Novel imaging modalities for burn assessment

Overview¶

Burn depth drives the central early decision in burn care: which wounds will heal with dressings and which need excision and grafting. The problem is that the tool used to make that decision, the clinician's eye, is unreliable. Clinical assessment of burn depth is accurate in only 60-75% of cases even when performed by an experienced burn surgeon ^[1]. Independent series put expert depth accuracy at roughly 67% for partial-thickness burns ^[3] and tissue-viability judgment at 50-70% ^[4]. Clinical examination alone is often insufficient to determine which indeterminate wounds will heal spontaneously and which require surgery ^[5]. The cost of getting it wrong is real: misclassified deep wounds heal slowly and scar, while overcalled superficial wounds are excised and grafted unnecessarily.

That gap has motivated four decades of work on objective imaging. The modalities cluster by what they actually measure. Perfusion-based techniques (laser Doppler, laser speckle, ICG fluorescence, thermography as a perfusion surrogate) infer depth from blood flow, on the principle that deeper injury destroys the dermal microcirculation. Structural and compositional techniques (optical coherence tomography, hyperspectral and multispectral imaging, spatial frequency domain imaging, photoacoustic, terahertz, near-infrared, confocal) read collagen denaturation, scattering, oxygenation, water content, or microanatomy directly. A separate stream applies machine learning to ordinary digital images, and a parallel set of digital tools targets burn extent (TBSA) and remote triage rather than depth. Reviews of these optical modalities frame them consistently as adjuncts to, not replacements for, the experienced surgeon ^[5][11].

This page organizes the evidence by what each family of techniques can do in clinical practice and how well it has been validated, then turns to the adoption barriers that have kept nearly all of these tools in the research literature rather than on the ward.

The validated comparator: laser Doppler imaging¶

Laser Doppler imaging (LDI) is the reference against which every newer modality is measured, so it anchors the rest of this page even though it is not itself novel. Across reviews it is described as the only evidence-based adjunct to clinical evaluation of burn depth ^[7], and the only technique shown to accurately predict wound outcome with a large weight of evidence, having been approved for burn-depth assessment by regulatory bodies including the FDA ^[1]. A systematic review of measurement-property quality found strong evidence for the construct validity of LDI and concluded it is currently the most favorable technique for assessing burn wound healing potential ^[10]. Reported LDI accuracy reaches roughly 96% as it scans large burned areas non-contact ^[9], with the technique characterized as exceeding 95% accuracy for objective measurement of healing potential ^[8].

LDI sets the bar in two ways. First, it is the comparator: many studies of thermography, laser speckle, hyperspectral imaging, and AI report their performance relative to LDI rather than to histology. Second, its limitations define the opening for newer tools. LDI requires a relatively expensive dedicated scanner, takes time per field, and is most accurate only after 48 hours post-injury. Reviews note that older modality comparisons consistently rank LDI and ICG video angiography as offering the best data-supported accuracy estimates among techniques in clinical use ^[6]. The newer optical and AI methods are largely attempts to match LDI's accuracy at lower cost, faster, earlier, or in settings where a dedicated scanner is not available.

Optical and spectral techniques¶

Thermography¶

Infrared thermography measures surface temperature as a perfusion surrogate, exploiting the observation that deeper burns run cooler. It is fast, non-contact, and inexpensive, and the arrival of smartphone-attachable cameras (FLIR ONE) lowered the entry cost further ^[15]. In a porcine-validated and clinical work, infrared thermal imaging classified burn depth with 87.2% accuracy versus 54.1% for clinical assessment ^[12], and a diagnostic-accuracy meta-analysis reported pooled sensitivity 0.84 and specificity 0.76 ^[13]. A separate meta-analysis put overall thermography accuracy at 84.8% with 63% sensitivity and 81.9% specificity, characterizing it as moderately accurate ^[14]. Dynamic approaches that cool the wound and track rewarming have correctly predicted graft need in indeterminate extremity burns ^[16].

The central caveat is physical. Thermography is subject to the systematic bias of evaporative cooling, which is why it is judged unlikely to challenge LDI as the gold standard ^[7]. In a head-to-head study the FLIR ONE reached 66.7% sensitivity and 76.7% specificity for healing within 21 days versus LDI's 93.3% sensitivity ^[15], and evaporative cooling at the wound surface can lead to overprediction of healing times ^[15].

