Computer Vision News - April 2020

3 Summary Spotlight News 31 TU Wien researchers develop Ultrafast machine vision with 2D material neural network image sensors Visual information is usually captured by a frame-based camera, converted into a digital format and processed afterwards using ML neural network . However, the large amount of results in low frame rates and high power consumption, calling for techniques to increase the efficiency of the subsequent signal processing. In this work the authors demonstrate that an image sensor can itself constitute an artificial neural network that can simultaneously sense and process optical images without FDA clears Fluidda's digital lung imaging platform for respiratory disease monitoring The fair magazine FierceBiotech tells us about Fluidda’s Broncholab software , a new digital imaging platform designed to give doctors a clear peek into a person’s lung function . For instance, the platform can give the pulmonologist information about how drug particles deposit throughout the lung when inhaled or indicate any trapped air pockets as well as volume and mapping of ventilation and nodules. You can find more articles about AI in pulmonology here . Read More Software is enabling medical 3D printing innovation The three main actors in the imaging space - Royal Philips, GE Healthcare and Siemens Healthineers - have made progress in creating 3D-printable files from medical scans. Software needs to be very advanced in order to produce patient-specific 3D-printed models for customized orthopedic implants and other medical devices . Uses of AI software-powered 3D printers to produce physical orthopedic devices is a major step forward from creating a 3D version of colon, heart and lungs . Read More latency. Once the training process is completed, the neural network can work alone at the level of the sensor, with no need for a computer. Read More