This video stockshots presents a series of EU-funded projects on Artificial Intelligence in some of these sectors: Internet of food and farm 2020 (Lithuania), MURAB (The Netherlands), Human brain project (Germany), Symbio-Tic (Sweden), Cyberlegs Plus Plus (Belgium), De-Enigma (The Netherlands), Vi-Das (Spain - The Netherlands).

computer

Belgium

road traffic

Horizon 2020 (T)

Germany

technology

Sweden

computer applications

medical devices (T)

Netherlands

medical device

robotisation

research

use of robotics and artificial intelligence (T)

scientific research

medical research

electronic device

farming sector

use of drones (T)

European Fund for Strategic Investments (T)

Lithuania

scientific profession

agricultural performance

European Investment Fund

mountain

car safety (T)

fight against cancer (T)

Infoclips (T)

Infoclip: Artificial Intelligence, EU-funded projects

Title: Infoclip: Artificial Intelligence, EU-funded projects

1. INTERNET OF FOOD AND FARM 2020, LITHUANIA

The drone operator takes the drone out of the car (2 shots)

Demonstration of one use case from the IoF2020 (Internet of Food &amp; Farm 2020) project, Kaunas and Vilnius, Lithuania.The IoF2020 project was funded by &euro;30 million from the EU&rsquo;s Horizon 2020 programme. It aims to accelerate adoption of IoT for securing sufficient, safe and healthy food and to strengthen competitiveness of farming and food chains in Europe. The use case in Lithuania uses Artificial Intelligence technologies to perform complex analyses of crop field hyperspectral images taken by a drone system and data collected in a laboratory to identify the composition of nutrients in crops. Thanks to these analyses they can determine which nutritional elements and how much of them a plant is lacking at different stages of its growth. It reduces the use of fertiliser, reduce the cost, yield and quality is increased.

The drone operator installs the hyperspectral camera on the drone

The drone operator plans the route taken by the drone over the field (2 shots)

The drone operator drives the drone (2 shots)

Drone views of the drone flying over the field while collecting hyperspectral imaging (4 shots)

On the same field where the hyperspectral images have been taken, the drone operator takes some samples of the crops (3 shots)

In the laboratory where they analyze the composition of the crops, the lab technician puts the samples in a dryer (4 shots)

The lab technician turns the dried crops into powder (3 shots)

In another part of the laboratory they put the powder of the samples in test tubes (3 shots)

The lab technician puts the test tubes in a distillation unit that measures the amount of nitrogen and protein in the crop (5 shots)

The lab technician puts other test tubes in an inductively coupled plasma atomic emission spectrometer that measures the exact composition of the crops (3 shots)

On a computer, using Artificial Intelligence complexe analysis of crop field hyperspectral images takes place. Thanks to the information collected in the laboratory put together with the hyperspectral images the system learns to recognize various indicators and identifies the composition of nutrients in corps (4 shots)

The final result is a map that will help farmers to identify which part of the field has a lack of some element, nitrogen or protein. In the map the red part means that this part of the field has a lack of Manganese and Calcium.

The radiologist looks at a breast MRI scan (5 shots)

Demonstration of the MURAB (MRI and Ultrasound Robotic Assisted Biopsy) robot on a silicone breast, Nijmegen, The Netherlands. The MURAB project was funded by &euro;4 million from the EU&rsquo;s Horizon 2020 programme. It aims to revolutionise the way cancer screening and muscle diseases are researched for patients and has the potential to save lives by early detection and treatment. Using Artificial Intelligence technology, the project developed a robot that make more accurate biopsies to diagnose cancer faster This will be achieved by identifying a target using Magnetic Resonance Imaging (MRI) and then use a robot with an ultrasound probe to match the images and navigate to the right location.

The MURAB robot scans the silicone breast thanks to the green markers put by the radiologist so the robot will know where it will have to execute the ultrasound scan and biopsy (4 shots)

Thanks to the information collected during the scan, the radiologist plans the path the robot will take to perform the MRI-ultrasound scan (3 shots)

The robot proceeds the ultrasound scan of the silicone breast (4 shots)

The robot combines the information collected from the MRI and from the ultrasound scans to identify and localize the target lesion inside the silicone breast (blue ball in the 3D imaging) taking care of tissue deformations

The robot equipped with an intelligent ultrasound probe aligns the biopsy needle with the lesion and the radiologist inserts the needle manually while the robotic arm keeps it aligned with the lesion inside the silicone breast. The MURAB robot reduces the number of hospital visits for the patient, biopsy is more accurate and it reduces the diagnostic time (3 shots)

Scientist cuts thin slices of the human brain preserved in paraffin wax with a microtome (3 shots)

The Human Brain Project is a FET Flagship launched in 2013 and funded by &euro;54 million under the EU&rsquo;s 7th Framework Programme. In 2016 the Human Brain Project Framework Partnership Agreement was funded by &euro;177 million under the EU&rsquo;s Horizon 2020 programme. In 2018 the ICEI (Interactive Computing E-Infrastructure) for the Human Brain Project was funded by &euro;25 million under the EU&rsquo;s Horizon 2020. The Human Brain Project aims to empower brain research toward understanding the human brain and its diseases to advance brain medicine and computing technology. At the Forschungszentrum J&uuml;lich, Germany, scientists are using Artificial Intelligence to create a 3D model of the human brain. To achieve this the Artificial Intelligence analyse thousands of ultrathin histological brain slices scanned into a computer and subsequently reconstruct them in 3D thanks to a supercomputing centre that process all the information. In the long term, the data obtained will improve understanding of how the brain works and enable earlier diagnosis and better treatment of psychiatric and neurological disorders. Researchers are also using these results to build brain-inspired computer.