Optical coherence tomography¶

Optical coherence tomography (OCT) and its polarization-sensitive variant image tissue microstructure and collagen birefringence, which thermal injury reduces through denaturation. OCT is non-invasive and can be used to diagnose burn depth in real time ^[20]. Combined dual-imaging approaches pairing OCT with pulse speckle imaging classified burn type with an ROC-AUC of 0.87 as early as one hour post-burn ^[19]. A combined reflectance confocal microscopy/OCT instrument used OCT to visualize deeper injuries and quantify collagen destruction by skin birefringence while confocal provided submicron epidermal detail ^[22]. OCT's narrow field of view and depth penetration limit it more to focused structural readout than whole-wound mapping.

Hyperspectral, multispectral, and short-wave infrared imaging¶

Spectral imaging measures reflectance across many wavelengths to derive oxygenation, hemoglobin, and scattering parameters. Early proof-of-concept work characterized partial-thickness burns from a hyperspectral image analyzed with a linear spectral unmixing model, flagging the technique as promising but not yet an established clinical method ^[86]. A systematic review and meta-analysis of spectral imaging for burn depth reported pooled sensitivity 86% and specificity 84%, with machine-learning integration improving classification ^[33]. Clinical work on hand burns found hyperspectral imaging differentiated burns and could distinguish superficial from deep partial burns on near-infrared perfusion features ^[34]; a parallel clinical series, however, found no reliable distinction between superficial and deep partial-thickness burns ^[35], underscoring the variability in this literature. Short-wave infrared multispectral imaging is an emerging line that distinguishes superficial from deep burns in animal and early human work ^[36]. A multispectral device with a CNN reached an AUC of 0.95, with accuracy lowest at 1-2 days and highest at 3-4 days, identifying time-since-injury as a significant covariate ^[37].

Spatial frequency domain imaging¶

Spatial frequency domain imaging (SFDI) maps tissue absorption and scattering across a wide field. In a porcine model, SFDI scattering parameters showed a strong negative correlation with histological burn depth (r-squared above 0.89) ^[30]. Combining SFDI reflectance with a support-vector-machine classifier predicted burn severity at 24 hours with 92.5% accuracy ^[31]. In a swine comparison against histology, SFDI reached 85% diagnostic accuracy, edging clinical analysis (83%) and outperforming laser speckle (75%) and thermography (73%) at 24 hours ^[32].

Laser speckle contrast imaging¶

Laser speckle contrast imaging (LSCI) measures microvascular perfusion across a wide field instantly and without contact, addressing some of LDI's speed limitations. Perfusion in the first week separates burns that heal within 14 days from those that do not ^[38], and a validity study reported good performance for predicting healing potential and characterized LSCI as highly feasible and patient-friendly ^[41]. In pediatric scalds, LSCI sensitivity and specificity reached 100% when combining early and 72-96 hour measurements into a perfusion trend ^[39], and adding blood-flow pulsatility raised day 0-2 prediction of surgical need to 100% sensitivity and specificity in one cohort ^[40]. Interobserver reliability is a strength: observers identify the same region of interest and produce observer-independent perfusion values irrespective of burn experience ^[42]. A critical examination tempers this, reporting low positive predictive value and concluding LSCI can overestimate severity because it fails to detect deep dermal blood flow ^[8].

Photoacoustic, near-infrared, and confocal imaging¶

Photoacoustic imaging (PAI) combines optical contrast with ultrasound resolution to image depth-resolved hemoglobin and the zone of stasis. Burn depths from PAI correlated strongly with histologically determined injury depth, with smaller measurement errors than LDI ^[44] and a maximum depth error of 140 micrometers in a real-time system ^[45]. Combined PAI/OCT has been proposed as a multi-parameter approach with clinical diagnostic potential ^[46]. Near-infrared spectroscopy reads oxygenation, total hemoglobin, and tissue water; it distinguishes superficial from full-thickness injuries ^[47] and has detected methemoglobin as an early viability biomarker highest in non-viable tissue by 24 hours ^[4]. In-vivo confocal-laser-scanning microscopy differentiates superficial-partial from deep-partial thickness burns at the histomorphological level ^[48], but its microscopic field limits whole-wound use.

Terahertz imaging¶

Terahertz time-domain spectroscopy is sensitive to tissue water and structure and has emerged as an early-triage tool. Using day-0 terahertz waveforms, one system predicted day-28 wound-healing outcome with 94.7% accuracy ^[49], and a portable handheld terahertz scanner with a deep neural network differentiated partial-, deep partial-, and full-thickness burns one hour post-injury (ROC-AUC 91%, 88%, 86%) regardless of etiology ^[50]. Terahertz spectral imaging has also been extended to early frostbite-depth assessment, with ROC-AUCs of 0.94, 0.85, and 0.87 across healthy tissue, partial-thickness, and full-thickness frostbite ^[51].

