Abstract
Purpose
Accurate assessment and monitoring of tissue perfusion are critical in diagnosing and managing vascular diseases of the hand. Traditional methods, including clinical examination, pulse oximetry, and Doppler ultrasound, have limitations in detecting early or subtle perfusion deficits. Hyperspectral imaging (HSI) is a noninvasive, real-time technology that generates quantitative perfusion maps by measuring oxygen saturation, oxyhemoglobin, and deoxyhemoglobin levels. To the best of our knowledge, this is the first study of its kind to evaluate the clinical use of HSI in assessing perfusion abnormalities in vascular conditions of the hand and its emerging applications in hand surgery.
Methods
A single center study was conducted from 2021 to 2025, involving 5 patients with various upper-extremity vascular conditions, each followed to early 2025. Perfusion images were obtained using the HyperView (HyperMed, Memphis, TN) system before and after surgical or procedural interventions. Results were compared against the contralateral hand or historical normal values from the prior-generation HyperView (OxyVu-1) system.
Results
Hyperspectral imaging effectively identified perfusion deficits, guided urgent clinical decision-making, and monitored postoperative improvements. In one case, HSI demonstrated a 5% increase in tissue oxygen saturation after ulnar artery reconstruction, correlating with successful wound healing. Another patient with Raynaud’s phenomenon showed a 23% perfusion increase following botulinum toxin injection. Hyperspectral imaging also detected thrombosis and monitored perioperative oxygen saturation changes in real-time.
Conclusions
Hyperspectral imaging offers a novel, objective approach to perfusion assessment in hand and upper-extremity vascular disease, addressing key limitations of traditional methods. Its integration into clinical workflows holds promise for enhancing real-time, data-driven decision-making in hand surgery and vascular medicine.
Type of Study/level of evidence
Diagnostic IV.
Key words: Hand disorders, Hyperspectral imaging, HyperView, Tissue perfusion, Vascular disease
Adequate evaluation of tissue perfusion is essential for diagnosing, guiding revascularization, and monitoring hand vascular disease. Oxygen delivery is critical for wound and disease healing, yet current perfusion assessment methods have limitations. The most commonly used modalities are pulse oximetry and Doppler ultrasound.1 However, these may miss early ischemic changes, as they require pulsatile blood flow. Angiography is invasive, costly, involves radiation and contrast dye, and lacks quantitative tissue perfusion metrics.
Hyperspectral imaging (HSI) is a novel technique providing real-time, objective, and user-independent assessment of oxygen saturation, oxyhemoglobin, and deoxyhemoglobin, creating two-dimensional perfusion maps. Unlike traditional methods, it is waveform-free and assesses large areas simultaneously. An example is HyperView (HyperMed, Memphis, TN), a Food and Drug Administration 510(k)-cleared device using visible light and wide-field diffuse reflectance spectroscopy to measure light reflection and absorption from capillaries 0.1 mm beneath the skin, assessing tissue metabolic state and microvasculature better than arterial oxygen metrics. Differences in chromophores between oxygenated and deoxygenated hemoglobin allows distinction. The camera is hand-held and uses an overlapping dual laser pointer for accurate focus. The software analyzes captured cutaneous tissue without patient contact, contrast, or dye.
Since its initial development in 1983 for earth remote sensing, HSI has proven effective in medicine but is not applied widely to hand conditions.2, 3, 4 Hyperspectral imaging has shown promise in predicting diabetic foot ulcer prognosis, burn healing in animal models, flap monitoring, and screening for lower-extremity peripheral vascular disease.5, 6, 7, 8
However, application of HSI in hand diseases requiring blood flow interventions remains unexplored. This study aims to evaluate its use in this setting and review the current literature on the emerging uses of HSI in the management of ischemic hand conditions.
Materials and Methods
This study was conducted at a single center between 2021 and early 2025 and included 5 patients who presented with various vascular and circulatory issues of the hand. Study patients were adults with confirmed hand vascular deformities who were able to provide informed consent and undergo HSI imaging. Patients were excluded if they had an active infection in the hand, known allergy to device materials, or inability to comply with imaging protocols. Institutional review board approval was obtained for this study. Informed consent and, when applicable, Health Insurance Portabiity and Accountability Act authorization was obtained from all participants.
