• AI in Medical Imaging - Bay Vision - Computer Vision, DL & AI in the Bay Area

    We're excited to present two fascinating speakers for our coming meetup: Facebook's own Tullie Murrell will talk about Robust, accelerated MRI acquisition using AI, and Ehsan Adeli from Stanford will talk about Confounders in Medical Studies and Deep Learning. It will be a great evening, and we can't wait to see you all there! Agenda: 18:00-18:30 Arrival, mingling, and Pizza 18:30-18:40 Opening words by Moshe Safran, recently named CEO of RSIP Vision USA 18:40-19:10 Tullie Murrell - Applied Research Scientist at FAIR (Facebook AI Research) - Robust, accelerated MRI acquisition using AI 19:10-19:40 Ehsan Adeli - Researcher at Stanford University - Confounders in Medical Studies and Deep Learning 19:40-20:00 - Q&A and further networking Tullie Murrell - Applied Research Scientist at FAIR (Facebook AI Research) Title: Robust, accelerated MRI acquisition using AI Abstract: An MRI scan can take anywhere between 15 and 60 minutes to be acquired. With AI, we believe that this can be accelerated by up to 10x - dramatically improving patient experience and making MRIs more accessible. This inspired the fastMRI project, a collaborative project between Facebook AI Research and NYU that investigates the use of AI to improve acquisition time. In this talk, i'll introduce the MRI reconstruction problem and how it can be approached with deep learning. I'll discuss some of the recent research in the area and highlight key challenges we've faced tackling it. Notably - acquiring data, evaluating reconstruction quality and bridging the gap between research and clinical usefulness. Ehsan Adeli - Researcher at Stanford University Title: Confounders in Medical Studies and Deep Learning Abstract: Presence of confounding effects is inarguably one of the most critical challenges in all medical applications. They influence both the exposure (input, e.g., neuroimages) and the outcome (output, e.g., diagnosis or clinical score) variables and may cause spurious associations when not properly controlled for. Confounding effect removal is particularly difficult for a wide range of state-of-the-art prediction models, including deep learning methods. These methods operate directly on images and extract features in an end-to-end manner. This prohibits removing confounding effects by traditional statistical analysis, which often requires precomputed features (image measurements, like brain regional volumes). In this talk, I will talk about methods to learn confounder-invariant discriminative features and conduct confounder-aware visualization of convolutional neural networks. This meetup is is made possible thanks to sponsorship by RSIP Vision.

    1
  • AI in Medical Imaging - Bay Vision - Computer Vision, DL & AI in the Bay Area

    We're happy to meet again soon to discuss everything that's new in Computer Vision, DL & AI. This time we'll be discussing AI in Medical Imaging, with two great speakers. Agenda: 18:00-18:30 Arrival, mingling, and Pizza 18:30-19:00 Professor Daniel L. Rubin, MD - the Department of Biomedical Data Science at Stanford University - AI Approaches to Medical Imaging 19:00-19:30 Professor Irina M. Conboy - the College of Engineering at UC Berkeley - Preventing, attenuating and reversing age-imposed diseases, as a class. 19:30-20:00 - Q&A and further networking Professor Daniel L. Rubin, MD - the Department of Biomedical Data Science at Stanford University Title: AI Approaches to Medical Imaging Abstract: Despite the exciting prospects of AI for automated detection, diagnosis, and decision making with images, there are several major challenges to be addressed in order to develop robust and clinically useful AI models. First, most current AI work focuses on detection and classification, but important clinical needs and opportunities for AI lie in making clinical predictions. Second, training AI models requires tremendous amounts of labeled data, and while there are abundant images in the historical clinical archives of healthcare institutions, it is difficult to label the images in large scale, which we address through deep learning methods with clinical texts. Third, it is challenging to tap into data from multiple institutions due to privacy and intellectual property concerns. In this talk we will highlight some of the exciting frontiers in AI in medical imaging and the implications for data-driven medicine, focusing on (1) application for AI methods in making clinical prediction, (2) ways to leverage the large amounts of unlabeled data to build AI models using weak learning methods on text, and (3) federated computational methods to create AI models from multi-institutional data without data sharing. Bio: Dr. Rubin is a tenured Professor of Biomedical Data Science, Radiology, and Medicine (Biomedical Informatics Research) at Stanford University and Director of Biomedical Informatics for the Stanford Cancer Institute. His NIH-funded research program focuses on artificial intelligence in medicine and quantitative imaging, integrating imaging with clinical and molecular data, and mining these Big Data to discover imaging phenotypes that can predict disease biology, define disease subtypes, and personalize treatment. Key contributions include discovering quantitative imaging phenotypes in radiology, pathology, and ophthalmology images that identify novel clinical subtypes of disease that help to determine treatments and improve clinical outcomes. He is a Fellow of the American College of Medical Informatics, Fellow of the American Institute for Medical and Biological Engineering (AIMBE) College, and Distinguished Investigator in the Academy for Radiology & Biomedical Imaging. Dr. Rubin has published over 250 peer-reviewed scientific paper and pending patents on 10 inventions. Professor Irina M. Conboy, PhD - the College of Engineering at UC Berkeley. Title: Preventing, attenuating and reversing age-imposed diseases, as a class. Abstract: Prof. Conboy will briefly discuss the published body of work in the arena of aging and rejuvenation, including a key direction on understanding age-imposed and pathological changes cell signaling networks that regulate tissue maintenance, repair and health. In the past few years this direction has been ramified into bio-orhtogonal proteomics and development of innovative digital bio-sensors that we collaboratively applied to the fields of aging and diagnostics of genetic diseases. Success in this research will improve our understanding of the determinants of homeostatic health and will enable novel rational approaches to treat a number of pathologies that manifest in later decades of life. This meetup is is made possible thanks to sponsorship by RSIP Vision.

