This study focuses on leveraging data-driven techniques to diagnose brain tumors through magnetic resonance imaging (MRI) images. Utilizing the rule of deep learning (DL), we introduce and fine-tune two robust frameworks, ResNet 50 and Inception V3, specifically designed for the classification of brain MRI images. Building upon the previous success of ResNet 50 and Inception V3 in classifying other medical imaging datasets, our investigation encompasses datasets with distinct characteristics, including one with four classes and another with two. The primary contribution of our research lies in the meticulous curation of these paired datasets. We have also integrated essential techniques, including Early Stopping and ReduceLROnPlateau, to refine the model through hyperparameter optimization. This involved adding extra layers, experimenting with various loss functions and learning rates, and incorporating dropout layers and regularization to ensure model convergence in predictions. Furthermore, strategic enhancements, such as customized pooling and regularization layers, have significantly elevated the accuracy of our models, resulting in remarkable classification accuracy. Notably, the pairing of ResNet 50 with the Nadam optimizer yields extraordinary accuracy rates, reaching 99.34% for gliomas, 93.52% for meningiomas, 98.68% for non-tumorous images, and 97.70% for pituitary tumors. These results underscore the transformative potential of our custom-made approach, achieving an aggregate testing accuracy of 97.68% for these four distinct classes. In a two-class dataset, Resnet 50 with the Adam optimizer excels, demonstrating better precision, recall, F1 score, and an overall accuracy of 99.84%. Moreover, it attains perfect per-class accuracy of 99.62% for ‘Tumor Positive’ and 100% for ‘Tumor Negative’, underscoring a remarkable advancement in the realm of brain tumor categorization. This research underscores the innovative possibilities of DL models and our specialized optimization methods in the domain of diagnosing brain cancer from MRI images.

Disease recognition has been revolutionized by autonomous systems in the rapidly developing field of medical technology. A crucial aspect of diagnosis involves the visual assessment and enumeration of white blood cells in microscopic peripheral blood smears. This practice yields invaluable insights into a patient’s health, enabling the identification of conditions of blood malignancies such as leukemia. Early identification of leukemia subtypes is paramount for tailoring appropriate therapeutic interventions and enhancing patient survival rates. However, traditional diagnostic techniques, which depend on visual assessment, are arbitrary, laborious, and prone to errors. The advent of ML technologies offers a promising avenue for more accurate and efficient leukemia classification. In this study, we introduced a novel approach to leukemia classification by integrating advanced image processing, diverse dataset utilization, and sophisticated feature extraction techniques, coupled with the development of TL models. Focused on improving accuracy of previous studies, our approach utilized Kaggle datasets for binary and multiclass classifications. Extensive image processing involved a novel LoGMH method, complemented by diverse augmentation techniques. Feature extraction employed DCNN, with subsequent utilization of extracted features to train various ML and TL models. Rigorous evaluation using traditional metrics revealed Inception-ResNet’s superior performance, surpassing other models with F1 scores of 96.07% and 95.89% for binary and multiclass classification, respectively. Our results notably surpass previous research, particularly in cases involving a higher number of classes. These findings promise to influence clinical decision support systems, guide future research, and potentially revolutionize cancer diagnostics beyond leukemia, impacting broader medical imaging and oncology domains.

The recent advances in Unmanned Aerial Vehicles (UAVs) technology permit to develop new usages for them. One of the current challenges is to operate UAVs as an autonomous swarm. In this domain we already proposed a new mobility model using Ant Colony Algorithms combined with chaotic dynamics (CACOC) to enhance the coverage of an area by a swarm of UAVs. In this paper we propose to consider this mobility model as waypoints for real UAVs. A control model of the UAVs is deployed to test the efficiency of the coverage of an area by the swarm. We have tested our approach in a realistic robotics simulator (V-Rep) which is connected with ROS. We compare the performance in terms of coverage using several metrics to ensure that this mobility model is efficient for real UAVs.

This paper focuses on the use of the Virtual Robotics Experimental Platform (V-REP) and the Robotics Operative System (ROS) working in parallel for design, test, and tuning of a vision based control system to command an Unmanned Aerial Vehicle (UAV). Here, is presented how to configure the V-REP and ROS to work in parallel, and the developed software in ROS for the pose estimation based on vision and for the design and use of a fuzzy logic control system. It is also explained how to interact with a virtual and a real quadrotor (QR) to command it for the specific task of aerial visual inspection task. The control system approach presented in this work is based on three fuzzy logic controllers (FLC) working in parallel on an external control loop based on the visual information. The three controllers were designed and tuned to command the vertical, longitudinal and lateral velocities of the UAV. The task to accomplish by the control system is to modify the position of the UAV in real time for the visual inspection of an object or specific parts of a structure. The virtual environment of the V-REP was used to tune manually the control system. Finally, the behavior of the tuned controllers was validated by a set of tests in a real environment with a quadrotor.

There is a growing need for precise and rapid detection methods in fields such as biomedical diagnostics, environmental monitoring, and chemical analysis. Surface plasmon resonance (SPR) sensors have been used for the detection and quantification of a wide range of analytes, including biomolecules, chemicals, and gases, in real-time. Despite the promising capabilities of SPR sensors, there remains a gap in creating a balance between having a large enough area to capture a significant number of analytes for detection and being small enough to ensure high sensitivity. This research aims to explore the design of a D-shaped SPR-based optical fiber sensor, focusing on the use of copper, gold, and silver thin films at optimized width and thickness of 10 µm and 45 nm, respectively, to improve the sensor’s performance. Employing a computational approach, this study examines the influence of the optimized width and refractive indices of metallic films on the sensor’s characteristics. The 10 µm width of the metallic thin film has been found to produce an optimal balance between the sensitivity and the dynamic range of the sensor. Leveraging on the ratio of the real and imaginary parts of the dielectric constant of the thin film metal provides insight into the optical properties and sensitivity at certain wavelengths. Within an analyte refractive index range of 1.37–1.42 and a wavelength range of 650–1200 nm, results indicate that silver outperforms gold and copper at the optimized width with a wavelength sensitivity, and detection accuracy of 12,300 nmRIU⁻¹, and 3.075, respectively. By optimizing the width of the metal thin film at 10 µm, a highly sensitive D-SPR is designed, allowing for enhanced sensor detection capabilities for a wide range of bioanalytes.

The current paper discusses the design, modeling and control of a Light weight robotic arm actuated by Shape Memory Alloy (SMA) actuators, usable for applications such as Aerial Manipulator. Compared to servo motor based robotic arm the proposed design has an added advantage of light weight and high force to mass ratio, but further introduces the problem of nonlinearities such as Hysteresis into the system. A nonlinear dynamic model of the hysteretic robotic arm is systematically developed to perform closed loop simulations. A Joint Space control is performed using Variable Structure Control and the closed loop performance is successfully verified by simulation studies.