AI and digital methods¶

A distinct stream applies machine learning to ordinary images rather than to specialized optics, which makes it attractive for low-cost, smartphone-based deployment. Convolutional neural networks segment wounds and classify depth with high reported metrics: a Mask R-CNN model deviated less from ground-truth TBSA than burn surgeons did on average ^[52], and a smartphone-image joint-task model achieved an R-squared of 0.9136 for TBSA estimation ^[53]. A multispectral-imaging device paired with deep learning improved surgeons' intra-operative viability judgment by 25% ^[56], and an intra-operative AI ensemble reached 81% sensitivity and 100% specificity for burn classification ^[55]. Systematic reviews report burn-depth classification accuracies above 83% ^[57] and across studies generally 68.9-95.4% ^[58], typically 70-90% relative to clinical judgment as the comparator ^[59].

The honest reading of the AI literature is that performance numbers run ahead of evidence quality. A general-purpose multimodal model was non-inferior to emergency physicians for region-level TBSA estimation but performed substantially worse for depth classification ^[54]. A PRISMA diagnostic-test-accuracy review concluded the evidence base is insufficient for deployment-ready use of AI burn-depth assessment ^[60], and a scoping review found that high reported performance does not necessarily imply clinical robustness or real-world accuracy ^[61]. Digital photography occupies a related niche: burn size can be assessed reliably and validly by experts from photographs, but burn depth cannot, with depth inter-rater reliability as low as 0.38 for experts ^[62]. Overall photographic diagnostic accuracy is roughly 67.5% for size and 66.0% for depth ^[63].

Burn extent (TBSA) and digital planimetry¶

A parallel set of digital tools targets extent rather than depth, where the failure mode is overestimation: physicians overestimate body surface area by 20-50% ^[75]. Smartphone TBSA apps and 3D photography reduce that error. The Mersey Burns app produced quicker and more accurate calculations than Lund-Browder charts ^[71]; EasyTBSA had the greatest accuracy among methods tested (-0.01% mean error) ^[72]; and a unit-level study tied a smartphone app to a significant improvement in referral-hospital TBSA agreement ^[73]. A systematic review found digital tools demonstrated superior accuracy and higher inter-rater reliability (ICC 0.986-0.998) than traditional methods ^[74]. 3D photography and scanning are valid and reliable for wound area: 3D photography matched digital planimetry with an ICC of 0.994 ^[76], and dedicated 3D systems (BurnCalc, Artec scanner) produced stable, reliable area measurements ^[77][78].

Telemedicine and remote triage¶

Telemedicine extends expert assessment to referring sites and is the most clinically mature digital application. Video-enhanced telemedicine improved burn-size estimation and changed management, including more accurate fluid resuscitation ^[64], and smartphone-based referral changed the referral pathway, avoiding or appropriately delaying admission in 66% of patients ^[65]. Smartphone telemedicine diagnosed the need for surgery accurately in 94.4% of cases ^[66], and telemedicine matched in-person examination almost perfectly for clinical decision and TBSA ^[67]. A systematic review found low-to-moderate-level evidence supporting virtual burn care's cost-effectiveness and its ability to improve assessment and triage ^[68], with another review reporting better triage and more accurate TBSA estimation by telehealth ^[69]. The consistent weak spot mirrors photography: remote burn-depth assessment is low-accuracy, with aggregated specialist depth ICCs around 0.53 ^[70].

Diagnostic accuracy and validation¶

Read across modalities, the validation picture has a recurring shape. Pooled diagnostic accuracy for the better-studied techniques clusters in the mid-80% range: thermography at pooled sensitivity 0.84 and specificity 0.76 ^[13], spectral imaging at 86% and 84% ^[33], SFDI at 85% against histology ^[32]. These figures beat clinical assessment's 60-75% baseline ^[1] but rarely match LDI's reported 95% ^[8]. Validation is complicated by the comparator problem: studies variously use histology, healing outcome at 14 or 21 days, LDI, or surgeon judgment as the reference standard, which makes cross-modality comparison unreliable. A spectral-imaging review explicitly flagged significant variability in methodologies and a lack of standardized ground-truthing ^[33].

The AI literature sharpens this concern. A diagnostic-test-accuracy meta-analysis judged the evidence base insufficient for deployment-ready AI burn-depth assessment ^[60], and a scoping review warned that high reported performance does not necessarily imply real-world accuracy ^[61]. Timing also matters: multispectral CNN accuracy was lowest at 1-2 days and highest at 3-4 days, with time-since-injury a significant covariate ^[37]. The most rigorous recent comparator work uses non-inferiority designs, where a multimodal model was non-inferior to physicians for TBSA but worse for depth ^[54].

