Bay Labs: Redefining Cardiovascular Imaging Through Deep Learning

Bay Labs is a privately held medical technology company at the forefront of applying artificial intelligence (AI) to cardiovascular imaging. Bay Labs is focused on increasing quality, value and access to medical imaging by combining deep learning and ultrasound. The company is developing products designed to help medical professionals of all skill levels perform and interpret high-quality echocardiography to ultimately benefit their patients.

Bay Labs was founded in San Francisco to make a major difference in healthcare by democratizing medical imaging. Initially, the founders were focused on reimagining the detection, monitoring and timing of treatment for cardiovascular disease with AI. According to the Centers for Disease Control, over 600,000 people in the US die from cardiovascular disease each year making it the cause of one in four deaths.

Bay Labs’ first product addresses the limitations in scaling echocardiography (ultrasound of the heart), the diagnostic gateway to cardiovascular treatment. Today, there are over 20 million echoes performed annually in the US. Yet, the skills needed for a sonographer or physician to become really good at acquiring quality ultrasound images of the heart, or interpreting them, require very long and intense training. The company is working to design solutions aimed at helping medical professionals accurately perform ultrasound exams by providing users real-time guidance for high-quality image capture, automatic clip selection to identify the best images obtained during the exam, and auto-interpretation of the images. Ultimately, these tools may help save time and improve reproducibility.

Bay Labs’ vision is to transform how diagnostic cardiovascular care is delivered in healthcare systems and in rural settings, enabling expanded access to high-quality echocardiography, more appropriate treatment from improved monitoring and detection, which may improve patient outcomes, while allowing for a reduction in cost. To achieve this vision, the company has assembled a team of experts in machine learning, visual neuroscience, physics, medical devices, regulatory affairs, and cardiology.

 

The Dynamic Leaders and Innovators

Charles Cadieu, Co-founder and CEO of Bay Labs, is both a neuroscientist and entrepreneur who has sought to push forward the company’s understanding of information processing in the brain. Charles currently leads Bay Labs on the mission to make medical imaging universally accessible to improve wellness and healthcare. Throughout his career, Charles has blended entrepreneurial and scientific pursuits to bring cutting-edge AI technologies to market and to increase the knowledge of ourselves. As an entrepreneur, he has started multiple companies and was a founding team member of IQ Engines, which was acquired by Yahoo! and now powers deep learning-based visual analytics across all Yahoo! properties. In his scientific work, he and his colleagues were the first to show that deep learning algorithms could rival the human visual system on specific tasks. Charles holds a BS and MEng from MIT in Electrical Engineering and Computer Science and a Ph.D. from UC Berkeley in Neuroscience.

Kilian Koepsell, Co-founder and CTO, leads the development team at Bay Labs. Kilian was a founding member of IQ Engines and the 2nd engineer at Fivestars Loyalty. Kilian has more than ten years of experience in artificial neural networks and visual neuroscience as a research scientist at the former Redwood Neuroscience Institute and at UC Berkeley. He holds a Ph.D. in theoretical physics from Hamburg University.

 

Transforming Heart Healthcare with Futuristic Solutions

Ultrasound has a unique potential to detect and aid in the management of cardiovascular disease because devices are portable, effective, and non-ionizing.  However, the process of performing the ultrasound exam and interpreting the results requires extremely skilled experts: sonographers and cardiologists.  Bay Labs is working to create an ultrasound solution that aids in image capture by providing deep learning-based real-time guidance, and that aids in interpretation by selecting the best images and providing deep learning-based measurements on them. The company’s software could enhance the current process of performing an ultrasound exam by enabling healthcare professionals like medical assistants, nurses and physician assistants in image capture, and by enabling cardiologists to expedite their reading of the exam.

Bay Labs-powered automated ultrasound solution, once fully developed, has the potential to transform how diagnostic care can be delivered in healthcare systems and in rural settings, enabling reduction in cost, more appropriate treatment from improved monitoring and detection, which may lead to improved patient outcomes.

The company currently has several clinical studies planned to test the solution in various clinical settings. A pilot to evaluate nurse practitioner usage began at Stanford in June 2018 and a second pilot will be announced shortly. Bay Labs has also demonstrated that with its technology, school teachers in Kenya can detect rheumatic heart disease, potentially changing the treatment paradigm for this disease. The company’s aim is to provide evidence that nurses and other medical professionals with limited training can acquire imagery that can support clinical decision-making and that our automated interpretations reduce variability and perform well on a wide range of patients.

 

Accelerating Innovation Through Networks and Partnerships

At Bay Labs, the detection, monitoring and timing of treatment for cardiovascular disease with artificial intelligence are reimagined. Its solutions are being developed to extend the use of ultrasound and enable disease detection and monitoring at a new scale. The company works with a trusted network of world-class clinical and academic partners including New York Presbyterian Hospital, Allina Health, Minneapolis Heart Institute and Northwestern Medicine, among others. Recently, Northwestern Medicine announced that Bay Labs was one of the four companies it is partnering with to utilize a $25 million grant to develop AI as a tool in the fight against heart disease.

 

Awards and Accolades Honouring Bay Labs’ AI Initiatives

Bay Labs is honored to be recognized in the industry, customers, organizations and publications around the world for its cutting-edge innovation. Below are some of them:

•  Bay Labs’ EchoMD AutoEF software received FDA clearance for fully automated AI echocardiogram analysis. This deep learning algorithm package automates both clip selection and ejection fraction calculation, and the company received the 510(k) from the FDA in June 2018. EF is the single most widely used metric of cardiac function and used as the basis for many clinical decisions.

•  In 2018, Business Insider Intelligence report included Bay Labs as one of the top five start-ups disrupting healthcare in the artificial intelligence space.

•  In 2017, Bay Labs was recognized as one of the most promising AI Startups in the social impact category at NVIDIA’s Inception competition.

•  Bay Labs is the proud recipient of a National Science Foundation grant to support the development of ultrasound guidance.

•  The company is now recognized by Analytics Insights Magazine as one of “The 10 Most Innovative Deep Learning Solutions Providers 2018”.

 

The Optimistic Future Driven by AI

At Bay Labs, the belief goes that AI applied to echocardiography will enable identification of heart disease at an earlier stage by making it possible for many more medical professionals to obtain fast, high-quality echocardiograms, and thereby allowing for early detection and better appropriate management of that condition. It also can help bring echocardiography to areas of the world where it’s not currently available but could greatly use it.