One in three people with diabetes mellitus is at risk of diabetic foot ulcers (DFUs), with over 10% amputated.
The current global pandemic has driven a significant change in healthcare delivery and disrupted DFU care and
limb preservation, leaving many patients with limited or no clinical care.
Clinicians must adopt a paradigm shift
from the hospital and clinic care to community-based point-of-care (POC) - to best triage chronic DFU cases that
are high-risk lesions requiring clinical care or hospitalization. There is an unmet clinical need for smart health
assessment tools for POC treatment of patients with DFUs onsite, where no wound care expertise is available.
Smartphone technologies for wound care are limited to 2D/3D wound image analysis for size/depth. They are
insufficient as stand-alone tools to assess and triage high-risk DFU lesions without wound expertise onsite.
Hence, additional clinical assessments (e.g., the extent of oxygen supply to wound) are required during POC of
Oxygenation measurements provide a sub-clinical physiological assessment that complements
clinical visual assessment. We recently developed a smartphone-based NIR imaging approach or SmartPhone
Oxygenation Tool (SPOT) to obtain visual tissue oxygenation measurements in wounds. Systematic
assessment of the skin tone and wound characteristics is critical during physiological imaging and has not been
investigated to date. Hence, our objective is to develop and validate a smartphone-based imaging approach
(or Smart Scanner) capable of visual and physiological analysis of DFUs across the spectrum of skin tones and
wound features via automated machine learning (ML) algorithms. Developing a smartphone-based optical device
via integration of existing NIR imaging technology, but towards smart health platform for physiological
assessment of DFUs, while accounting for varying skin colors and wound types using ML algorithms is
innovative. The specific aims are: (i) Account for the effect of skin tones on oxygenation measurements by
applying light propagation models and machine learning algorithms and validate via phantom and in-vivo studies.
(ii) Analyze tissue curvatures and account for depth variations in-vivo oxygenation maps via studies on control
subjects (~15 cases). (iii) Differentiate wound tissue types and validate physiological imaging using the SPOT
device via DFU studies (~25 cases). The expected outcomes are: (i) Develop our Smart Scanner (SPOT
device + app) to obtain accurate tissue oxygenation maps across different skin tones and wound tissue types;
(ii) Validate our SPOT device to differentiate DFUs with high-risk lesions that require clinical care, from low-risk
cases. Incidence of DFUs and related amputation rates differ by race/ethnicity, and are higher in African
Americans, Hispanic and Native Americans compared to Caucasians. In the long term, SPOT can be used as
a smart health tool to pre-screen or triage DFUs with high-risk lesions to clinical care and thus minimize potential
amputations in any racial/ethnic group (with varying skin tones).