Biomedical Engineering

Subtask 1 Biomedical Technologies

Main activity 1 Clinical applications of diagnostics and therapy in surgery (Ronja, Bibi)

The primary objective of the activity is the automatic segmentation of liver tissue from patient CT scans and the creation of a suitable 3D model for use in subsequent processes. This issue will be addressed in collaboration with physicians from FNO, VP7.

The project focuses on the design, implementation, and in vitro and in vivo testing of a high-frequency biopsy needle with puncture site treatment using high-frequency current to stop potential bleeding.

Currently, biopsy needles are used both in operating rooms and clinical outpatient clinics, guided through the abdominal wall into the liver, with a significant risk of subsequent bleeding into the abdominal cavity. Especially in outpatient procedures, where the needle puncture is guided using ultrasound (USG), the risk of post-procedure bleeding is minimized.

During the course of the project, emphasis will be placed on the use of electrosurgical methods, particularly radiofrequency techniques, for coagulating the puncture site of the biopsy needle. It is anticipated that a prototype of the instrument and electronics will be manufactured by a reputable prototype manufacturer, so that, in the event of successful tests, this product could be modified for potential subsequent preclinical animal testing.

Outputs:

The project will result in a two- to three-part concept of a biopsy needle utilizing high-frequency tissue stimulation to achieve coagulation. In the final phase of the project, the high-frequency biopsy needle will be tested for functionality and usability.

The activity will be implemented in several stages:

• Analysis of existing methods in surgery – liver biopsy and resection
• Development of a prototype for preclinical animal testing
• Development of an improved prototype based on further measurements
• Validation of results, intellectual property protection, and publication activities
• Preparation of a proposal for a clinical study

Main activity 2 Clinical applications of diagnosis and therapy in gastroenterology

This activity focuses on basic research and involves the design of a methodology for measuring and evaluating the electrical activity of the stomach to support the diagnosis of its physiological state. The output of the activity is a measurement methodology consisting of a defined minimal electrode layout, recording rules, and preprocessing methods, as well as a data evaluation methodology that outlines procedures for extracting classifiers and a fuzzy expert system designed for their evaluation.

The research will include in vitro abdominal measurement of the electrogastrographic (EGG) signal to verify the correlation between food intake records and the sensation of satiety, as well as long-term EGG recordings. The aim is to address this issue in collaboration with the Faculty of Medicine, University of Ostrava (LF OU), and University Hospital Ostrava (FNO), developing a technical solution for food intake limitation, measurement, and stimulation of gastric signals, and exploring alternative approaches to obesity management.

The activity will be implemented in several stages:

• Analysis of existing methods in cardiology and surgery
• Development of a prototype device for electrogastrography measurement
• Development of a gastroenterological monitoring method for food intake
• Validation of results, intellectual property protection, and publication activities

Main activity 3 Sensors for diagnosis and therapy

The activity encompasses two main research directions: whole-body analyses (impedance and tomographic) and the development of testers and simulators, specifically in the field of cardiotocography.

The activity will be implemented in several stages:

• Analysis of existing simulators and testers for medical instrumentation
• Analysis of the availability of individual components for testers/simulators
• Development of simulators and testers
• Validation of results, intellectual property protection, and publication activities

Subtask 2 Telemedicine Technologies and Systems

The aim of the subprogram is the development of new telemedicine solutions for remote monitoring of the health status of specific target groups in various areas of telemedicine application. Activities focus on the development and testing of new telemedicine technologies that enable the capture of biological signals and other manifestations of the human body, such as physical activity or behavioural changes, using measurements of both biological and non-biological signals provided by the surrounding environment of the monitored individual.

In the area of new technology development, the emphasis is on non-intrusive wearable technologies and methods for contactless measurement of biological signals. The focus is on telemedicine applications in various medical fields where long-term monitoring of biological signals is beneficial, particularly for chronically ill patients and the elderly.

The activity also includes the development of telemedicine solutions for therapeutic processes, such as in rehabilitation, orthotics, prosthetics, and other therapeutic areas.

Additionally, the activities involve the development of algorithms for automated evaluation of observed manifestations in both short-term and long-term monitoring, utilizing artificial intelligence approaches.