Perfusion images were taken before and after surgical interventions or procedures and, if possible, compared to either the patient’s healthy opposite hand or standardized normal ranges for palm and fingertip perfusion offered by the previous iteration of the HyperView camera (OxyVu-1). The HyperView camera is calibrated by the manufacturer, but verification of calibration and laser alignment can be checked every three months. The camera is calibrated in a darkened room with an ambient temperature of 65 to 78 °F (18 to 26 °C). The HyperView system has an infrared temperature sensor that measures surface temperatures and gives feedback as needed on temperatures that may alter results. It is recommended that the camera be placed perpendicular to the skin surface to align the two laser pointers for focus, to capture an accurate image. No other standardized protocols, including the use of a tripod device, were necessary for accuracy.
Hyperspectral imaging was performed using the HyperView (HyperMed, Memphis TN) system. The HyperView camera delivers visible light waves to the skin’s surface. The HyperView camera has a spectrometer that measures various wavelengths of delivered, reflected and absorbed light from the skin’s surface, creating a gradient map with pixels of data corresponding to 0.1 mm area. Oxyhemoglobin and deoxyhemoglobin have unique spectroscopic characteristics, specifically, they have differing absorption peaks and magnitudes of each chromophore’s contribution, allowing for the HyperView system’s ability to differentiate between the two while normalizing for melanin absorption (Fig. 1). Hyperspectral tissue oxygen saturation (HT-Sat %) is calculated as hyperspectral tissue oxyhemoglobin (HT-Oxy) divided by the sum of HT-Oxy and hyperspectral tissue deoxyhemoglobin (HT-Deoxy), multiplied by 100% (Fig. 2). The perfusion map reflects these data in a color-coded gradient, and a boundary tool can select regions of the image to analyze.
Figure 1.
Schematic of the mechanism of the HyperView camera, adapted from the HyperView manual. Created with BioRender.com.
Figure 2.
Hyperspectral tissue oxygen saturation (HT-Sat %) is calculated as hyperspectral tissue oxyhemoglobin (HT-Oxy) divided by the sum of HT-Oxy and hyperspectral tissue deoxyhemoglobin (HT-Deoxy), multiplied by 100%.
Results
Case 1
A 70-year-old man with a history of leukemia on chronic steroids was evaluated for an acute atraumatic ulceration of the right index finger. Angiography demonstrated slow filling through the radial artery and a corkscrew anomaly of the ulnar artery near the wrist. Preoperative HSI, focused on the volar aspect of the right index finger distal interphalangeal joint, demonstrated a decreased HT-Sat % of 39 (Fig. 3). The patient underwent resection of the ulnar artery corkscrew anomaly and reconstruction with a vein graft. Compared to preoperative HSI, a 5% improvement in the same region was recorded one day after surgery. Given the patient’s steroid requirements, his ulcer healed slowly. Tissue saturation was monitored for at least four months after surgery, with ultimate improvement to saturation of 58%, as measured at his right IF volar P3. This value was comparable to the HT-Sat % of his healthy left hand DP. The ulcer ultimately healed uneventfully.
Figure 3.
Case 1 Hyperspectral images of right index finger (pathological finger) before surgery and after surgery, compared to left index finger (comparison finger).
Case 2
A 51-year-old woman presented with a longstanding history of symptomatic Raynaud’s disease as a sequela of limited scleroderma. In September 2024, she received 200 units of botulinum toxin across all 10 fingers, and she was followed up one month later. Before injection, HSI recorded a HT-Sat % of 27, HT-Oxy of 139, and HT-Deoxy of 51 on the volar side of her left middle finger. One month later, she reported a reduction in pain and HSI of the same finger demonstrated an HT-Sat % of 50, HT-Oxy of 105, and HT-Deoxy of 107 (Fig. 4).
Figure 4.
Improvement in finger perfusion measured by hyperspectral imaging one month after botulinum toxin injection for Raynaud’s secondary to limited scleroderma.
