OUSD (R&E) CRITICAL TECHNOLOGY AREA(S): Combat Casualty Care OBJECTIVE: Design, build, and validate results of a non-contact Laser Vital Signs Monitor (ncLVSM) in the form of a stand-alone lightweight portable cellphone-sized self-steering laser vibrometry device, constructed using integrated photonics. DESCRIPTION: Main vital signs (VS) consist of core body temperature (Tc), heart rate (HR; pulse), respiratory rate (RR) and blood pressure (BP) An extended set includes oxygen saturation (SpO2), level of consciousness, and pain. VS of battlefield wounded are critical for Medic's triage at the point-of-injury (POI) and monitoring during prolonged field care (PFC). Standard manual means for monitoring VS are inefficient when first responders are focused on priorities of hemorrhage cessation and wound care. One approach is to use non-contact VSMs. Non-contact VSMs have been developed recently with optical, radar, thermal, and Laser Doppler Vibrometry (LDV) technology, and passive and active acoustic sensing. These devices suffer an innate limitation due to steering and localization control, requiring either manual direction or a restriction in the subject's position (e.g. [1]). The ncLVSM created in this project will use integrated photonics-based optical, laser, and LDV technology [2] capable of steering using computer vision and learned-sensing control. LDV can make non-contact vibration measurements of a surface struck by a laser [3]. It was shown [4] that LDV with manual steering can record a patient's Arterial Waveform (AWF) when signal is acquired from a body pulse point. Analysis of the AWF signal [5,6] yields HR and BP (as systolic BP (SBP) and diastolic BP (DBP)). The patient's RR is interrogated from signals acquired from chest expansions. The Tc is computed from the HR using for example the ECTemp algorithm [7] or another multi-wavelength thermography technique. The ncLVSM will include an onboard camera and computer, then use computer vision incorporating pose recognition [8-10] to steer the interrogation laser beams. Pulse points are located by morphing a gender-specific standard anatomy surface mesh with labeled pulse points onto the patient's body surface, with locations adjusted for movement by tracking software. ncLVSM is intended to operate hands-free, for example, attached to the front of a first-responder's jacket, helmet, or unmanned aerial vehicle. PHASE I: The main goal of Phase I is a feasibility study in the development of a portable ncLVSM device. The ncLVSM must be designed to acquire data to compute the Tc, HR, RR and BP as SBP and DBP. The major components of the ncLVSM are to include the laser diode, silicon photonics for laser transmit and receive components, camera, computer processor(s), circuit board, rechargeable battery, transmission antenna, on/off power switch and small alphanumeric display screen. A thermally sensitive camera is an optional feature for nighttime pose recognition. As the first deliverable, a physical, electronics, optical, photonics and circuit design of the final ncLVSM product is to be completed to prove feasibility. The designs may include commercial components accompanying custom-designed photonics components. The physical design of the ncLVSM must have a form factor of approximately the width and height of a cellphone, must be appropriate for rugged civilian or battlefield applications, and must operate throughout the range of arctic to desert temperatures. The ncLVSM should operate by battery for a minimum of two hours of combined time use prior to battery recharging or replacement. Innovation is encouraged in each design aspect to create a lighter, more rugged, longer charged device. The device is to contain a computer processor(s) capable of performing the computations necessary to redirect the laser beam for locating and tracking a pulse point in case of movement. The VS should be computed at approximately 1Hz, displayed on the device, transmitted to external devices in real-time, and stored for later download. A second deliverable is a CAD computer model of the device, accompanied by a physical 3D printed model of the device. A third deliverable is a schematic description of the data acquisition process and software for each task. Existing software and planned software in the scheme should be indicated. A practically attainable AWF analysis methodology must be described. PHASE II: The overall objective of Phase II is to produce one fully operational portable ncLVSM prototype. The prototype device must perform these tasks: recognize the subject body form; recognize the body pose; morph the body surface