Main activity 1 Therapeutic telemedicine

Development of new biotelemetric sensor systems focused on therapy, implementing technical solutions into standard therapeutic aids, provides professionals—therapists—with a better overview of the rehabilitation process. This allows for more frequent adjustments tailored to the individual without the need for frequent in-person visits and patient check-ups. For patients, these solutions offer information on the progress of treatment, the accuracy, and the effectiveness of the therapeutic process, thereby increasing trust in the therapist and enhancing motivation to undergo rehabilitation.

The activity will be implemented in the following stages:

• Technologies for monitoring therapy in a home environment
• Smart orthotics
• Validation of results, intellectual property protection, and publication activities

Main activity 2 Diagnostic telemedicine

The field of diagnostic telemedicine is currently highly developed, particularly in the area of home diagnostic medical devices and wearable devices for non-diagnostic monitoring of biological signals. However, almost all known implemented technical solutions require some level of cooperation from the monitored individual, ranging from actively performing independent measurements of biological signals to the minimum requirement of regularly charging wearable devices. Within the project, we will focus on developing fully contactless sensor systems for monitoring basic biological signals, specifically radar systems.

The goal of the research activity is to develop radar systems capable of monitoring selected biological signals with appropriate quality in the natural environment of the monitored individual’s home, without restricting their physical activities. The research will focus both on technical solutions and the potential use of MIMO antenna systems, as well as on modern algorithms for preprocessing measured signals to eliminate random physical movements. The developed sensor systems will subsequently be tested and validated in various application scenarios, from monitoring diabetics and individuals with metabolic disorders to applications in remote home care for seniors or individuals in recovery.

The second area of development in telemetric diagnostic solutions within this project is closely related to physical activities and their impact on selected conditions, such as the manifestations of chronic diabetes. This involves the development of a unique wearable diagnostic device for measuring shear forces between the limb and the surface or footwear. A new technical solution will be designed to measure shear forces during walking in a diagnostic environment, such as a motion laboratory or, subsequently, a doctor’s office. The newly developed measurement system will enable early diagnosis and prediction of soft tissue damage in the foot, for instance, in cases of diabetes mellitus, as well as in other applications.

The activity will be implemented in the following stages:

• Development of methods for contactless monitoring of biological signals
• Collaboration with VP4
• Validation of results, intellectual property protection, and publication activities

Main activity 3 Activity of daily living

Monitoring daily living activities is a complex technical task that combines various sensor systems and approaches for analysing measured signals. It has a wide range of applications, from pure telemetric systems for monitoring behavioural changes in individuals long-term affected by chronic illnesses, detecting acute behavioural changes caused by sudden health issues, identifying dangerous situations such as falls in a home environment, to home automation systems designed to improve the quality of life for individuals with disabilities or the elderly. This sub-activity is therefore crucial for collaboration across the various work packages of the LERCO project.

Developing solutions for monitoring daily living activities requires specialized testing environments in the form of living labs, which allow for thorough validation of proposed technical solutions and algorithms in a nearly natural setting while also enabling precise measurement of monitored parameters. The project will utilize the living labs at VŠB-TUO, which support both the development and subsequent validation of new technical solutions as well as long-term testing in real-world living environments.

The first area under development within this activity is the methodology for monitoring and evaluating individuals' daily rhythms in applications defined in collaboration with personnel from other work packages of the LERCO project. These include monitoring the level of daily activities that affect a healthy lifestyle in relation to the surrounding environment and its quality, for instance, suitability for outdoor sports. Another scenario involves monitoring and supporting individuals with metabolic disorders to maintain the effectiveness of their treatment. Additionally, the activity focuses on tracking the progression of chronic illnesses in the elderly population. A standalone task involves fall detection using contactless measurement systems, which can be easily implemented in the homes of seniors or individuals at risk of falls, as well as in hospital environments.

All the application areas require specific sensor solutions and new algorithms for automated decision-making systems.

Moreover, new sensor solutions will be developed within this activity for monitoring non-biological parameters, specifically those related to environmental monitoring and quality, which are influenced by human activities. This will be followed by the development of home automation elements, enabling individuals with disabilities to control their living environment.

