====== Medical Technology Group ====== ~~NOCACHE~~ ===== Research ===== Our research in the field of medical engineering focuses on the interdisciplinary interface of intensive care medicine, data analysis, machine learning, control engineering and embedded software. Here, we are particularly interested in data-based early detection of disease patterns and device malfunctions, as well as model-based control of organ support systems. {{:forschung:medtechsynergienen_en_cut.png?nolink575|}} \\ ===== Projects ===== ==== Aix-Neo-Guard ==== ### The digitalisation of the intensive care environment enables the consolidation of patient data from diagnostics and therapy, resulting in multi-parametric, high-frequency data collections. While such databases have already been established for adults and are publicly accessible, there is a lack of data in the field of neonatal and paediatric intensive care medicine for systemically modelling a wide range of clinical pictures or using artificial intelligence (AI) methods. {{:forschung:ang_logo_1.png?nolink&170 |}}The AIx-Neo-Guard project, which uses and expands the data set generated by a previous Nanni project, has the overarching goal of improving intensive care diagnostics and therapy in neonatal, paediatric and adolescent medicine by increasing treatment safety and improving medical training. The use of AI-based algorithms enables the early detection of treatment complications and provides deeper insights into pathophysiological relationships. In addition, a physiological model for modelling pulmonary gas exchange under mechanical ventilation is being developed, which could directly support both the training of medical staff and therapy in the future. As a first step, AI methods such as random forest or recurrent neural networks have already been applied to high-resolution time series data in order to automatically recognise patient-ventilator asynchronies. \\ \\ \\ Contact: [[:en:lehrstuhl:mitarbeiter:oprea]] \\ \\ ### ==== Explainable AI ==== In the context of the SMITH project, as well as the project "Aix-Neo-Guard", we also deal with the topic of explainable and interpretable artificial intelligence, or explainable and interpretable machine learning. Here, all aspects of data processing are taken into consideration: A clear and good visual representation of the available data to make it easier for physicians to work with the data; Frameworks that can extend existing machine learning methods with a selection of explainability and interpretability; Machine learning methods developed from scratch for explainability and interpretability; Extended evaluation methods for machine learning that go beyond individual values to show the capabilities and weaknesses of the methods. \\ \\ \\ Contact: [[en:lehrstuhl:mitarbeiter:fonck]], [[en:lehrstuhl:mitarbeiter:kruschewsky]], [[en:lehrstuhl:mitarbeiter:oprea]] \\ \\ ==== SMITH ==== ### Within the [[https://www.smith.care/|SMITH-Projektes]], innovative IT solutions are being developed to improve medical patient care. With the help of data integration centers (DIZ) and a marketplace developed within the project, the interoperable use of data and patient-oriented research is enabled. Three use cases will be used to demonstrate the added value of this data interoperability. In the first methodological use case "Phenotype pipeline" (PheP), innovative data analytic methods and tools are developed which make medical data accessible. [[en:forschung:projekte:smith|{{:forschung:projekte:smith:smith_icon.png?nolink&200 |}}]] \\ \\ With the help of two clinical use cases, the approach underlying the main objective will be demonstrated.\\ In the use case ASIC (Algorithmic Surveillance of ICU Patients), the data generated in intensive care units is continuously evaluated in order to automatically monitor the condition of patients to enable rapid therapeutic intervention. The main focus is on Acute Respiratory Distress Syndrome (ARDS) - acute respiratory failure. ARDS has a very high mortality rate, which is mainly due to the fact that the disease is often detected too late. Automated monitoring is intended to enable early diagnosis and consequently improve patient treatment. \\ The HELP clinical use case focuses on the targeted use of antibiotics to combat bacterial infections at an early stage. Innovative technologies are to be used to support infectious diseases in normal and intensive care units. \\ \\ The work on Informatik 11 takes place within the framework of the use case ASIC. On this basis, we are primarily researching data plausibility and the classification of ARDS in secondary data.