Listen to this articleBelow is our recent interview with Moshe Safran from RSIP Vision:
Q: Could you provide our readers with a brief introduction to RSIP Vision?
A: With pleasure. Our company mission is to drive technological innovation in the medical imaging space through advanced AI and computer vision software. We have been in this space for over 25 years. Our modules are integrated in industry leading medical devices, and have supported treatment of thousands of patients undergoing procedures and surgeries for use cases in multiple medical disciplines. We also provide deep research and customized algorithm development capabilities to medical device companies to give them an edge over their competition. Our multidisciplinary team includes over 50 algorithm experts, MD’s and medical image annotation specialists, who work together to develop practical AI modules that ensure precision, reduce time to market, cut costs and allow core R&D teams to focus on key initiatives. Our advanced technologies can be leveraged by a range of medical devices within leading facilities worldwide, ensuring our customers remain at the forefront of the latest medical advances.
Recommended: OlaClick Provides An Alternative For Restaurants Who Want To Grow Their Own Sales Channel Without Paying Commissions
Q: Can you give us more insights into Artificial Intelligence and deep learning for medical devices and applications?
A: Medical imaging is used to empower physicians with precise visualization before and during complex procedures and surgeries. It helps guide and inform the actions and decisions of physicians to ensure they’re working with as much precision as possible. This is where artificial intelligence comes into play. By powering our solutions with AI and advanced learning algorithms, we’re able to enhance how surgeons can use these medical images. Whether by providing a 3D rendering of a patient’s anatomy or by directing what procedure will best fit the patient’s needs, AI helps doctors to navigate and support decisions based on patient imaging. AI can also provide additional insight, safety warnings and proactive high-risk indicators to the medical team(s) based on information it has learned from previous surgeries.
There are endless uses for medical imaging technology today. The rule of thumb regarding deep learning advancement is: if a human can classify or measure an image, and if the proper dataset is available, they can train AI technologies to do the same at a similar or sometimes even superior level. This saves time, money and unnecessary errors from being made. Having a full-time employee segment or classify images all day can get quite costly, and it’s not uncommon for a misread to occur. AI can efficiently and effectively process the images at hand with peak accuracy, thereby allowing doctors to focus their time on treating and supporting patients.
Q: How do you see application of computer vision in medical imaging?
A: There’s a distinction to be made between the concepts of deep learning and computer vision. Computer vision refers to any computer program or automated algorithm that analyzes an image and extracts information. State of the art AI has the ability to build upon the foundational programming entered by an engineer, thereby expanding these abilities by learning to solve complicated problems from examples in an automated fashion. However, if you don’t have proper engineering and medical personnel in place to implement this technology, you can’t teach AI how to do something. There are plenty of AI companies that provide platforms capable of processing canned algorithms by default, for example for detecting defects in an industrial process. However, for the medical field, this isn’t sufficient. Specific medical domain expertise and multiple algorithmic methods must be carefully applied to achieve an effective solution.
Q: Can you give us a few examples of current and future use cases?
A: RSIP Vision focuses on use cases for interventions and surgeries, and enhances images to guide precise medical intervention. We’ve recently announced a coronary artery-focused, AI-based software module that receives a contrast CT scan as input and uses it to generate a 3D model of the coronary arteries. It also provides customizable measurements, such as vessel dimensions and stenosis detection. The model can typically be integrated into our customers’ existing software pipeline, but it can also work as a standalone solution. The technology under the hood is AI-based and vendor-neutral, so it allows processing studies with no limitation of the source.
We’ve also done extensive work within ultrasounds for cardiac analysis. The new software module we developed is used to help with quick evaluation and diagnosis, enabling medical teams at various point-of-care settings to perform a highly-accurate heart evaluation quickly and receive an immediate, onsite diagnostic from the Parasternal Long Axis (PLAX) view. This PLAX view, an essential part of the point-of-care ultrasound (POCUS) protocol, enables the measurement of the left ventricle of the heart and reviews the function of the heart as a whole. This new vendor-neutral technology optimizes existing medical devices to save more lives and improve day-to-day clinical care.
Lasty, our AI-based segmentation and measurement tool is used for detecting objects of interest and their boundaries quickly and automatically, making surgical and diagnostic measurements easier and more accurate to improve/enhance treatment decisions. The domain-agnostic algorithmic module requires minimal work on the part of the user to deliver an accurate 3D visualization and analysis of patient anatomy. The solution runs automatically, is applicable across medical imaging verticals and modalities, and helps avoid human error factors such as fatigue and misreads (which may result in mistakes in measurement).
Recommended: Fantasy Football Startup Sorare Raises Over $40 Million In Series A Funding, Announces A New Game With Ubisoft
Q: What can we expect from RSIP Vision in the next 6 months? What are your plans?
A: We have multiple initiatives running in additional medical imaging AI domains – specifically in orthopedics, advanced endoscopy, urology and echocardiology. We plan to develop more in-house capabilities of novel applications of these technologies. We’ll also continue to work with medical advisers, including cardiologists and radiologists, to bridge the gap between understanding the needs of the medical field and what capabilities are already created. As a company, we’re set up for success by understanding technology and having a medical team who has the ability to understand the connection to medical device companies, which we will continue to expand on in the future.