    7
  • AI in Medical Imaging - Bay Vision - Computer Vision, DL & AI in the Bay Area

    We're happy to meet again soon to discuss everything that's new in Computer Vision, DL & AI. This time we'll be discussing AI in Medical Imaging, with two great speakers: John Axerio-Cilies CTO and Founder, Arterys Arterys: Tearing down the Walls to bring ML to Healthcare Arterys facilitates the global advancement of medicine through data, artificial intelligence and technology. Because a significant proportion of the world's medical data resides in medical images, Arterys set out to tackle several issues around the space, including the enormous workloads radiologists face, the lack of accuracy with many of today's tools, and the need for increased consistency across practices. The company was the first to receive FDA clearance for a cloud-based product with Artificial Intelligence and it continues to focus on solving some of radiology's most pressing needs. Ron Soferman CEO, RSIP Vision Challenges in project management for Medical AI The Medical field includes some of the most complex challenges in the AI world: medical AI has to tackle the huge diversity which exists in biology and at the same time analyze the huge quantity of data available in medical imaging (and video). Artificial Intelligence is already able to provide many answers to the questions asked at the time of taking medical decisions: diagnosis, treatment, alternatives and the like. This is how medical AI solutions are invented, validated and implemented in the clinical studio, the operating room and the medical devices. Agenda: 18:00-18:30 Arrival, mingling, and Pizza 18:30-19:00 John Axerio-Cilies - CTO and Founder, Arterys - Tearing down the Walls to bring ML to Healthcare. 19:05-19:35 Ron Soferman - CEO, RSIP Vision - Challenges in project management for Medical AI 19:35-20:00 - Q&A and further networking This meetup is is made possible thanks to sponsorship by RSIP Vision.

    2
  • AI in Medical Imaging - Bay Vision - Computer Vision, DL & AI in the Bay Area

    We're meeting again soon to discuss everything that's new in Computer Vision, DL & AI. This time we'll go into AI in Medical Imaging, with three great speakers: Dr. Aïcha BenTaieb - Research Scientist at Roche "Artificial Pathology: Computerized Deep Learning Models for Cancer Diagnosis". To learn more about Dr. BenTaieb, please visit https://www.rsipvision.com/MICCAI2016-Thursday/12/ Dr. Mayank Kumar - Head of research at Gauss Surgical "Computer vision and AI for real-time monitoring of surgical blood loss" To learn more about Gauss Surgical, please visit https://www.rsipvision.com/BayVision-2018spring/4/ Dr. Michiel Schaap - Director of Imaging Science at Heartflow "The HeartFlow Analysis - Deep Learning in Clinical Practice" To learn more about Heartflow, please visit https://www.rsipvision.com/MICCAI2017-Tuesday/#15 Agenda: 18:00-18:15 Arrival and mingling 18:15-18:25 Opening words by Ralph Anzarouth, Marketing Manager at RSIP Vision (Meetup sponsors) and Editor of magazine Computer Vision News. 18:25-18:55 Dr. Aïcha BenTaieb - Research Scientist at Roche 19:00-19:30 Dr. Mayank Kumar - head of research at Gauss Surgical 19:30-20:00 Dr. Michiel Schaap - Director of Imaging Science at Heartflow This meetup is is made possible thanks to sponsorship by RSIP Vision.

