Brain tumors are also known as intracranial disease which is occurred due to cause of uncontrolled cell growth in the brain. Detecting and classifying the brain tumors at the initial stage plays a crucial role in saving the patient’s life. Radiologist uses MRI scans to identify and classify the various types of brain tumors in a manual approach. However, it is inaccurate and time-consuming with the large number of images. Among machine learning convolutional neural network (CNN) is one kind of significant algorithm that can extract the features automatically with high accuracy. The drawback of this algorithm is that it can extract features without knowing micro and macro features which occurs overfitting. The proposed architecture of Parallel CNN (PCNN) can extract the features by knowing the micro and macro features from two separate window sizes. At first, augmenting the normalized data using geometric transformation to enhance the number of images. Then, Micro and macro features are extracted using the proposed architecture PCNN alongside batch normalization to reduce the overfitting problem. Finally, three types of tumors glioma, meningioma, and pituitary are classified using various types of classifiers like Softmax, KNN, and SVM. The proposed DPCNN –SVM obtained the best accuracy 96.7% with the special features compared with the other pertained model.

Increased electrical activity in the brain causes epilepsy, which causes seizures resulting in various medical complications that can sometimes be fatal. Doctors use electroencephalography (EEG) for profiling and diagnosis of epilepsy. According to the World Health Organization (WHO), approximately 50 million people worldwide have epilepsy, making it one of the most common neurological disorders globally. This number represents about 0.7% of the global population. The conventional method of EEG analysis employed by medical professionals is a visual investigation that is time-consuming and requires expertise because of the variability in EEG signals. This paper describes a method for detecting epileptic seizures in EEG signals by combining signal processing and machine learning techniques. SVM and other machine learning techniques detect anomalies in the input EEG signal. To extract features, DWT is used for decomposition to sub-bands. The proposed method aims to improve the accuracy of the machine learning model while using as few features as possible.

The classification results show an accuracy of 100% with just one feature, Mean Absolute Value, from datasets A&E. With additional features, the overall accuracy remains high at 99%, with specificity and sensitivity values of 97.2% and 99.1%, respectively. These results outperform previous research on the same dataset, demonstrating the effectiveness of our approach. This research contributes to developing more accurate and efficient epilepsy diagnosis systems, potentially improving patient outcomes.

A detailed analysis of the inertial signals input is required when using deep learning models for Parkinson's Disease detection. This work explores the possibility of reducing the input size of the models studying the most appropriate frequency range and determines if it is feasible to evaluate subjects with different sensor locations than those used during training. For experimentation, 3.2-second windows are used to classify signals between Parkinson's patients and control subjects, applying Fast Fourier Transform to the inertial signals and following a Leave-One-Subject-Out Cross-Validation methodology over the PD-BioStampRC21 dataset. It has been observed that the frequency range of 0 to 5 Hz offers a classification accuracy rate of 75.75 ± 0.62% using the five available sensors for training and evaluation, which is close to the model's performance over the entire frequency range, from 0 to 15,625 Hz, which is 77.46 ± 0.60%. Regarding the information transfer between sensors located in different body parts, it was observed that training and evaluating the model using data from the right forearm resulted in an accuracy of 65.17 ± 0.69%. When the model was trained with data from the opposite forearm, the accuracy was similar, at 63.57 ± 0.69%. Likewise, comparable results were found when using data from the other forearm and when training and evaluating with opposite thighs, with accuracy reductions not exceeding 3%.

**Abstract**

This research presents the development and analysis of a microwave sensor designed with a microstrip band stop filter, aimed at applications in electrical engineering and food quality assessment. The sensor employs parallel coupled lines within the microstrip, integrating a band stop filter at 2.45 GHz on an FR4 substrate. The primary objective is to evaluate preserved fish samples to demonstrate the sensor's efficacy and applicability. Measurements were conducted using a KEYSIGHT model E5063A network analyzer, covering a frequency range from 0.1 GHz to 3 GHz. The analysis focuses on the frequency response of the insertion loss (S21) across specified frequencies. The results indicate a significant correlation between the percentage shift in the transmission coefficient and the frequency, even when the sample range was meticulously adjusted. These findings underscore the potential of microwave sensors in monitoring the physical properties of preserved food, particularly within food production and quality control processes. The sensor facilitates rapid and precise assessments of food properties, highlighting its broad applicability in various sectors of the food industry. Furthermore, this research contributes to the advancement of microwave technology, suggesting new pathways for future studies and applications in engineering and industrial contexts. The integration of microstrip technology with band stop filters in sensor design presents a novel approach that enhances the accuracy and efficiency of food quality monitoring systems. This study not only establishes a foundation for further technological developments but also emphasizes the interdisciplinary nature of modern engineering solutions, combining principles of electrical engineering with practical applications in the food industry. This innovative approach could lead to more sophisticated and reliable methods for ensuring food safety and quality.