Brain slices are put in boxes and then stored on shelves (3 shots)

Brain slices are placed on a glass plate (2 shots)

Glass plates are moved to another room and then placed into frames (4 shots)

Glass plates are put in a scanner and scanned into the computer (3 shots)

The scientist programs the scanner and analyses the scans (5 shots)

All these scanned brain slices (approximately 7400) are put together by the artificial intelligence to create a 3D-brain Atlas (4 shots)

Supercomputing center - supercomputer are used to analyse all these datas (1 trillion bytes) to reconstruct a 3D volume of the brain (5 shots)

The engineer pre-programs the path the robot will take (3 shots)

Demonstration of the SYMBIO-TIC (Symbiotic Human-Robot Collaborative Assembly: Technologies, Innovations and Competitiveness) project on an industrial robot, Sk&ouml;vde, Sweden. The SYMBIO-TIC project was funded by &euro;6 million from the EU&rsquo;s Horizon 2020 programme. It aims to create a safe, dynamic, intuitive and cost-effective industrial working environment through an immersive and symbiotic collaboration between human workers and robots. Thanks to kinect sensors the robot can create a 3D environment where it can detect the worker and obstacles to avoid any collision. Artificial Intelligence is used to process the information collected by the sensors, identify the worker, make a classification and adapt its reactions to the circumstances. The benefits that the project can bring about include lower costs, increased safety, better working conditions and higher profitability through improved adaptability, flexibility, performance and seamless integration

The human worker and the robot are working together to assemble a mass balancing system used in car engine (8 shots)

Demonstration of the collision avoidance - thanks to kinect sensors the robot can create a 3D environment where it can detect the worker and every obstacle to avoid any collision. AI is used to process the information collected by the sensors, identify the worker, make a classification and adapt its reactions (7 shots)

A man with an amputated leg arrives in the BruBotics office of the Vrije Universiteit Brussel (VUB)

Demonstration of the CYBERLEGS Plus Plus (The CYBERnetic LowEr-Limb CoGnitive Ortho-prosthesis Plus Plus) prosthesis, Brussels, Belgium. The CYBERLEGS Plus Plus was funded by &euro;4 million from the EU&rsquo;s Horizon 2020 programme. The global goal of the CYBERLEGs Plus Plus project is to validate the technical and economic viability of the CYBERLEGS project funded by &euro;2,5 million from the EU&rsquo;s 7th Framework Programme. It develops wearable exoskeletons and prosthesis to enhance/restore the mobility of transfemoral amputees and to enable them to perform locomotion tasks such as ground-level walking, walking up and down slopes, climbing/descending stairs, standing up, sitting down and turning in scenarios of real life. It collects data thanks to embedded sensors and sensing devices displayed on the body. These informations are analysed by algorithm to classify the locomotion mode of the upcoming step before the heel strike occurs, in order to ensure a safe interaction with the ground.

The researcher helps the amputee to take off his usual prosthesis (2 shots)

The researcher removes the bottom part of the prosthesis and puts the Cyberlegs prosthesis instead (8 shots)

Wireless sensors which are inertial measurement units are put on the healthy leg and on the abdomen to detect the movements of the body (5 shots)

The prosthesis is switched on and programmed in the software (4 shots)

The researcher helps the amputee to go to the treadmill

The amputee walks with the Cyberlegs prosthesis on the treadmill (6 shots)

The sensors put on the body and the sensors embedded in the prosthesis give informations about the position and the movements of the body. Artificial intelligence helps the data being analyzed by algorithms to detect what the person is doing and what he is intending to do to make sure the prosthesis is actuated in the right way (3 shots)

6. DE-ENIGMA, THE NETHERLANDS ; Disclaimer: a non-autistic child was used as a model in this video

The person selects on a tablet the emotion to teach to the child. The child must then select on another tablet the part of the face which shows the emotion and then has to reproduce the emotion showed by the Zeno robot (7 shots). DISCLAIMER: a non-autistic child was used as a model in this video.

Demonstration of the DE-ENIGMA (Multi-Modal Human-Robot Interaction for Teaching and Expanding Social Imagination in Autistic Children) project on an humanoid robot, Twente, Netherlands. The DE-ENIGMA project was funded by &euro;4 million from the EU&rsquo;s Horizon 2020 programme. It aims to develop an humanoid robot called Zeno to interact with autistic children and teach them emotions. During the project, Zeno has gone through several design phases, getting &lsquo;smarter&rsquo; every time thanks to Artificial Intelligence. This technology enables to realise robust, context-sensitive (such as user- and culture-specific), multimodal (including facial, bodily, vocal and verbal cues) and naturalistic human-robot interaction (HRI) aimed at enhancing the social imagination skills of children with autism. DISCLAIMER: a non-autistic child was used as a model in this video.

Zeno is detecting the facial expressions of the child, practices social interaction and therefore being more socially intelligent (4 shots) DISCLAIMER: a non-autistic child was used as a model in this video

Drone views of the car used in the VI-DAS project (3 shots)

Demonstration of the VI-DAS (Vision Inspired Driver Assistance Systems) project, San Sebastian, Spain and Eindhoven, Netherlands. The VI-DAS project was funded by &euro;6 million from the EU&rsquo;s Horizon 2020 programme. It aims to design next generation of camera based driver assistance/driving automation capable to handle critical situations like driving control handover and handback by providing scene understanding capabilities. It uses Artificial Intelligence to analyse data collected by sensors inside and outside the car to better understand driver moves and develop the Advanced driver-assistance systems (ADAS) and improve safety and efficiency.

A driver executes different critical driving situations. Thanks to sensors inside the car, the AI analyses all the data to better understand driver, vehicle and scene contact in order to develop the Advanced driver-assistance systems (ADAS) and improve safety and efficiency (8 shots)