Assessing Hand Perfusion via Video Using Pigment Enhancement Technology

PURPOSE: Timely and accurate triage of upper extremity injuries is critical and can be challenging, especially in patients with higher skin melanin content.1-3 This pilot study evaluates a Eulerian Video Magnification (EVM) algorithm4 combined with color channel waveform extraction to enable video-based measurement of hand and finger perfusion across all skin types. We aimed to obtain preliminary data on how this technology impacts the accuracy and confidence of clinicians in detection of finger ischemia via video. METHOD: Videos of 10 volunteers with Fitzpatrick skin types III-VI were taken in a controlled environment during normal perfusion and tourniquet-induced ischemia. Videos were EVM processed (EVM), and red/green/blue color channel characteristics were extracted to produce waveforms. These videos were assessed by surgeons with a range of expertise in hand injuries. The videos were randomized and presented in one of three different ways: unprocessed, EVM, and EVM with accompanying waveform (EVM+waveform). Survey respondents indicated whether the video showed an ischemic or perfused finger(s), or if unable to tell. We used group comparisons to evaluate response accuracy across video types, skin tones, and respondent groups, and inter-rater agreement analysis using the kappa statistic. RESULTS: Of 51 surveys sent, 25 (49%) responded. The frequencies of correct responses were statistically significantly higher in the EVM+waveform category compared to unprocessed or EVM videos. Participants assessing unprocessed and EVM videos had a decrease in accuracy with increasing skin melanin content. EVM+waveform outperformed the others, and that was consistent across the different levels of experience/training. Also, there was higher agreement amongst responses for the EVM+waveform group compared to unprocessed or EVM for all questions. Accuracy and agreement in the EVM+waveform group was consistent across all skin types evaluated. CONCLUSION: Video-based EVM processing combined with waveform extraction improved assessment of perfusion in the distal upper extremity. These technologies have potential future applications including triage, post-surgery vascular assessment, and telemedicine. REFERENCES: 1. Chae MP, Rozen WM, Whitaker IS, et al. Current evidence for postoperative monitoring of microvascular free flaps: a systematic review. Ann Plast Surg. May 2015;74(5):621-32. doi:10.1097/SAP.0b013e3181f8cb32 2. Polfer EM, Zimmerman RM, Tefera E, Katz RD, Higgins JP, Means KR, Jr. The Effect of Skin Pigmentation on Determination of Limb Ischemia. J Hand Surg Am. Jan 2018;43(1):24-32 e1. doi:10.1016/j.jhsa.2017.09.002 3. Mahmoudi E, Swiatek PR, Chung KC, Ayanian JZ. Racial Variation in Treatment of Traumatic Finger/Thumb Amputation: A National Comparative Study of Replantation and Revision Amputation. Plast Reconstr Surg. Mar 2016;137(3):576e-585e. doi:10.1097/01.prs.0000479969.14557.9d 4. Wu H-Y, Rubinstein M, Shih E, Guttag J, Durand F, Freeman W. Eulerian Video Magnification for Revealing Subtle Changes in the World. ACM Trans Graph. 2012;31(4)doi:10.1145/2185520.2185561