to simultaneously localize and classify suspicious breast lesions from entire mammograms. The proposed system first preprocesses the raw images, then recognizes abnormal regions as… More
  •   Views:484       Downloads:386        Download PDF

  • Integrated CWT-CNN for Epilepsy Detection Using Multiclass EEG Dataset
  • Abstract Electroencephalography is a common clinical procedure to record brain signals generated by human activity. EEGs are useful in Brain controlled interfaces and other intelligent Neuroscience applications, but manual analysis of these brainwaves is complicated and time-consuming even for the experts of neuroscience. Various EEG analysis and classification techniques have been proposed to address this problem however, the conventional classification methods require identification and learning of specific EEG characteristics beforehand. Deep learning models can learn features from data without having in depth knowledge of data and prior feature identification. One of the great implementations of deep learning is Convolutional Neural Network… More
  •   Views:350       Downloads:301        Download PDF

  • Hybrid Segmentation Scheme for Skin Features Extraction Using Dermoscopy Images
  • Abstract Objective and quantitative assessment of skin conditions is essential for cosmeceutical studies and research on skin aging and skin regeneration. Various handcraft-based image processing methods have been proposed to evaluate skin conditions objectively, but they have unavoidable disadvantages when used to analyze skin features accurately. This study proposes a hybrid segmentation scheme consisting of Deeplab v3+ with an Inception-ResNet-v2 backbone, LightGBM, and morphological processing (MP) to overcome the shortcomings of handcraft-based approaches. First, we apply Deeplab v3+ with an Inception-ResNet-v2 backbone for pixel segmentation of skin wrinkles and cells. Then, LightGBM and MP are used to enhance the pixel segmentation… More
  •   Views:280       Downloads:281        Download PDF

  • Segmentation and Classification of Stomach Abnormalities Using Deep Learning
  • Abstract An automated system is proposed for the detection and classification of GI abnormalities. The proposed method operates under two pipeline procedures: (a) segmentation of the bleeding infection region and (b) classification of GI abnormalities by deep learning. The first bleeding region is segmented using a hybrid approach. The threshold is applied to each channel extracted from the original RGB image. Later, all channels are merged through mutual information and pixel-based techniques. As a result, the image is segmented. Texture and deep learning features are extracted in the proposed classification task. The transfer learning (TL) approach is used for the extraction… More
  •   Views:367       Downloads:290        Download PDF