Case 3
A 52-year-old man presented to the emergency department with a left ring finger crush injury. The digit was cool, had delayed capillary refill, and decreased oxygen saturation on pulse oximetry compared to the other fingers. X-ray showed no fracture. Initial conservative treatment with aspirin, a calcium channel blocker, and warming failed to improve the digit. Hyperspectral imaging revealed decreased perfusion relative to uninjured fingers. Angiogram confirmed traumatic occlusion of both proper digital arteries at the metacarpophalangeal joint flexion crease. The patient underwent microvascular repair of the radial and ulnar arteries at the metacarpophalangeal joint crease the next day. After surgery, the finger was well perfused with normal capillary refill and strong Doppler signals at the pulp tip. Hyperspectral imaging showed improved distal perfusion (Fig. 5). The postoperative course was uneventful, and the patient was discharged.
Figure 5.
Hyperspectral images of a left ring finger crush injury. Numerical values were lost.
Case 4
A patient presented to the clinic with a diagnosis of complex regional pain syndrome in his right hand for the past six months. The HT-Sat %, HT-Oxy, and HT-Deoxy of his right midpalm were 44, 69, and 89, respectively. The values on his left, nonpathological hand were 51, 89, and 85. Hyperspectral imaging demonstrated an objective decrease in HT-Sat % on the hand with CRPS (Fig. 6). In this case, decreased perfusion identified by hyperspectral imaging is not intended to suggest a causal or diagnostic mechanism for CRPS. Rather, this finding represents an objective physiologic observation in a patient with established CRPS.
Figure 6.
Hyperspectral imaging reveals decreased perfusion in the right hand of a patient with complex regional pain syndrome compared to the unaffected left hand.
Case 5
A 45-year-old woman with bilateral Raynaud’s disease secondary to systemic sclerosis was evaluated for surgical sympathectomy. She had no history of finger ulcerations and was on maximal medical therapy including calcium channel blockers, sildenafil, and prior botulinum toxin injections that had lost efficacy. Examination showed deep purple fingertip discoloration and absent Doppler signals.
Before left palmar sympathectomy, distal middle finger measurements were HT-Sat 41%, HT-Oxy 91, and HT-Deoxy 131. Eleven days after surgery, these improved to HT-Sat 49%, HT-Oxy 107, and HT-Deoxy 109; at 34 days, values further improved to HT-Sat 53%, HT-Oxy 135, and HT-Deoxy 119. The reduction in HSI values observed on day 15 after sympathectomy may reflect transient vasospasm, postoperative edema, or environmental influences, such as room temperature (Fig. 7).
Figure 7.
Hyperspectral imaging demonstrates improved fingertip perfusion following left palmar sympathectomy in a patient with severe Raynaud’s disease secondary to systemic sclerosis.
The right hand sympathectomy showed preoperative distal middle finger values of HT-Sat 47%, HT-Oxy 116, and HT-Deoxy 131. Six days after surgery, values were HT-Sat 50%, HT-Oxy 106, and HT-Deoxy 107; at 15 days after surgery, measurements were HT-Sat 44%, HT-Oxy 90, and HT-Deoxy 115.
Clinically, the patient had notable symptom that improvement following both procedures.
Discussion
The results highlight HSI as a valuable, noninvasive tool for assessing and monitoring hand perfusion in various vascular conditions. Hyperspecteral imaging provides physiologic assessment of tissue perfusion and can be used as an adjunct to angiography when detailed vascular anatomy is required. It provides real-time, quantitative measurements of tissue oxygenation and offers certain advantages over traditional methods, including noninvasive assessment and broader surface area coverage. However, further studies are needed to compare outcomes directly between HSI and established techniques.
Our study demonstrated the use of HSI through multiple clinical scenarios. In Case 1, HSI effectively documented perfusion improvement following surgical revascularization in a patient with atraumatic ulceration and vascular anomalies, providing clear quantitative data on healing progression without disrupting care.
Case 2 showed the effectiveness of HSI in objectively quantifying perfusion improvements after botulinum toxin injections in Raynaud’s disease secondary to limited scleroderma. Given the challenge of insurance coverage for these injections because of limited efficacy evidence, the objective data of HSI may help justify treatment and improve patient access.