The object movement monitoring in three-dimensional space in agriculture allows for adaptation to current needs or planned implementation as a result of predictive analysis. The 3D movement monitoring has the potential for particulate matter movement analysis during handling and processing. Our research employs an Inertial Measurement Unit (IMU) sensor for in-situ 3D tracking of particulate matter, allowing us to capture detailed movement patterns and enhance our understanding of particle dynamics during the particulate matter interactions with agricultural tools. The sensor utilizes bluetooth technology, a magnetometer modul LSM303 with accelerometer for triggering the sensor, 3-axis accelerometer and 3-axis gyroscope modul LSM6DSL, internal memory and processor unit, allowing it to accurately monitor and record changes in particulate matter movement continuously during the motion. The aim of this contribution was to develop an economically accessible smart sensor, implemented based on a battery-operated concept. The assembly requirement aimed to achieve optimal dimensions to minimize disturbance to particle flow. The influence of layers of particulate matter on the IMU sensor is less pronounced compared to the use of alternative tracing methods. This concept is effective for subsurface tracing beneath specific layers of particulate matter. As a result, XYZ trajectories describing the movement paths of particle clusters within the volume of particulate matter were obtained. These results can be used for predictive analysis of particulate matter behavior. Such findings can significantly contribute to a better understanding of particulate matter dynamics in interaction with tools and its impact on agricultural processes.

Human gesture recognition using electromyography (EMG) signals holds high potential in enhancing the functionality of human-machine interfaces, prosthetic devices, and sports performance analysis. This work proposes a gesture classification system based on electromyography. This system has been designed to improve the accuracy of forearm gesture classification by leveraging advanced signal processing and deep learning techniques to optimize classification accuracy. The system is composed of two main modules: a signal processing module able to perform several transforms (Short-Time Fourier Transform and Constant-Q-Transform) and a classification module based on Convolutional Neural Networks (CNNs). The dataset employed in this study "Latent Factors Limiting the Performance of sEMG-Interfaces" comprises EMG signals collected via a bracelet equipped with 8 distinct sensors, capable of capturing a wide range of forearm muscle activities. The experimental process is composed of two main phases. Firstly, we employed a k-fold cross-validation methodology to systematically assess and validate the model's performance across different subsets of the data for hyperparameter tunning. Secondly, the best system configuration was evaluated over a new subset reporting significant improvements. The baseline neural network architecture reported an accuracy of 85.0 ± 0.13 % in classifying gestures. Through rigorous hyperparameter tuning and the application of various mathematical transformations to the EMG features, we managed to enhance the classification accuracy to 90.0 ± 0.12 % (an absolute improvement of 5% compared to the baseline for a 5-class problem). When comparing to previous works, we improved the F-score from 85.5%, to 89.3% for a 4-class problem (left, right, up and down).

This paper describes the development of a Human Activity Recognition (HAR) system based on deep learning for classifying full body activities using inertial signals. The HAR system is divided in several modules: a preprocessing module for extracting relevant features from the inertial signals window by windows, a machine learning algorithm for classifying the windows and a postprocessing module for integrating the information along several windows. Regarding the preprocessing module, several transformations are implemented and evaluated. For the ML module, several algorithms are evaluated including several deep learning architectures. This evaluation has been carried-out over the HARTH dataset. This public dataset contains recordings from 22 participants wearing two 3-axial Axivity AX3 accelerometers for 2 hours in a free-living setting. Sixteen different activities were recorded and annotated accordingly. This paper describes the fine-tuning process of several machine learning algorithms and analyses their performance with different sets of activities. The best results show an accuracy of 90% and 93% for 12 and 9 activities respectively. These results have been compared to the results reported in previous worlks. To the author's knowledge, these analyses provide the best state of the art results over this public dataset. Additionally, this paper includes several analyses of the confusion between the different activities and the contribution of every accelerometer in the global performance.

The integration of functionalized fabrics within smart textiles enables the monitoring of a range of physiological parameters, including electrocardiogram (ECG), respiration, temperature, and moisture.

