00 Informatik, Wissen, Systeme
Filtern
Dokumenttyp
Sprache
- Englisch (18)
Volltext vorhanden
- ja (18)
Gehört zur Bibliographie
- nein (18)
Schlagworte
- Maschinelles Lernen (5)
- machine learning (4)
- Convolutional Neural Network (3)
- Künstliche Intelligenz (3)
- Alkoholmissbrauch (2)
- Anomalie <Medizin> (2)
- Anomalieerkennung (2)
- App <Programm> (2)
- Intervention <Medizin> (2)
- Jugend (2)
- Nachhaltigkeit (2)
- Software (2)
- alcohol (2)
- anomaly detection (2)
- internet (2)
- young people (2)
- "Visual Knowledge Communication" (research project) (1)
- Agent <Künstliche Intelligenz> (1)
- Algorithmus (1)
- Autofahren (1)
- Autonomer Roboter (1)
- Bild (1)
- Bilderkennung (1)
- Bildverarbeitung (1)
- Branch-and-Bound-Methode (1)
- CNN (1)
- Computeranimation (1)
- DNN (1)
- Deep learning (1)
- Demontage (1)
- Diskriminanzanalyse (1)
- Docker (1)
- EHR (1)
- Elektronische Patientenakte (1)
- Energieverbrauch (1)
- Engagement (1)
- Fahrerassistenzsystem (1)
- Forschungsprojekt (1)
- Fuzzy-Logik (1)
- GA-ANFIS (1)
- GA-FUZZY (1)
- GA-GA-FUZZY (1)
- Gehirn-Computer-Schnittstelle (1)
- Genetischer Algorithmus (1)
- Gesundheitsökonomie (1)
- Green-IT (1)
- Handy (1)
- ICU (1)
- Industrieroboter (1)
- Informatik (1)
- Intensivpflege (1)
- Intensivstation (1)
- Interdisziplinäre Forschung (1)
- Internet (1)
- Internet der Dinge (1)
- Internet of Things (1)
- Kehlkopf (1)
- Klassifikation (1)
- Konturfindung (1)
- Künstliche Beatmung (1)
- Laborparameter (1)
- Medizinische Informatik (1)
- NIR-Spektroskopie (1)
- Neuro-Fuzzy-System (1)
- Neuronales Netz (1)
- Neurowissenschaften (1)
- Nichtlineare Diffusion (1)
- Online-Algorithmus (1)
- Peer Review (1)
- Programmierumgebung (1)
- Prozessmanagement (1)
- Psychologie (1)
- Rapid Prototyping (1)
- Raucherentwöhnung (1)
- Rekursives neuronales Netz (1)
- Robotertechnik (1)
- Selbstorganisierende Karte (1)
- Sepsis (1)
- Soziale Norm (1)
- Statistik (1)
- Stimmband (1)
- Student (1)
- Suchverfahren (1)
- Taxonomie (1)
- Teer (1)
- Unterstützungssystem <Informatik> (1)
- VentAI (1)
- Verteiltes System (1)
- Visuelle Kommunikation (1)
- Wissensvermittlung (1)
- Zigarettenrauch (1)
- adaptive neuro fuzzy system (1)
- animations (1)
- artificial intelligence (1)
- brain-computer interface (1)
- branch and bound search algorithm (1)
- classification (1)
- cluster-management (1)
- containerization (1)
- convolutional neural network (1)
- convolutional neural networks (1)
- critical care (1)
- critically-ill patient (1)
- data-oriented business process (1)
- deep learning (1)
- disassembly plan (1)
- disassembly process (1)
- distributed computing (1)
- driver-assisting system (1)
- driving performance (1)
- edge detection (1)
- energy awareness (1)
- energy-aware software (1)
- engagement taxonomy (1)
- environmental criteria for software (1)
- fingerprint recognition (1)
- fuzzy logic (1)
- genetic algorithm (1)
- green software (1)
- human-robot-collaboration (1)
- image processing (1)
- image recognition (1)
- incremental learning (1)
- indoor localization (1)
- inferencing (1)
- informed software agent (1)
- intelligent robot assistant (1)
- intensive care (1)
- internet of things (1)
- knowledge-intensive process (1)
- lab values (1)
- laryngeal high-speed video (1)
- learning vector quantization (1)
- lightGBM (1)
- linear discriminant analysis (LDA) (1)
- long short-term memory (1)
- maximal function (1)
- medical consumables (1)
- medical economics (1)
- medical informatics (1)
- mobile phone (1)
- model of software impacts (1)
- near-infrared spectroscopy (1)
- nonlinear diffusion (1)
- online learning (1)
- ontology (1)
- peer review process (1)
- problem drinking (1)
- product model (1)
- rapid prototyping (1)
- reinforcement learning algorithm (1)
- resource efficiency (1)
- robot system (1)
- scalability (1)
- self organizing maps (SOM) (1)
- sharp function (1)
- signal processing (1)
- smoking cessation (1)
- social norms (1)
- software energy measurements (1)
- students (1)
- sustainability criteria (1)
- sustainability indicators (1)
- sustainability management (1)
- sustainable software (1)
- tar (1)
- text messaging (1)
- time series classification (1)
- tobacco (1)
- ventilation (1)
- visual programming environments (1)
- vocal fold vibration (1)
- web (1)
- web servers (1)
The aim of this work was to develop and evaluate the reinforcement learning algorithm VentAI, which is able to suggest a dynamically optimized mechanical ventilation regime for critically-ill patients. We built, validated and tested its performance on 11,943 events of volume-controlled mechanical ventilation derived from 61,532 distinct ICU admissions and tested it on an independent, secondary dataset (200,859 ICU stays; 25,086 mechanical ventilation events). A patient “data fingerprint” of 44 features was extracted as multidimensional time series in 4-hour time steps. We used a Markov decision process, including a reward system and a Q-learning approach, to find the optimized settings for positive end-expiratory pressure (PEEP), fraction of inspired oxygen (FiO2) and ideal body weight-adjusted tidal volume (Vt). The observed outcome was in-hospital or 90-day mortality. VentAI reached a significantly increased estimated performance return of 83.3 (primary dataset) and 84.1 (secondary dataset) compared to physicians’ standard clinical care (51.1). The number of recommended action changes per mechanically ventilated patient constantly exceeded those of the clinicians. VentAI chose 202.9% more frequently ventilation regimes with lower Vt (5–7.5 mL/kg), but 50.8% less for regimes with higher Vt (7.5–10 mL/kg). VentAI recommended 29.3% more frequently PEEP levels of 5–7 cm H2O and 53.6% more frequently PEEP levels of 7–9 cmH2O. VentAI avoided high (>55%) FiO2 values (59.8% decrease), while preferring the range of 50–55% (140.3% increase). In conclusion, VentAI provides reproducible high performance by dynamically choosing an optimized, individualized ventilation strategy and thus might be of benefit for critically ill patients.