In Case 3, HSI complemented standard diagnostics by confirming perfusion deficits before surgery and demonstrating postoperative improvements after microvascular repair in a traumatic crush injury, aiding clinical decisions.
Case 4 saw HSI objectively identify reduced perfusion in a patient with complex regional pain syndrome, providing quantitative physiologic information in a condition often assessed using subjective symptoms.
Finally, Case 5 tracked perfusion improvements following palmar sympathectomy in a patient with severe bilateral Raynaud’s secondary to systemic sclerosis, while highlighting variability caused by external factors, such as lighting and temperature, underscoring the need for standardized imaging protocols.
Collectively, these cases illustrate the diverse clinical applications and advantages of HSI in hand vascular assessment, emphasizing its potential to enhance clinical decision-making and patient outcomes (Fig. 8). Beyond clinical benefits, HSI could contribute considerably to research by evaluating treatments in clinical trials and refining diagnostic criteria for hand vascular conditions. Furthermore, HSI-generated perfusion maps offer intuitive visuals for patients, improving communication and monitoring of treatment progress.
Figure 8.
Hyperspectral imaging assessments of tissue perfusion in cases 1, 2, 4, and 5. The figure compares before surgery and postoperative perfusion maps, along with representative physiologic and pathologic perfusion patterns, highlighting differences in oxygenation and perfusion distribution.
Hyperspectral imaging has limitations. It lacks established baseline reference values for every anatomical region, making its data harder to interpret. Though not intended to serve as control values, prior studies using the OxyVu-1 camera suggest average perfusion values for the palm (Table).9,10 Fingertips may have similar HT-Sat % values to the palm but likely differ in HT-Oxy and HT-Deoxy because of higher capillary density.
Table 1.
Normal Values for Hyperspectral Data
| Source/Area of Measurement | HT-Oxy | HT-Deoxy | HT-O2 % |
|---|---|---|---|
| Control index fingertip values | 124±16 | 73±10 | 63±4% |
| Control midpalm values | 88±9 | 70±18 | 57±8% |
| Literature midpalm values∗ | 70±21 | 56± 19 | 55±13% |
Literature values obtained using OxyVu-1 iteration of hyperspectral camera.
The HyperView system adjusts for melanin, but highly pigmented or heterogeneous areas still may produce inaccurate readings. However, our study only interrogated palmar skin, with much less pigmentation heterogeneity. Environmental factors also impact results — temperatures <80 °F (27 °C), recent nicotine or vasoconstrictor use, and incomplete acclimation to ambient temperature can lower oxyhemoglobin and oxygen saturation values. One trial demonstrated that cooling followed by reperfusion allowed for more accurate assessment of perforator perfusion during surgical flap harvest, highlighting how vascular reactivity can alter results. Conversely, vasodilation from substances like caffeine also may affect readings, though these effects remain under-studied. Importantly, perfusion values may not correlate with patient symptoms or clinical improvement, and it is unclear if normal ranges could be standardized across different devices. Our study did not include a control cohort or repeated same-day or longitudinal measurements in asymptomatic individuals. Future studies incorporating left-to-right comparisons in healthy controls and repeated measurements over time may help establish baseline variability and strengthen interpretation of results in different pathologies.
At present, to the best of our knowledge no dedicated reimbursement codes exist for hyperspectral imaging in hand perfusion assessment, and cost coverage may require institutional or research funding. The HyperView system is intended for routine clinical use and does not require formal operator training. Images were obtained following the manufacturer’s instruction manual, which includes guidance on lighting, room temperature, patient positioning, acclimation time, and minimizing motion. Because standardized reference values for hyperspectral perfusion metrics are not yet available, interpretation focused on within-patient comparisons, including before and after intervention measurements and comparison with the contralateral hand when available. The Table indicates normal values for hyperspectral data.
The purpose of this study is to report an existing, simple to use, technology which can assist in the management of vascular disorders of the upper extremity. We are not currently recommending it be used for surgical decision-making but, rather for monitoring both before and after intervention.
