| NSF Org: |
TI Translational Impacts |
| Recipient: |
|
| Initial Amendment Date: | April 13, 2016 |
| Latest Amendment Date: | September 21, 2018 |
| Award Number: | 1556103 |
| Award Instrument: | Standard Grant |
| Program Manager: |
Peter Atherton
patherto@nsf.gov (703)292-8772 TI Translational Impacts TIP Directorate for Technology, Innovation, and Partnerships |
| Start Date: | April 15, 2016 |
| End Date: | March 31, 2020 (Estimated) |
| Total Intended Award Amount: | $740,788.00 |
| Total Awarded Amount to Date: | $1,376,062.00 |
| Funds Obligated to Date: |
FY 2018 = $635,274.00 |
| History of Investigator: |
|
| Recipient Sponsored Research Office: |
2000 SIERRA POINT PKWY STE 800 BRISBANE CA US 94005-1889 (415)424-5616 |
| Sponsor Congressional District: |
|
| Primary Place of Performance: |
490 Post St. Suite 824 San Francisco CA US 94102-1409 |
| Primary Place of
Performance Congressional District: |
|
| Unique Entity Identifier (UEI): |
|
| Parent UEI: |
|
| NSF Program(s): | SBIR Phase II |
| Primary Program Source: |
01001819DB NSF RESEARCH & RELATED ACTIVIT |
| Program Reference Code(s): |
|
| Program Element Code(s): |
|
| Award Agency Code: | 4900 |
| Fund Agency Code: | 4900 |
| Assistance Listing Number(s): | 47.084 |
ABSTRACT
The broader impact/commercial potential of this Small Business Innovation Research (SBIR) Phase II project will be in the field of healthcare. The United States spends approximately $9,000 per person per year on healthcare. Ultrasound medical imaging is a medical imaging technology that could lower costs by providing an alternative to higher-cost imaging techniques. The technology created during this Phase II project is expected to increase the quality, value, and accessibility of medical ultrasound, which would in turn reduce medical imaging costs in the US healthcare system. Furthermore, the company's technology is expected to bring ultrasound to more clinical settings and improve system-wide efficiencies in the diagnosis and treatment of disease. The technology also has commercial potential in the international market, with $5.8B spent annually on medical ultrasound devices worldwide. Finally, by improving the utility of ultrasound, the technology will lead to improved patient care and may ultimately save lives.
This Small Business Innovation Research (SBIR) Phase II project will develop deep learning technology for ultrasound imaging in medicine. Ultrasound imaging has numerous benefits including real-time image acquisition, non-invasive scanning, low-cost devices, and no known side-effects (it is non-ionizing). However, variability in quality has encumbered its adoption and utility. As a result, more expensive imaging is typically utilized, often exposing patients to ionizing radiation. Our objective is to develop, improve, and test machine learning techniques, based on deep learning, to improve ultrasound acquisition and interpretation. We expect this project will create novel technologies that make ultrasound easier to use and improve the quality of ultrasound examinations. The end result will improve the quality, value, and accessibility of medical ultrasound examinations, will result in cost savings to the healthcare system, will produce improvements in patient care, and will support a sustainable business opportunity.
PROJECT OUTCOMES REPORT
Disclaimer
This Project Outcomes Report for the General Public is displayed verbatim as submitted by the Principal Investigator (PI) for this award. Any opinions, findings, and conclusions or recommendations expressed in this Report are those of the PI and do not necessarily reflect the views of the National Science Foundation; NSF has not approved or endorsed its content.
As part of this Phase II project, we developed, implemented, and verified Caption GuidanceTM, a software that assists medical professionals in the acquisition of cardiac ultrasound images. Caption Guidance uses artificial intelligence to provide real-time guidance, diagnostic quality assessment, and automated capture of ultrasound images. This enables healthcare providers--even those without prior ultrasound experience--to obtain diagnostic-quality cardiac images. Empowering more clinicians with ultrasound acquisition capabilities brings the known benefits of ultrasound to more patients, helping elevate and standardize quality of care while also assisting hospitals in realizing valuable cost and time savings. It also enables broader use of ultrasound to detect disease earlier--when there is the highest potential for impact--in more accessible and convenient settings such as primary care offices and even patients′ homes.
Caption Guidance is equipped with numerous features that together act as a co-pilot for clinicians when performing an ultrasound exam. The software emulates the actions that an expert sonographer would take to optimize the image, including providing real-time guidance on how to manipulate the transducer, and automated feedback on diagnostic image quality. Following a brief training on the software, any healthcare professional can capture diagnostic-quality ultrasound images thanks to AI guiding them through every step of the scanning and image-capture process in real time. This expands the number of clinicians able to perform ultrasound from ~74,000 sonographers in the U.S. to an estimated 725,000 medical assistants, 722,000 licensed practice nurses, 3.1 million registered nurses, and many more.
The project included extensive performance testing, including a pivotal multicenter prospective clinical trial conducted with Northwestern Medicine and Minneapolis Heart Institute at Allina Health evaluating the use of Caption Guidance by registered nurses (RNs) with no prior ultrasound experience.
In this study, eight RNs with no prior ultrasound experience used Caption Guidance to perform ultrasound exams on 240 patients following a short training course. Patients were stratified to include a wide range of body-mass index and cardiac pathologies. The RNs acquired limited echo exams of 10 views each. Each exam was assessed by a panel of 5 expert cardiologists to determine if the exam was of sufficient quality to make a set of specific qualitative visual assessments.
Caption Guidance successfully met all four primary endpoints, meeting the pre-specified criteria for study success, by acquiring images of sufficient quality for specific clinical assessments. Namely, the RNs successfully acquired limited echo exams for qualitative visual assessments of left ventricular size: 98.8%, 95% Confidence Interval (CI) [96.7, 100]; left ventricular function: 98.8%, CI [96.7, 100]; right ventricular size: 92.5%, CI [88.1, 96.9]; and pericardial effusion: 98.8%, CI [96.7, 100].
In 2018, the FDA granted Breakthrough Device designation to Caption Guidance. To qualify for such designation, a device must provide for more effective treatment or diagnosis of a life-threatening or irreversibly debilitating disease or condition, and meet additional criteria including being a breakthrough technology with no approved alternatives and offers significant advantages over existing alternatives.
Caption Guidance was then granted landmark marketing authorization via the De Novo pathway, a regulatory pathway reserved for novel technologies. The granting of this De Novo was groundbreaking, as Caption Guidance was the first medical software authorized by the FDA that provides real-time AI guidance for medical imaging acquisition.
Caption Guidance was also approved for new technology add-on payments (NTAP) by the Centers for Medicare and Medicaid Services (CMS) in 2021. Medicare patients receiving inpatient care and eligible for NTAP have a newly created ID-10 procedure code corresponding to Caption Guidance. The NTAP designation is awarded to new medical technologies and services which are expected to substantially improve the diagnosis or treatment of Medicare beneficiaries.
Caption Guidance software can be integrated onto compatible ultrasound devices. It is paired with Caption Health′s Caption InterpretationTM which includes an automated calculation of ejection fraction, based on the cardiac images captured. Together, these capabilities are offered to the market as Caption AITM.
Last Modified: 02/01/2022
Modified by: Charles F Cadieu
Please report errors in award information by writing to: awardsearch@nsf.gov.
