Lung CT image segmentation is a key process in many applications such as lung cancer detection. It is considered a challenging problem due to existing similar image densities in the pulmonary structures, different types of scanners, and scanning protocols. Most of the current semi-automatic segmentation methods rely on human factors therefore it might suffer from lack of accuracy. Another shortcoming of these methods is their high false-positive rate. In recent years, several approaches, based on a deep learning framework, have been effectively applied in medical image segmentation. Among existing deep neural networks, the U-Net has provided great success in this field. In this paper, we propose a deep neural network architecture to perform an automatic lung CT image segmentation process. In the proposed method, several extensive preprocessing techniques are applied to raw CT images. Then, ground truths corresponding to these images are extracted via some morphological operations and manual reforms. Finally, all the prepared images with the corresponding ground truth are fed into a modified U-Net in which the encoder is replaced with a pre-trained ResNet-34 network (referred to as Res BCDU-Net). In the architecture, we employ BConvLSTM (Bidirectional Convolutional Long Short-term Memory)as an advanced integrator module instead of simple traditional concatenators. This is to merge the extracted feature maps of the corresponding contracting path into the previous expansion of the up-convolutional layer. Finally, a densely connected convolutional layer is utilized for the contracting path. The results of our extensive experiments on lung CT images (LIDC-IDRI database) confirm the effectiveness of the proposed method where a dice coefficient index of 97.31% is achieved.

Accurate segmentation and classification of brain tumors from Magnetic Resonance Imaging (MRI) remain key challenges in medical image analysis, primarily due to the lack of high-quality, balanced, and diverse datasets with expert annotations. In this work, we address this gap by introducing BRISC, a dataset designed for brain tumor segmentation and classification tasks, featuring high-resolution segmentation masks. The dataset comprises 6,000 contrast-enhanced T1-weighted MRI scans, which were collated from multiple public datasets that lacked segmentation labels. Our primary contribution is the subsequent expert annotation of these images, performed by certified radiologists and physicians. It includes three major tumor types, namely glioma, meningioma, and pituitary, as well as non-tumorous cases. Each sample includes high-resolution labels and is categorized across axial, sagittal, and coronal imaging planes to facilitate robust model development and cross-view generalization. To demonstrate the utility of the dataset, we provide benchmark results for both tasks using standard deep learning models. The BRISC dataset is made publicly available.

Melanoma is the deadliest type of cancerous cell, which is developed when melanocytes, melanin producing cell, starts its uncontrolled growth. If not detected and cured in its situ, it might decrease the chances of survival of patients. The diagnosis of a melanoma lesion is still a challenging task due to its visual similarities with benign lesions. In this paper, a fuzzy logic-based image segmentation along with a modified deep learning model is proposed for skin cancer detection. The highlight of the paper is its dermoscopic image enhancement using pre-processing techniques, infusion of mathematical logics, standard deviation methods, and the L-R fuzzy defuzzification method to enhance the results of segmentation. These pre-processing steps are developed to improve the visibility of lesion by removing artefacts such as hair follicles, dermoscopic scales, etc. Thereafter, the image is enhanced by histogram equalization method, and it is segmented by proposed method prior to performing the detection phase. The modified model employs a deep neural network algorithm, You Look Only Once (YOLO), which is established on the application of Deep convolutional neural network (DCNN) for detection of melanoma lesion from digital and dermoscopic lesion images. The YOLO model is composed of a series of DCNN layers we have added more depth by adding convolutional layer and residual connections. Moreover, we have introduced feature concatenation at different layers which combines multi-scale features. Our experimental results confirm that YOLO provides a better accuracy score and is faster than most of the pre-existing classifiers. The classifier is trained with 2000 and 8695 dermoscopic images from ISIC 2017 and ISIC 2018 datasets, whereas PH2 datasets along with both the previously mentioned datasets are used for testing the proposed algorithm.

The main objective of this work is to establish a framework for processing and evaluating the lower limb electromyography (EMG) signals ready to be fed to a rehabilitation robot. We design and build a knee rehabilitation robot that works with surface EMG (sEMG) signals. In our device, the muscle forces are estimated from sEMG signals using several machine learning techniques, i.e. support vector machine (SVM), support vector regression (SVR) and random forest (RF). In order to improve the estimation accuracy, we devise genetic algorithm (GA) for parameter optimisation and feature extraction within the proposed methods. At the same time, a load cell and a wearable inertial measurement unit (IMU) are mounted on the robot to measure the muscle force and knee joint angle, respectively. Various performance measures have been employed to assess the performance of the proposed system. Our extensive experiments and comparison with related works revealed a high estimation accuracy of 98.67% for lower limb muscles. The main advantage of the proposed techniques is high estimation accuracy leading to improved performance of the therapy while muscle models become especially sensitive to the tendon stiffness and the slack length.

In this article, an innovative deep learning-driven framework, adapted for the identification of diseases in strawberry plants, is proposed. Our approach encompasses a comprehensive embedded electronic system, incorporating sensor data acquisition and image capturing from the plants. These images are seamlessly transmitted to the cloud through a dedicated gateway for subsequent analysis. The research introduces a novel model, ResNet9-SE, a modified ResNet architecture featuring two squeeze-and-excitation (SE) blocks strategically positioned within the network to enhance performance. The key advantage gained is achieving fewer parameters and occupying less memory while preserving a high diagnosis accuracy. The proposed model is evaluated using in-house collected data and a publicly available dataset. The experimental outcomes demonstrate the exceptional classification accuracy of the ResNet9-SE model (99.7%), coupled with significantly reduced computation costs, affirming its suitability for deployment in embedded systems.

In recent years, the use of electroporation process has attracted much attention, due to its application in various industrial and medical fields. Electroporation is a microbiology technique which creates tiny holes in the cell membrane by the applied electric field. The electroporation process needs high-voltage pulses to provide the required electric field. To generate high-voltage pulses, a pulse generator device must be used. High-voltage pulse generators can be mainly divided into two major groups: Classical pulse generators and power electronics-based pulse generators. As their name suggests, the first group is associated with the primary and elementary pulse generators like Marx generators, and the second group is associated with the pulse generators that have been updated with the advancement of power electronics like Modular Multilevel Converters. These two major groups are also divided into several subgroups which are reviewed in detail in this paper. This study reviews the literature presented in the field of pulse power and pulse generators proper for the electroporation process and addresses their strengths and weaknesses. Several tables are provided to highlight and discuss the characteristics of each subgroup. Finally, a comparative study among different groups of pulse generators is performed which is followed by a classification performance analysis.