The activity will be implemented in the following stages:

• Assistive technologies to enhance quality of life and prevent critical situations in home (hospital) environments
• Monitoring activities through non-biological parameter sensing
• Validation of results, intellectual property protection, and publication activities

Subtask 3 Cardiovascular system

Main activity 1 New methods for cardiac output measurement (CO)

The aim of the research is to develop and validate a new diagnostic method for cardiac output measurement based on the dilution principle. The development of this method includes the creation of technical tools, specifically a continuous glucose sensor and a catheter for measurement within the bloodstream. According to initial research, the method achieves higher accuracy than existing dilution techniques. The activity will involve a comparison of this method with conventional cardiac output measurement methods and its preclinical validation on an animal model. The activity also includes the development of algorithms for accurate determination of cardiac output from the measured data.

The activity will be implemented in the following stages:

• Analysis of technical solutions for continuous glucose measurement
• Design and implementation of a sensor for continuous glucose measurement in whole blood
• Validation of the sensor prototype on an animal model
• Optimization of the continuous sensor for online application in a catheter
• Validation of the Glycodilution method on an animal model
• Validation of results, intellectual property protection, and publication activities

 Main activity 2 Measurement and processing of ECG in radiodiagnostics

The main activity focuses on two areas.

Development of Polymer and Capacitive Electrodes

Currently, so-called MRI-compatible electrodes (mostly carbon) and conductors (optical fibers) are available, which are non-ferromagnetic. However, these electrodes and conductors also have their limitations. During ECG signal acquisition in MRI examinations, interference is typically induced in the measured ECG signal, which can be completely distorted by this interference. The aim is to develop new biopotential electrodes and conductors made of materials that allow safe and high-quality ECG signal acquisition during not only MRI but also other imaging methods. The objective of this activity is to analyze polymer electrodes and verify their performance compared to standard electrodes in terms of signal quality during imaging methods and the level of generated image artifacts.

Myocardial Infarction Diagnosis

Insufficient blood supply to the heart—ischemic heart disease (IHD)—is one of the leading causes of mortality worldwide. Therefore, it is essential to address the use of vectorcardiography (VCG) as an additional electrocardiographic diagnostic method and its application for the automatic diagnosis of IHD. VCG offers a non-invasive, accessible, and cost-effective examination that can detect proper heart function, flag warning signs, assess the state of IHD during routine clinical examinations, and provide supplementary information alongside the conventional 12-lead ECG method. VCG quantitatively describes the heart’s electrical space (EPS) through a set of properties suitable for further processing and automatic evaluation of IHD using computational techniques and classification algorithms.

The primary goal is to design an algorithm for the automatic classification of myocardial infarction (MI) patients based on VCG recordings, utilizing the extensively studied Laufberger octant VCG theory for quantitative EPS description and applying cybernetic approaches to signal processing. Another goal is to create a database of verified physiological VCG recordings to enhance the classification capabilities of the proposed algorithms, enabling the detection of both acute and stable forms of IHD and providing detailed information on the myocardial condition, extent, and localization of IHD. Optimization of the algorithm is possible at all stages of the classification process discussed in the methodology, including data acquisition, preprocessing, feature selection, significant EPS properties, and classification methods, to refine electrocardiographic diagnostics.

The activity will be carried out in the following stages:

• Development of polymer and capacitive electrodes
• Diagnosis of myocardial infarction
• Validation of results, intellectual property protection, and publication activities

Subtask 4 Medical data processing

The research plan primarily focuses on the use of recent intelligent methods for processing image data in selected biomedical applications, particularly machine learning, image registration, and segmentation methods, which serve for the automatic analysis of image signals. Within this plan, methods will be developed for the creation of a Brain-Computer Interface (BCI) system for rehabilitation purposes, the development of an autonomous biometric system for patient identification in oncology, the development of a system for the automatic diagnosis of articular cartilage, and the creation of a system for liver virtualization during surgical procedures.

Main activity 1 Development of algorithms based on fuzzy logic

One of the essential operations in image processing is segmentation, which leads to the identification, subsequent quantification, and classification of objects of interest from medical images. These methods have the potential to fully replace the subjective annotation of target tissues in the context of clinical diagnostics. This activity will focus on developing modern methods for detecting and recognizing selected biological structures from medical images. The aim is to develop advanced region-based methods with elements of artificial intelligence, utilizing principles of fuzzy logic and machine learning, which have the potential for both the automatic detection of specific biological tissues and the interpretation of the corresponding tissue regions.

