Ultrasonic tomography (UST) represents a powerful non-invasive diagnostic technique for monitoring and analyzing internal processes within industrial reactors. Despite its potential, UST-based reconstructions are often challenged by the ill-posed nature of the inverse problem, limited measurements, and the presence of noise. To address these limitations, this study introduces a novel differential neural network architecture that enhances conventional deep learning models by incorporating a specialized differential layer. This layer processes two parallel input streams and operates on their residuals, thereby amplifying subtle variations in the data critical for accurate tomographic reconstructions. This study aims to empirically validate the concept of the efficacy of differentiated architecture. Reconstruction performance was evaluated using established quantitative metrics. Results demonstrate that models incorporating the differential layer consistently outperform their standard counterparts, delivering higher resolution, and superior noise robustness.
The project aims to develop an intelligent system that has the potential to significantly improve patient care by examining and supporting the emotional state of people with chronic diseases or undergoing rehabilitation. The system will use advanced technologies such as image analysis, natural language processing and understanding (NLP/NLU), and a personal assistant module. Acting as a virtual companion, this module will support the user in everyday tasks such as setting reminders and managing schedules and provide emotional support through interactive conversations. The project also involves the development of an analytical module that automatically generates analyses and reports based on the collected data. A patient-oriented system will be created during the work to collect essential data and support him in rehabilitation. On the other hand, the system will cooperate with a doctor who can make a preliminary diagnosis and develop further treatment based on the patient's data. As feedback, the patient will receive health reports along with their interpretation and treatment recommendations created by the doctor.
15 December 2019 in Informatyka, Automatyka, Pomiary w Gospodarce i Ochronie Środowiska
The article presents a solution based on a cyber-physical system in which data collected from measuring sensors was analysed for prediction in the production process control system. The presented technology was based on intelligent sensors as part of the solution for Industry 4.0. The main purpose of the work is to reduce data and select the appropriate covariate to optimise modelling of defects using the Cox model for a specific mechanical system. The reliability of machines and devices in the production process is a condition for ensuring continuity of production. Predicting damage, especially its movement, gives the ability to monitor the current state of the machine. In a broader perspective, this enables streamlining the production process, service planning or control. This ensures production continuity and optimal performance. The presented model is a regressive survival analysis model that allows you to calculate the probability of failure occurring over a given period of time.
The aim of the research was to develop a new device for monitoring and diagnosing functional disorders of the lower urinary tract, which will enable measurement of muscle tension (EMG) and electrical impedance tomography (EIT) with the possibility of electrostimulation of bladder reconstruction.
The article presents the capabilities of the LETS Web system, which uses Internet of Things (IoT) technology to analyze medical data to optimize diagnostic processes. The article focuses on implementing decision tree algorithms that analyze electrocardiogram (ECG) data to identify cardiac conditions. The study used three variants of decision trees that differed in structure and ECG parameters. Each variant was tested for its ability to accurately classify cardiac health conditions ranging from simple arrhythmias to complex arrhythmic changes. The study showed that modifications to the structure of the decision trees significantly affected their effectiveness. The most advanced variant of the tree, using multivariate data analysis, showed the highest efficiency in diagnosing complex conditions. The effectiveness of the different variants of decision trees varied, confirming the importance of selecting the suitable diagnostic model for the specifics of the data and clinical goals.