SBIR Phase I: Integrating Patient Photographs with Medical Imaging Examinations to Reduce Medical Errors

The broader impact/commercial potential of this Small Business Innovation Research (SBIR) Phase I project is widespread adoption of a technology with a direct cost-­savings in healthcare. It is estimated that nearly 5,000 patients are harmed each year due to wrong-patient errors in medical imaging. Additionally, even a 10% improvement in efficiency in radiologists' performance would translate to ~$900 million savings to the overall health care budget. Finally, an indirect, but impactful outcome is the potential to improve the relevance and accuracy of medical imaging reports. It is anticipated that pairing the medical imaging studies to the patient photographs of the patient's face and upper body, where tubes and wires are often located, can provide important clinical information in a rapid and efficient manner. In the broadest terms, this novel technology can be translated to provide patient authentication and identification for the digital data that is being generated by hundreds of new medical devices. The proposed project concerns the seamless integration of a patient Identification system in hospitals to improve patient safety and radiology throughput. The intellectual merit of this work is a transformative approach overcoming the failure of existing patient identification methods while harnessing the power of an embedded camera system to improve patient care. Toward this goal this work will develop an imaging synchronization technology that automatically obtains and embeds a photograph of the patient's face and upper body onto the x-ray taken during an imaging examination. To test and establish feasibility, the initial target of this technology will be portable x-ray machines brought to the patient's bedside in high volume settings. Specifically, the proposed work will 1) incorporate face-­tracking technology to ensure the patient's face and torso are always captured by the camera, 2) conduct a feasibly study in a real-world setting by deploying the technology on ~6 portable radiography machines at a large academic hospital where it will be integrated with the hospital's Picture Archiving and Communication System (PACS), and 3) perform preliminary workflow studies. The long term objectives are to 1) increase the detection rate of wrong-­patient errors by embedding an intrinsic, externally visible biometric identifier with medical imaging studies; and 2) decrease turn-­around time by decreasing interpretation time.