Generate near-infinite permutations of complex, domain-specific, 3D environments
highlights
- Time series analysis to extract actionable insights
- Using ‘hidden’ information in your data to create additional functionalities in your device or application
- Deep learning with recurrent neural networks comprised of convolutional layers, long short-term memory cells and gated-recurrent unit
increase functionality by actionable knowledge extraction.
The term ‘actionable knowledge’ refers to the extraction of information, such as patterns or anomalies, from data that can be used to initiate follow-up actions. For example, identifying anomalous breathing patterns from sensor data (pressures, flows) in a respiratory device to create a warning system for the physician.
creating new functionalities with smart algorithms.
In many applications, the available sensor data is however very noisy due to high variability in the use environment or poor sensor quality. Advanced statistical- and machine-learning techniques enable us to extract actionable knowledge in these situations. With the application and integration/embedding of these smart algorithms in your system we can create new functionalities. For example, simultaneously detecting the respiratory effort and heartbeat signal from a single surface electromyography sensor.
starting from a concrete use-case is a success factor for applying AI.
products & services