Of the limited existing literature on the use of HIS in hand pathologies, studies on Raynaud’s and systemic sclerosis are the most common. One prospective study found that, after a cold provocation challenge, Raynaud’s patients had a greater decline in oxyhemoglobin and oxygen saturation from baseline compared to healthy controls, supporting HSI as a quantitative tool for assessing Raynaud’s severity and activity.11
Most of the literature applies to HSI use in diseases of the lower extremity, such as diabetic foot ulcers and peripheral vascular disease.5 Khaodhiar et al.9 found that HSI had the capacity to identify with statistical significance changes in tissue immediately in the vicinity of diabetic foot ulcer when comparing ulcers that healed to those that did not.Yudovsky et al.7 used HSI to analyze skin surrounding a diabetic foot ulcer, creating a model to predict whether the ulcer would expand. They also used HSI to analyze changes in oxyhemoglobin and deoxyhemoglobin in nonulcerated sites that eventually ulcerated in o develop an ulcer formation prediction index.7 Early prognosis with HIS could improve preventive care and assess amputation risk.
Given its promise, it has similarly been applied in assessing peripheral artery disease. Hyperspectral imaging was shown to be unaffected by external confounders, such as temperature and inter-operator variability when used to assess oxygen saturation before and after surgical or endovascular therapy in 25 patients with peripheral artery disease.12 Hyperspectral imaging metrics could enhance disease characterization made with solely an ankle-brachial index and could be used to monitor perfusion.12 Additionally, it was used to compare oxyhemoglobin levels during brief occlusion (cuff) in people with and without peripheral vascular disease, suggesting its potential as a screening tool before clinical symptoms emerge and its ability to identify patients lacking a hyperemic response after proximal arterial occlusion. As a noninvasive, low-risk technique, HSI may help detect early peripheral vascular disease and inform management decisions.
Kohler et al.8 studied the use of HIS in monitoring flap perfusion in 22 patients who had an overall flap revision rate of 28%. They found that all patients with oxygen saturation or near-infrared perfusion <40 at 16–28 hours after surgery required revision, while none with saturation or perfusion >40 did.8 However, surface perfusion varies greatly by tissue type and location; for example, fingertip oxygen saturation differs from that on the abdomen, making regional comparisons difficult. No definitive absolute values yet correlate with healing or symptoms like pain. Current data suggest tissue oxygenation of ≥50 or may support normal wound healing, while levels <30 predict poor healing.13
Hyperspectral imaging might be a useful adjunct in determining the extent of burn and radiation-induced tissue injuries.6 Using mice models, HSI proved to be useful tool in distinguishing between intermediate-dermal, deep-dermal, and full thickness burn wounds. Measured hemoglobin and oxygen saturation via HSI could be improve early burn assessment and treatment options for burn patients.
Recent work using video-based deep learning models to extract imaging photoplethysmography waveforms from smartphone videos has demonstrated similarly the feasibility of noncontact classification of perfused versus ischemic fingers in control volunteers and acutely injured patients, further supporting the role of image-derived perfusion analytics as an adjunct in hand vascular assessment.14 In parallel, pigment-enhancing video technologies, such as Eulerian Video Magnification with waveform output, have been shown to improve ischemia detection even by inexperienced evaluators across a range of skin tones, underscoring the broader potential for objective, image-based perfusion assessment tools in hand vascular disease.15 However, unlike pigment-enhancing video or video-based photoplethysmography, which infer perfusion from pulsatile color changes, HSI directly measures oxyhemoglobin, deoxyhemoglobin, and tissue oxygen saturation to create quantitative perfusion maps.
As implications for future work, HSI shows promise as a user-friendly, nonharmful method for immediate perfusion assessment but requires further study for safe, widespread use. The impact of vasoactive substances on HSI values should be quantified to improve data validity. Similarly, the influence of topical products warrants investigation; the manufacturer of the HyperView System recommends thorough skin cleansing before imaging. Standardizing cleansing protocols and imaging best practices, including image timing relative to procedures, hand positioning, digit selection, and camera view angles (lateral, ventral, and so forth), is essential to enhance data quality and consistency.
Conflicts of Interest
No benefits in any form have been received or will be received related directly to this article.
