A system for real-time multivariate feature combination of endoscopic mitral valve simulator training data.
Reinhard FuchsKarel M Van PraetRichard BieckJörg KempfertDavid HolzheyMarkus KoflerMichael Andrew BorgerStephan JacobsVolkmar FalkThomas NeumuthPublished in: International journal of computer assisted radiology and surgery (2022)
The results support the validity of our setup and feature calculation, while their analysis shows significant distinctions and can be used to identify the used endoscopic view mode, something not apparent when analyzing time tables of each exercise attempt. The presented work is therefore a first step toward future developments, with which multivariate feature vectors can be classified automatically in real-time to evaluate endoscopic training and track learning progress.
Keyphrases
- ultrasound guided
- mitral valve
- machine learning
- deep learning
- virtual reality
- data analysis
- high intensity
- physical activity
- endoscopic submucosal dissection
- electronic health record
- heart failure
- artificial intelligence
- left atrial
- magnetic resonance imaging
- left ventricular
- computed tomography
- diffusion weighted imaging