Care delivery at primary and urgent clinics is critically hindered by doctor shortage, time constraints and the inappropriate use of infrastructures. Rising aging population and unhealthy lifestyle are expected to drive the increase of chronic disease and the national health expenditures associated (20% of the U.S. gross domestic product by 2025). Telemedicine has been proposed as a possible solution to reduce the cost of access to care minimizing the dependence of patients on health professionals. However, current solutions that connect patients with doctors dont allow accurate diagnosis due to the lack or inappropriate use of medical devices. AdviNow proposes to implement computer vision to offer a clinically reliable approach to perform self- evaluation of vital parameters. AdviNow will develop the first Augmented Reality (AR) algorithms that will guide any non-trained people to a correct usage of Class 2 medical devices at the point of need. The new approach will represent the core component of a novel clinical decision support system (CDSS) which will be integrated into telemedicine platforms, kiosks or any medical office to automate medical visits, improving chronic diseases management. In this Phase I application, we will develop a minimally viable application based on heart stethoscope exams, in order to assess the feasibility of the proposed approach for multiple medical devices to effectively monitor chronic diseases. Specific activities will include: (1) Development of AR-driven engine for the correct application of heart stethoscope; (2) development of a basic front-end application for Real-time instructions; (3) establishment of a classifier for heart sound abnormalities. We expect to achieve real-time (<300ms) estimation of phonocardiogram key points and accurate validation of the stethoscope position based on camera-capture and audio-recorded data. We expect that our system will recognize normal and abnormal heart sounds with >85% accuracy. Finally, a small feasibility test will be performed with healthy volunteers to verify their ability to position the stethoscope in less than 10 seconds with an error of less than one inch. The successful completion of this SBIR Phase I project will lay the foundation of a Phase II SBIR proposal for an external validation of the system to be performed in collaboration with urgent care clinics and patients affected by chronic heart conditions. The mission of AdviNow Medical is to develop a disruptive solution for the delivery of healthcare that is safe, effective, timely, efficient, and available in centralized and decentralized locations. By reducing the time for routine analysis, the actual doctor-patient encounter will be more meaningful, improving the clinical outcome as well as patient satisfaction. This invention promises to disrupt healthcare, as providers will improve cost management and profitability by visiting more patients.
Public Health Relevance Statement: PROJECT NARRATIVE New strategies for accurate and efficient delivery of healthcare to chronic disease patients are required. This project will develop the first Augmented Reality (AR) algorithms which will guide patients to use Class 2 medical devices for self-assessment of vital parameters to monitor the health status of chronic conditions. The project is expected to introduce a novel clinical decision support system (CDSS) to Primary and Urgent care clinics and remote settings to allow improved medical outcomes and cost management of chronic diseases.
NIH Spending Category: Assistive Technology; Bioengineering; Cardiovascular; Clinical Research; Health Services; Heart Disease; Machine Learning and Artificial Intelligence; Networking and Information Technology R&D (NITRD)
Project Terms: Abdomen; accurate diagnosis; Affect; aging population; Algorithms; Artificial Intelligence; Augmented Reality; Automation; Automobile Driving; base; Blood; care delivery; Caring; Chest; Chronic; Chronic Disease; Classification; Client satisfaction; Clinic; Clinical; clinical decision support; Clinical Decision Support Systems; Collaborations; computational intelligence; Computer software; Computer Vision Systems; Computers; convolutional neural network; cost; Data; Data Set; Decentralization; Dependence; design; Development; Diabetes Mellitus; Diagnosis; Diagnostic; Diffusion; Disease Management; Drops; Ear; Environment; Event; FDA approved; Foundations; Frequencies; Health; health care delivery; health care quality; Health Expenditures; Health Professional; Health Services Accessibility; Health Status; Healthcare; Healthcare Systems; healthy volunteer; Heart; Heart Abnormalities; Heart Diseases; Heart Sounds; Human; human subject; Image; improved; Infrastructure; innovation; insight; Instruction; interest; invention; Joints; Location; Lung; Measurement; Medical; Medical Device; Medical Records; Medicine; Mission; Modeling; Monitor; Nose; novel; novel strategies; Outcome; Oxygen; Patient-Focused Outcomes; Patients; Performance; Pharyngeal structure; Phase; Physical Examination; Positioning Attribute; predictive modeling; pressure; prevent; Primary Health Care; Process; Provider; Research Personnel; respiratory; Self Assessment; Self-Examination; Site; Small Business Innovation Research Grant; software development; sound; Stethoscopes; support tools; System; Technology; Telemedicine; Temperature; Testing; Therapeutic; Time; tool; unhealthy lifestyle; urgent care; Validation; Visit; Weight