Special considerations¶

Several modalities have specific roles in populations where clinical assessment is hardest. In pediatric scalds, laser speckle contrast imaging is well validated, with lower perfusion values associated with longer healing and care periods ^[43] and very high sensitivity and specificity from combined early and 72-96 hour measurements ^[39]. Thermography has been studied as a portable pediatric adjunct, comparable to LDI for diagnosing superficial partial-thickness burns in one cohort and useful where LDI is unavailable ^[81], and thermal imaging has contributed to depth determination in pediatric acute burns ^[17]. AI applied to polarized-light photography has also been used to improve depth assessment of pediatric scalds ^[21]. Point-of-care fluorescence imaging (MolecuLight) has been integrated into a pediatric burn program for infection detection ^[28].

Beyond thermal burns, terahertz spectral imaging has been applied to early frostbite-depth assessment ^[51], CNN models have graded skin frostbite ^[79], and ICG microangiography has monitored frostbite progression during hyperbaric oxygen therapy to guide treatment planning ^[80]. For low-resource and remote settings, telemedicine and smartphone tools are the practical answer, extending expert triage where dedicated optical scanners are not available ^[69]. A UK review noted that NICE recognized both LDI and MolecuLight as valuable tools with potential to improve outcomes and reduce costs ^[82].

A separate use of fluorescence imaging targets infection rather than depth. Bacterial autofluorescence imaging detected graft-site bacterial load with 86% sensitivity and 98% specificity for predicting graft loss ^[27], improved infection detection when added to standard assessment ^[28], and across a scoping review improved bacterial-detection accuracy versus clinical assessment alone ^[29]. ICG fluorescence angiography itself differentiates burns that will heal from those that will not ^[23] and can guide intra-operative necrosis excision ^[25], with one systematic review reporting 100% accuracy versus 50% for clinical assessment in indeterminate wounds ^[24] while another found its accuracy for distinguishing superficial from deep partial-thickness burns too limited for routine depth use ^[26].

Controversies and Evidence Gaps¶

The defining controversy is the gap between published performance and clinical adoption. Despite dozens of modalities and decades of work, none has achieved broad routine use, and reviews are candid that LDI remains the only technique with a large evidence base and regulatory clearance ^[1][9]. An implementation study of thermal imaging found the barriers are organizational as much as technical: physicians' attitudes and perceived value, low compatibility with existing workflow, and limited knowledge of the evidence ^[18]. Cost, the need for dedicated hardware, and the absence of reimbursement compound the problem.

The validation literature is thin and heterogeneous. Most studies are small, retrospective, low-level descriptive or observational designs ^[68][87], and a machine-learning review calls for systematic analysis of how input modalities, training sets, and classifiers affect reported accuracy ^[59]. A diagnostic-test-accuracy meta-analysis concluded the AI evidence base is not deployment-ready ^[60], and a scoping review cautioned that high reported accuracy may not translate to real-world performance ^[61]. There is no standardized ground-truthing approach across spectral-imaging studies ^[33].

Skin-tone fairness is an under-studied gap with direct equity implications. Epidermal melanin confounds spectral measurement: it strongly absorbs at visible wavelengths and biases the scattering and absorption coefficients derived by spatial frequency domain imaging in pigmented skin ^[83], and high intersubject variation in measured scattering at visible wavelengths coincides with large melanin extinction coefficients ^[84]. On the AI side, narrative and scoping reviews flag algorithmic bias across skin tones and limited dataset diversity as persistent barriers to safe, equitable deployment ^[85]. Pediatric-specific datasets and external validation against accepted comparators across skin tones and ages remain limited ^[88][61].

Finally, the comparator question itself is unresolved. The literature has not converged on whether the reference standard for validating a new modality should be histology, healing outcome, LDI, or expert consensus, and the choice materially changes reported accuracy. Until that standardizes, and until larger multi-institutional studies with external validation appear, these modalities will remain adjuncts that inform rather than replace the burn clinician's judgment ^[54][57].