The activity also aims to develop hybrid methods that, in addition to detection, will enable the quantification of tissue characteristics based on selected geometric and intensity-based features. The development process will include robust testing of segmentation quality compared to the gold standard, as well as assessing the robustness of the methods against additive deterministic noise with dynamic intensity, to objectively evaluate segmentation performance under variable imaging conditions.

The activity will be carried out in the following stages:

• Review of the current state of knowledge in modelling biological tissue from medical images
• Design and implementation of algorithms for the detection and modelling of biological tissue from medical images
• Validation of algorithms for accuracy evaluation
• Validation of results, intellectual property protection, and publication activities

Main activity 2 Augmented Reality in Surgery

The application of augmented reality methods in surgery represents one of the modern trends in liver surgery planning. These methods aim to virtualize the surgical environment, areas of interest, the overall morphological structure of the liver, and the surgical tools used during the operation to simulate surgical procedures and individual tasks, thereby minimizing patient harm.

The goal of this activity is to design and implement a comprehensive software system capable of simulating these procedures as part of the surgical operation. In the first stage, a thorough in-depth review of the current state of scientific methods and procedures used in augmented reality applications for planning therapeutic surgical procedures in clinical practice will be conducted to design an optimal augmented reality system for liver surgery. The system is expected to utilize imaging data from clinical CT scans and real recordings from the operation.

The aim is to develop intelligent registration algorithms to align these recordings, enabling the surgeon to simultaneously virtualize morphological details of the area of interest within the real scene visible during surgery, including the simulation of surgical tools. This software solution will incorporate artificial intelligence methods to automatically extract the liver region from various imaging modalities. These models will then be used for the aforementioned image domain registration. The proposed tool for virtualizing surgical procedures will have broad applications in surgical simulation, offering extensive possibilities for use.

The activity will be implemented in the following stages (in some phases in close collaboration with VP7):

• Analysis of the current state of augmented reality in surgery
• Development of an augmented reality system for surgery
• Creation of a 3D liver model
• Optimization of algorithms and finalization of the 3D liver model
• Visualization using the augmented reality system
• Validation of results, intellectual property protection, and publication activities

Main activity 3 Image processing (ID patient on radiotherapy)

The aim of this activity is the design and implementation of an autonomous biometric system for identifying patients undergoing indicated therapeutic oncology treatment. It is common for oncology patients to experience apathy, disorientation, and difficulty communicating with medical staff. These challenges have led to the development of an autonomous biometric recognition system capable of classifying and identifying specific patients based on unique biometric features. This system will be linked to a database containing the patient’s information, enabling medical staff to quickly and accurately set the appropriate procedures for the prescribed oncology treatment.

The first phase of the activity will involve a review of the current state of knowledge in facial feature classification to select the optimal methods for designing the biometric system. A robust objective analysis of the system will be conducted, emphasizing robustness in the context of potential visual changes in facial features during oncology treatment to ensure high recognition efficiency under variable conditions. Robustness is a key advantage over conventional facial classification methods, which often rely on standardized images and struggle to adapt to dynamic changes in facial features.

This comprehensive biometric system will be implemented into an autonomous hardware solution based on FPGA architecture to accelerate computations and enable simple installation without concerns about the reliability or computational capacity of standard computers.

The activity will be carried out in the following stages (in some phases in close collaboration with VP7):

• Analysis of the current state of knowledge for patient ID in radiotherapy
• Design of a software system for patient classification in oncology
• Implementation of the patient ID system in oncology
• Validation of results, intellectual property protection, and publication activities

Main activity 4 Real time medical signal processing

The goal of this activity is to design an autonomous system that creates an interface between electrical signals originating from the brain and selected rehabilitation procedures, with the aim of optimizing the therapeutic effect and managing these procedures. The primary application area of this system will be the optimization of therapeutic procedures for patients with deficits following strokes, spinal cord injuries, degenerative nervous system diseases such as ALS, and as an extended therapeutic option for muscular dystrophy.

The activity will be carried out in the following stages:

• Analysis of existing recent methods for real-time biological signal processing
• Design and implementation of a real-time biological signal processing model
• Verification and testing of the model for practical application
• Intellectual property protection and publication activities