    2
  • Developments in Artificial Intelligence for Autonomous Vehicles - Meetup #2

    After seeing how successful the first meeting was, we decided that the next meeting will also be about recent developments in Artificial Intelligence for Autonomous Vehicles. Agenda: 18:00-18:30 Arrival and mingling 18:30-18:40 Opening words by Shmulik Shpiro - EVP Global Business Development & Strategy at RSIP Vision, the Meetup sponsors. 18:40-19:10 Mohammad Musa, Founder and CEO at Deepen.ai 19:15-19:45 Eran Dor, Co-Founder & CEO at CRadar.Ai 19:45-20:00 Q&A Introduction will be made by Shmulik Shpiro - EVP Global Business Development & Strategy at RSIP Vision 1st speaker: Mohammad Musa - 3D Point level segmentation Mohammad Musa is Founder & CEO at Deepen.ai If you want to read more about Deepen.ai: https://www.rsipvision.com/BayVision-2018spring/8/ 2nd speaker: Eran Dor Eran is Co-Founder & CEO at CRadar.Ai, a company that owns Hardware IP that creates a Radar Signal that is at least 100x cleaner than what exists in the market today. Eran will discuss ways to utilize this signal to improve and change the way Radar sensors detect, classify and perceive the cars surroundings. This meetup is possible thanks to the sponsorship of RSIP Vision.

    5
  • Developments in Artificial Intelligence for Autonomous Vehicles

    Sunnyvale Community Center

    Our 1st meeting will include 2 expert speakers - Modar Alaoui, Founder and CEO at Eyeris, and Abhijit Thatte, AEye's new VP of Software - who will both talk about recent developments in Artificial Intelligence for Autonomous Vehicles. Agenda: 18:00-18:30 Arrival and mingling 18:30-18:40 Opening words by Ron Soferman, CEO of RSIP Vision, the Meetup sponsors. 18:40-19:10 Modar Alaoui, Founder and CEO at Eyeris 19:15-19:45 Abhijit Thatte, AEye's new VP of Software 19:45-20:00 Q&A Lecture details: 1st speaker: Modar Alaoui Title: In-Vehicle Scene Understanding AI Enabled with Automotive-grade AI Chip. Abstract: This session covers the latest advancements and advantages of inferencing a portfolio of vision AI neural networks in real-time from multiple cameras on an automotive-grade custom ASIC inside autonomous and highly automated vehicles. Enabling new real-world use cases for enhanced safety and optimized comfort, this session will further cover how this next generation of AI chips will enable efficient inference capable of generating new types of data and monetization models in this third living space. About Modar Alaoui: Modar is a technologist entrepreneur and expert in vision AI for human behavior understanding, modeling and prediction. He is currently the founder and CEO of Eyeris, the worldwide pioneer and leader of In- vehicle Scene Understanding (ISU) AI inside autonomous and Highly Automated Vehicles (HAVs). Modar graduated from Concordia University Montréal, with a concentration on Human Behavior Understanding (HBU) using artificial intelligence technologies. He combines over a decade of experience in computer vision and applied human-centric AI for a wide range of enterprise applications. His work focuses on bridging human-machine interaction with predictive invisible interfaces through body, face, object and surface understanding AI. He is a frequent speaker and keynoter on “human-centric ambient intelligence” as the next frontier in AI. Modar is a winner of several technology and innovation awards and has been featured on Bloomberg, the Wall Street Journal, Time magazine, CNBC and many major international media publications spanning over 12 different languages for his work. You can read about Eyeris at: https://www.rsipvision.com/CVPR2017-Saturday/#10 2nd speaker: Abhijit Thatte Title: Perception of 3D point clouds using deep learning Summary: This presentation will focus on a comparison of multiple deep neural network architectures for perception of 3D point clouds. It will cover architectural overview, training, validation, speed, accuracy, hardware requirements etc. It will highlight how the low level perception plays into high level scene understanding, behavior prediction and control. Abhijit Thatte is AEye's new VP of Software. You can read about AEye at: https://www.rsipvision.com/BayVision-2018spring/6/

    12