mesh of standard anatomy into the form of the patient; locate the pulse points on the patient from labels on the standard body mesh; acquire AWF VS; acquire signal from chest expansions; display the VS; store the information; decide if information is to be transmitted to an external device; and, continuously repeat this process. The first goal of Phase II is to produce a prototype hardware based on the silicon photonics and electronics design of Phase I. The emphasis should be focused on hardware integration and operation during this stage. This task will produce the first deliverable, a functioning prototype of the ncLVSM that acquires observable LDV signals from an inanimate phantom. The project then requires the design and programming of software operations detailed above. A second deliverable is the demonstration of the fully functional prototype ncLVSM in the desired cellphone form factor, complete with the computer software needed to perform signal acquisition and all functions for computation, display, data storage and transmission. Laser power deposition must be demonstrated to not exceed FDA guidelines. The ncLVSM must be demonstrated to acquire data from a human subject, under an IRB-approved research protocol. Subject movement should be included to demonstrate operation in non-static conditions. The third deliverable consists of 1) Providing one fully functional prototype ncLVSM device, accompanied by details of the electronics and integrated photonics design. 2) All software code that includes validation test reports and other relevant reports. 3) A regulatory strategy that reflects a clear plan on how FDA clearance will be obtained. Early FDA coordination may be considered to assist with the regulatory strategy for obtaining approval for use as a medical monitoring device. PHASE III DUAL USE APPLICATIONS: An aim would be to develop training software, sample input and create manuals for the system. Due to the device's small size and likely modest price, the main target for the product is the mass commercial market, i.e. primary care physicians, clinics, and EMT use. ncLVSM use when mounted on an UAV, to provide an unmanned triage capability, is another application for both the military and civilian markets. This phase shall include FDA submission with the goal of FDA approval. In conjunction with FDA submission, the contractor can develop scaled up manufacturing of the technology that follows FDA quality regulations. Utility is enhanced if the device was easily able to transmit VS data in a manner to be accessible to phone internet application(s), enabling telemedicine and potentially integrating with artificial intelligence. REFERENCES: 8. Laser can detect your heartbeat and breathing from a metre away , New Scientist, May 16, 2018. 9. Li Y, L Marais, H Khettab, Z Quan, A Aasmul, R Leinders, R Schuler, PE Morrissey, S Greenwald, P Segers, M Vanslembouck, RM Bruno, P Boutyrie, P O'Brien, M De Melis, R Baets, Silicon photonics-based laser Doppler vibrometer array for carotid-femoral pulse wave velocity (PWV) measurement, Biomedical Optics Express 11(7) 3913- 3926 (2020). 10. Kroschel, Kristian, Laser Doppler Vibrometry for Non-Contact Diagnostics, Springer Bioanalysis Series, 2020 11. Desjardins, CL, LT. Antontelli, E Soares, A remote and non-contact method for obtaining the blood-pulse waveform with a laser Doppler vibrometer, Proc SPIE 6430, Advanced Biomedical and Clinical Diagnostic Systems V, 64301C (6 Feb 2007). 12. Esper SA, MR Pinsky, Arterial Waveform Analysis, Best Practice & Research Clinical Anaesthesiology 28, 363-380 (2014). 13. Nirmalan M, PM Dark, Broader analysis of arterial pressure wave form analysis, Continuing Education in Anaethesia, Critical Care & Pain 14(6), 285-290 (2014) 14. Hunt, AP, MJ Buller, MJ Maley, JT Costello, IB Stewart, Validity of a noninvasive estimation of deep body temperature when wearing personal protective equipment during exercise and recovery, Mil Med Res 6: 2, 2019. 15. Shotton, J, A Fitzgibbon, M Cook, T Sharp, M Finocchio, R Moore, A Kipman, A Blake, Real-Time Human Pose Recognition in Parts from Single Depth Images, Computer Vision and Pattern Recognition 2011, 12897-1304, 2011. 16. Gamra, Ben, Miniar, Moulay A.Akhloufi, A review of deep learning techniques for 2D and 3D human pose estimation, Image and Vision Computing, 114, 104282, 2021. 17. Wang, J, STan, Xg Zhen, S Xu, F Zheng, Z He, L Shao, Deep 3D human pose estimation: A review, Computer Vision and Image Understanding, 210 193225, 2021. KEYWORDS: Laser, vibrometry, vital signs, biosensor, photonics, portable, self-steering, noncontact, monitor.
