Artificial Intelligence in coming to Medicine in a big wave. From making diagnosis in various medical conditions, suggesting the most appropriate treatment, finding of the latest advancement in the literature, to predicting the prognosis and outcome of the disease, AI is offering unprecedented opportunities to improve care of our patients. Taking digestive tract cancer as an example, AI-assisted image analysis aids the detection of colorectal neoplasia during colonoscopy, provide so-called optical biopsy of the lesions, integrate genomic, epigenetic and metagenomics data to provide new classification and sub-classification of the cancers, and provide evidence-based suggestion what is the optimal therapy of the condition. Furthermore, AI-assisted surgery, through semi-automated and (in future) fully-automated robotic surgery, will take over at least some part of the surgical treatment of such malignant lesions. This is a defining moment in Medicine. What is the future role of doctors and nurses? How should we train medical students and pharmacists and physiotherapists in the future? And, importantly, when things go wrong, such as reaching a wrong diagnosis and/or mishap occurring in the treatment of a patient, who should take the responsibility? This lecture will offer a peek into the future.

Computer vision has seen major success in learning to recognize objects from massive “disembodied” Web photo collections labeled by human annotators. Yet cognitive science tells us that perception develops in the context of acting the world---and without intensive supervision. Meanwhile, many realistic vision tasks require not only categorizing a well-composed human-taken photo, but also actively deciding where to look in the first place. In the context of these challenges, we are exploring how machine perception benefits from anticipating the sights and sounds an agent will experience as a function of its own actions. Based on this premise, we introduce methods for learning to look around intelligently in novel environments, learning from video how to interact with objects, and perceiving audio-visual streams for both semantic and spatial context. Together, these are steps towards first-person perception, where interaction with the world is itself a supervisory signal.

The ImPACT Tough Robotics Challenge is a national project of Japan Cabinet Office (period: 2014-18, researchers: 62 PIs and 300 researchers). It focused on tough technologies of robotics to give solutions to disaster response, recovery and preparedness, and achieved the following results.

• Cyber Rescue Canine, a digitally empowered rescue dog, wearing a suit for monitoring its behavior (motion, map, image and action) and conditions (health and enthusiasm).
• Active Scope Camera, a serpentine robot for search in debris that crawls and levitates in few-cm gaps, with visual, auditory and haptic sense for navigation and victim search. Dragon Firefighter, a serpentine robot flying into buildings by water jets to extinguish origins of fire.
• Serpentine robots for plant inspection with high mobility in ducts, at pipes, on uneven terrain, on vertical ladders, and over high steps. Omni gripper that can grasp wide variety of targets including sharp edges without positioning.
• An UAV that is robust in strong wind 20 m/s, in rain 300 mm/h, at change of payload, and at stop of propellers, with hearing and localizing voice coming from ground during flight.
• A 4-legged robot for plant inspection in risky places, with high mobility on rubbles, on vertical ladders, and over high steps. A 30-cm robot hand that can keep grip of 50-kg loads without electricity.
• A construction robot of double-swing dual-arm mechanism with both high-power and preciseness, with force and touch bilateral feedback without sensor at the end-effector, and teleoperation support by real and virtual bird’s-eye-view images.

Our scientific understanding of haptic interaction is still evolving, both because what you feel greatly depends on how you move, and because engineered sensors, actuators, and algorithms typically struggle to match human capabilities.  Consequently, few computer and machine interfaces provide the human operator with high-fidelity touch feedback or carefully analyze the physical signals generated during haptic interactions, limiting their usability.  The crucial role of the sense of touch is also deeply appreciated by researchers working to create autonomous robots that can competently manipulate everyday objects and safely interact with humans in unstructured environments.  My team works in all of these related areas, aiming to sharpen our understanding of haptic interaction while simultaneously inventing helpful human-computer, human-machine, and human-robot systems that take advantage of the unique capabilities of the sense of touch.  This talk will showcase key examples from our ongoing research, including Haptipedia (www.haptipedia.org), social-physical human-robot interaction for exercise, and large fabric-based tactile sensors that employ electrical resistance tomography.  I will close by sharing suggestions on recruiting and leading a diverse team of researchers.