\\ \\ [[en:forschung:projekte:smith|For more information, click here.]] \\ \\ \\ Contact: [[en:lehrstuhl:mitarbeiter:fonck]], [[en:lehrstuhl:mitarbeiter:kruschewsky]] ### \\ \\ ==== Clean Hands ==== ### The Clean Hands project is dedicated to achieving hand hygiene standards in a medical context. Inadequate hand disinfection repeatedly leads to hospital-acquired (nosocomial) infections. The relevant hand disinfection methods are generally based on alcohol-based disinfectant solutions. During the disinfection process, this alcohol evaporates on the user's skin surface. This evaporation removes energy from the wetted skin area in the form of heat. The evaporative cooling that occurs during a disinfection process can be measured using thermography and conclusions can be drawn from these measurements about the quality of the wetting of the hand with disinfectant. {{:forschung:saubere_heande_uv.jpg?nolink&207|}}{{:forschung:saubere_heande_symbol.jpg?nolink&200|}} \\ \\ \\ Ansprechpartner: [[:lehrstuhl:mitarbeiter:stollenwerk]] ### \\ \\ ==== SmartLungControl ==== Within the DFG project SmartLungControl, a concept for a demand-adapted control and safety monitoring of a long-term artificial lung outside the intensive care unit is being researched. In a first step, an analysis of already existing retrospective patient and animal test data is performed, followed by a systematic expert survey according to the Delphi method. In the second step, a safety concept and a draft regulation of the artificial lung will be developed. For this purpose, among other things, new sensor concepts for both reliability measures and demand-based adaptation will be investigated. Finally, the developed pilot concept will be validated and tested under different framework conditions and in defined critical operating conditions in-silico, in-vitro and in-vivo in animal experiments. \\ \\ \\ Contact: [[:en:lehrstuhl:mitarbeiter:wiartalla]] \\ \\ ==== NANNI ==== {{ :forschung:projekte:nanni:nanni_logo_final_1280.jpg?nolink&150|}} Within the framework of the BMBF-funded project Nanni (Neonatology Ventilator Device with Adaptive User Support), the partners Löwenstein Medical, the University Hospital Aachen and for RWTH Aachen University the Chair of Computer Science 11 have joined forces. In this project, the prototype of a new generation of early and neonatal ventilators is being developed. On this basis, we are primarily researching the regulation of arterial CO₂-partial pressure and the detection of problems in the artificial ventilation of newborns. \\ \\ \\ Contact: [[en:lehrstuhl:mitarbeiter:pfannschmidt]] \\ \\ ==== AutoMock ==== {{ :forschung:projekte:automocklogo.png?nolink&100|}} Within the framework of the BMBF project AutoMock, an automated mock loop is being developed for the long-term investigation and optimization of organ perfusion under a wide variety of circumstances. On this basis, the influence of perfusion parameters and pharmaceutical influence will be investigated. Furthermore, the test stand is also suitable for testing perfused medical devices. \\ \\ [[forschung:projekte:automock|For more information, click here.]] \\ \\ \\ Contact: [[en:lehrstuhl:mitarbeiter:wiartalla]] \\ \\ ===== Team ===== | **Member** | **Position / Project** | | [[en:lehrstuhl:mitarbeiter:stollenwerk]] | Group leader | | [[lehrstuhl:mitarbeiter:goldermann]] | [[en:forschung:medtech#smith|SMITH]] | | [[en:lehrstuhl:mitarbeiter:fonck]] | [[en:forschung:medtech#smith|SMITH]] | | [[en:lehrstuhl:mitarbeiter:kruschewsky]] | [[en:forschung:medtech#explainable_ai|Explainable AI]] | | [[en:lehrstuhl:mitarbeiter:oprea]] | [[en:forschung:medtech#aix-neo-guard|Aix-Neo-Guard]] | | [[en:lehrstuhl:mitarbeiter:pfannschmidt]] | [[en:forschung:medtech#nanni|NANNI]] | | [[en:lehrstuhl:mitarbeiter:wiartalla]] | [[en:forschung:medtech#smartlungcontrol|SmartLungControl]] | \\ \\ {{:forschung:p1089197_dt.jpg?nolink&575|}} \\ \\ ===== Theses ===== [[https://embedded.rwth-aachen.de/doku.php?id=en:lehre:abschlussarbeiten|The currently advertised theses can be found here.]] \\ \\ If you have a general interest in writing a thesis in the field of medical engineering and cannot commit to any of the topics mentioned, you are also welcome to send your application to the entire medical engineering group: [[medtech-abschlussarbeiten@embedded.rwth-aachen.de]] \\ \\ ===== Publications =====