References
- 1.Considine J. The reliability of clinical indicators of oxygenation: a literature review. Contemp Nurse. 2005;18(3):258–267. doi: 10.5172/conu.18.3.258. [DOI] [PubMed] [Google Scholar]
- 2.Goetz A. Three decades of hyperspectral remote sensing of the Earth: a personal view. Remote Sensing of Environment. Remote Sens Environ. 2009;113:S5–S16. [Google Scholar]
- 3.Lu G., Fei B. Medical hyperspectral imaging: a review. J Biomed Opt. 2014;19(1) doi: 10.1117/1.JBO.19.1.010901. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Linek M., Felicio-Briegel A., Freymüller C., et al. Evaluation of hyperspectral imaging to quantify perfusion changes during the modified Allen test. Lasers Surg Med. 2022;54(2):245–255. doi: 10.1002/lsm.23479. [DOI] [PubMed] [Google Scholar]
- 5.Sumpio B.J., Citoni G., Chin J.A., Sumpio B.E. Use of hyperspectral imaging to assess endothelial dysfunction in peripheral arterial disease. J Vasc Surg. 2016;64(4):1066–1073. doi: 10.1016/j.jvs.2016.03.463. [DOI] [PubMed] [Google Scholar]
- 6.Chin M.S., Babchenko O., Lujan-Hernandez J., Nobel L., Ignotz R., Lalikos J.F. Hyperspectral imaging for burn depth assessment in an animal model. Plast Reconstr Surg Glob Open. 2016;3(12) doi: 10.1097/GOX.0000000000000558. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Yudovsky D., Nouvong A., Pilon L. Hyperspectral imaging in diabetic foot wound care. J Diabetes Sci Technol. 2010;4(5):1099–1113. doi: 10.1177/193229681000400508. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Kohler L.H., Köhler H., Kohler S., et al. Hyperspectral Imaging (HSI) as a new diagnostic tool in free flap monitoring for soft tissue reconstruction: a proof of concept study. BMC Surg. 2021;21(1):222. doi: 10.1186/s12893-021-01232-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Khaodhiar L., Dinh T., Schomacker K.T., et al. The use of medical hyperspectral technology to evaluate microcirculatory changes in diabetic foot ulcers and to predict clinical outcomes. Diabetes Care. 2007;30(4):903–910. doi: 10.2337/dc06-2209. [DOI] [PubMed] [Google Scholar]
- 10.Greenman R.L., Panasyuk S., Wang X., et al. Early changes in the skin microcirculation and muscle metabolism of the diabetic foot. Lancet. 2005;366(9498):1711–1717. doi: 10.1016/S0140-6736(05)67696-9. [DOI] [PubMed] [Google Scholar]
- 11.Teaw S., Gupta A., Williams A., et al. Hyperspectral imaging in systemic sclerosis-associated Raynaud phenomenon. Arthritis Res Ther. 2023;25(1):10. doi: 10.1186/s13075-023-02990-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Grambow E., Sandkühler N.A., Groß J., et al. Evaluation of hyperspectral imaging for follow-up assessment after revascularization in peripheral artery disease. J Clin Med. 2022;11(3):758. doi: 10.3390/jcm11030758. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Jafari-Saraf L., Wilson S.E., Gordon I.L. Hyperspectral image measurements of skin hemoglobin compared with transcutaneous PO2 measurements. Ann Vasc Surg. 2012;26(4):537–548. doi: 10.1016/j.avsg.2011.12.002. [DOI] [PubMed] [Google Scholar]
- 14.Shenoy V.R., Kingston C.Q., Singh M., et al. Perfusion assessment of healthy and injured hands using video-based deep learning models. Plast Reconstr Surg. 2025;156(6):909–918. doi: 10.1097/PRS.0000000000012225. [DOI] [PubMed] [Google Scholar]
- 15.Horowitz M., Iskandarova A., Yohe G., Rahman S., Durr N., Giladi A.M. Inexperienced evaluator identification of hand ischemia via video processed with pigment-enhancing technology. J Hand Surg Glob Online. 2024;7(1):72–78. doi: 10.1016/j.jhsg.2024.11.002. [DOI] [PMC free article] [PubMed] [Google Scholar]