ABSTRACT
A focused imaging system such as a camera will reflect light directly back at a light source in a retro-reflection (RR) or cat-eye reflection. RRs provide a signal that is largely independent of distance providing a way to probe cameras at very long ranges. We find that RRs provide a rich source of information on a target camera that can be used for a variety of remote sensing tasks to characterize a target camera including predictions of rotation and camera focusing depth as well as cell phone model classification. We capture three RR datasets to explore these problems with both large commercial lenses and a variety of cell phones. We then train machine learning models that take as input a RR and predict different parameters of the target camera. Our work has applications as an input device, in privacy protection, identification, and image validation.
ABSTRACT
BACKGROUND: Decreased autologous flap vascular perfusion can lead to secondary procedures. Fluorescence angiography during surgery reduces the probability of repeat surgery but suffers from interpretation variability. Recently, the OnLume Avata System was developed, which evaluates real-time vascular perfusion in ambient light. This study aims to predict complications in autologous breast reconstruction using measures of relative intensity (RI) and relative area (RA). METHODS: Patients undergoing autologous breast reconstruction underwent intraoperative tissue perfusion assessment using the OnLume Avata System. Post-hoc image annotation was completed by labeling areas of the flap interpreted to be "Well Perfused," "Questionably Perfused," and "Under Perfused." RIs and RAs were calculated for the marked areas. Primary complications of interest were overall complication rate, fat and mastectomy skin flap necrosis, and surgical revision. Logistic regression was applied to determine the odds of developing a complication based on RI and RA for each image. RESULTS: A total of 25 patients (45 flaps) were included. In total, 17 patients (68%) developed at least one complication. Patients who developed any complication (p = 0.02) or underwent a surgical revision for complications (p = 0.02) had statistically lower RI of under-perfused portions of the flap. Patients with greater areas of under-perfused flap had a significantly higher risk of developing fat necrosis (odds ratio [OR]: 5.71, p = 0.03) and required a revision operation (OR: 1.10, p = 0.01). CONCLUSION: Image-based interpretation using the OnLume Avata System correlated with the risk of developing postoperative complications that standard fluorescence imaging systems may not appreciate. This information can benefit surgeons to improve perfusion assessment and intraoperative decision-making.
ABSTRACT
Nipple-areolar complex (NAC) necrosis is a devastating complication in nipple-sparing mastectomies (NSMs) that significantly impacts patient's quality of life. The use of fluorescence angiography for intraoperative assessment of mastectomy skin flap perfusion in NSM has been successfully described and can be utilized to help guide surgical decision-making. Recently, a novel fluorescence-guided surgical imager was developed, OnLume Avata System (OnLume Surgical, Madison, WI), which provides intraoperative evaluation of vascular perfusion in ambient light. In this case report, we describe the use of OnLume fluorescence-guided surgery technology to help aid in clinical decision-making for two breast reconstruction cases with concern for intraoperative nipple hypoperfusion.