Textile pressure sensors represent a recent field area of development within the field of wearable technology and smart textiles.

These sensors are integrated into fabrics, thereby enabling the detection of pressure changes through touch or body movement, thus providing a seamless interface between the user and the digital world.

Textile pressure sensors are primarily developed using conductive fibers or yarns, which are woven into or coated upon fabrics.

The underlying electric principles are as follows: 1) the use of materials whose electrical resistance changes under pressure, or 2) the modification of the sensor geometry which results in a change in electrical capacitance.

The aforementioned materials comprise conductive polymers, carbon nanotubes, and metallic nanoparticles in a deposited ink. The integration of these materials into textiles can be achieved through a variety of techniques.

As the technology matures, the status of intellectual property (IP) rights assumes greater importance for stakeholders seeking to define an exploitation strategy for their innovations.

The objective of this paper is to provide a comprehensive analysis and benchmarking of the intellectual property rights (IPR) scenario for textile pressure sensors. Indeed, as the field progresses, it will be necessary to implement ongoing adaptations to IP strategies and legal frameworks in order to effectively address the emerging challenges and opportunities.

A number of patent databases have been employed in order to evaluate the patent landscape pertaining to textile pressure sensors. This has involved the utilization of a combination of keywords and classification codes.

A preliminary examination of the data yielded from Espacenet (across a total of 185 patent documents) indicates a greater prevalence of patents pertaining to capacitive sensors in comparison to those related to piezoresistive sensors.

Diabetes mellitus is a chronic metabolic disorder characterized by dysregulation of blood glucose, which can lead to a range of serious health complications if not properly managed. Continuous glucose monitoring (CGM) is a cutting-edge technology that tracks glucose levels in real-time, providing continuous and detailed information about glucose fluctuations throughout the days. The CGM data can be leveraged to train deep learning models forecasting blood glucose levels. Several deep learning based glucose prediction models have been developed for diabetes populations, but their generalizability to other populations such as prediabetic individuals remains largely unknown. Prediabetes is a condition where blood glucose levels are higher than normal but not yet high enough to be classified as diabetes. It is a critical stage where intervention can prevent the progression to type 2 diabetes. To fill in the knowledge gap, we developed Long Short-Term Memory (LSTM) glucose prediction models tailored for three distinct populations: type 1 diabetes (T1D), type 2 diabetes (T2D), and prediabetic (PRED) individuals. We evaluated the internal and external validity of these models. The results showed that the model constructed with the prediabetic dataset demonstrated the best internal and external validity in predicting glucose levels across all three test sets, achieving a normalized RMSE (NRMSE) of 0.21 mg/dL, 0.11 mg/dL, 0.25mg/dL when tested on the prediabetic, T1D, and T2D test sets, respectively.

Rapid urbanization causes a high concentration of human population and economic activities that leads to the changes in landscape and spatial growth of the cities. Landscape features play a key role in understanding land use and land (LULC) dynamics of urban areas. This work aims to analyse and quantify the changes in LULC over 24 years (1999 to 2023) in Lucknow District of India. It focuses on different land use types including Built-up Area, Cropland, Water Body, Vegetation, and Fallow Land, using USGS satellite imagery. Multi-temporal Landsat satellite data from the years 1999, 2008, 2015, and 2023 were employed to prepare LULC maps including major classes namely built-up area, cropland, water body, vegetation, and fallow land. Several landscape metrics like Number of Patches (NP), Patch Density (PD), Largest Patch Index (LPI), Landscape Shape Index (LSI), Edge Density (ED), and Total Edge (TE) were calculated to analyse spatial patterns and changes of LULC categories. The study revealed significant changes in the landscape of Lucknow District, characterized by variations in the extent and distribution of the land use categories. Key findings include a remarkable increase in built-up area from 9.04% in 1999 to 25.91% in 2023, and a decrease in vegetation from 26.01% in 1999 to 11.71% in 2023. The PD and ED showed an increased fragmentation, especially in built-up areas where PD increased from 9.18 patches/100 ha in 1999 to 11.85 patches/100 ha in 2023. The LPI for built-up areas significantly grew, indicating larger continuous urban regions. The findings of this study emphasize the importance of monitoring landscape changes using multi-temporal remote sensing images over urban landscapes. Analysing landscape metrics helps to understand the ongoing changes in LULC, providing essential information for effective sustainable land management practices.