References¶

[1] Monstrey S, Hoeksema H, Verbelen J, Pirayesh A, Blondeel P. "Assessment of burn depth and burn wound healing potential." Burns : journal of the International Society for Burn Injuries 2008. PMID: 18511202 ↩

[2] Jaber GJ, Caddia G, Santoro C, Morelli Coppola M, Al Sammour C, di Summa PG. "Assessing Burn Depth by Indocyanine Green Angiography: A Systematic Review and Algorithm for Clinical Use." Annals of burns and fire disasters 2026. PMID: 41782773

[3] Lee JJ, et al. "Comparing Artificial Intelligence Guided Image Assessment to Current Methods of Burn Assessment." Journal of burn care & research : official publication of the American Burn Association 2025. PMID: 38918900 ↩

[4] Leung G, Duta D, Perry J, Leonardi L, Fish J, Cross K. "Rapid tissue viability evaluation using methemoglobin as a biomarker in burns." International journal of burns and trauma 2018. PMID: 30515351 ↩

[5] Kaiser M, Yafi A, Cinat M, Choi B, Durkin AJ. "Noninvasive assessment of burn wound severity using optical technology: a review of current and future modalities." Burns : journal of the International Society for Burn Injuries 2011. PMID: 21185123 ↩

[6] Devgan L, Bhat S, Aylward S, Spence RJ. "Modalities for the assessment of burn wound depth." Journal of burns and wounds 2006. PMID: 16921415 ↩

[7] Burke-Smith A, Collier J, Jones I. "A comparison of non-invasive imaging modalities: Infrared thermography, spectrophotometric intracutaneous analysis and laser Doppler imaging for the assessment of adult burns." Burns : journal of the International Society for Burn Injuries 2015. PMID: 26421694 ↩

[8] De Mey K, et al. "Validity of laser speckle contrast imaging for predicting wound healing potential in burns: A critical examination." Burns : journal of the International Society for Burn Injuries 2025. PMID: 40112657 ↩

[9] Claes KEY, et al. "Evidence Based Burn Depth Assessment Using Laser-Based Technologies: Where Do We Stand?." Journal of burn care & research : official publication of the American Burn Association 2021. PMID: 33128377 ↩

[10] Jaspers MEH, van Haasterecht L, van Zuijlen PPM, Mokkink LB. "A systematic review on the quality of measurement techniques for the assessment of burn wound depth or healing potential." Burns : journal of the International Society for Burn Injuries 2019. PMID: 29941159 ↩

[11] Allen J, Howell K. "Microvascular imaging: techniques and opportunities for clinical physiological measurements." Physiological measurement 2014. PMID: 24910968 ↩

[12] Singer AJ, Relan P, Beto L, Jones-Koliski L, Sandoval S, Clark RA. "Infrared Thermal Imaging Has the Potential to Reduce Unnecessary Surgery and Delays to Necessary Surgery in Burn Patients." Journal of burn care & research : official publication of the American Burn Association 2016. PMID: 26720102 ↩

[13] Asif A, Poyiatzis C, Raheman FJ, Rojoa DM. "The Use of Infrared Thermography (IRT) in Burns Depth Assessment: A Diagnostic Accuracy Meta-Analysis." European burn journal 2022. PMID: 39599957 ↩

[14] Pedrosa R, et al. "Accuracy of infrared thermography evaluation in burn wound healing: a systematic review and meta-analysis." Journal of wound care 2024. PMID: 38588060 ↩

[15] Goel J, et al. "A prospective study comparing the FLIR ONE with laser Doppler imaging in the assessment of burn depth by a tertiary burns unit in the United Kingdom." Scars, burns & healing 2020. PMID: 33489315 ↩

[16] Simmons JD, et al. "Early Assessment of Burn Depth with Far Infrared Time-Lapse Thermography." Journal of the American College of Surgeons 2018. PMID: 29409904 ↩

[17] Ganon S, Guédon A, Cassier S, Atlan M. "Contribution of thermal imaging in determining the depth of pediatric acute burns." Burns : journal of the International Society for Burn Injuries 2020. PMID: 31864785 ↩

[18] de Haan J, Stoop M, van Zuijlen PPM, Pijpe A. "Thermal Imaging for Burn Wound Depth Assessment: A Mixed-Methods Implementation Study." Journal of clinical medicine 2024. PMID: 38610828 ↩

[19] Ganapathy P, et al. "Dual-imaging system for burn depth diagnosis." Burns : journal of the International Society for Burn Injuries 2014. PMID: 23790396 ↩

[20] Liu SH, Xie WG, Kremer M, Machens HG, Lankenau EM, Huettmann G. "[Diagnosis of the deep partial-thickness burn wound of Skh-1 mouse with Optical Coherence Tomography]." Zhonghua shao shang za zhi = Zhonghua shaoshang zazhi = Chinese journal of burns 2010. PMID: 21029684 ↩

[21] Cirillo MD, Mirdell R, Sjöberg F, Pham TD. "Improving burn depth assessment for pediatric scalds by AI based on semantic segmentation of polarized light photography images." Burns : journal of the International Society for Burn Injuries 2021. PMID: 33947595 ↩

[22] Iftimia N, et al. "Combined reflectance confocal microscopy/optical coherence tomography imaging for skin burn assessment." Biomedical optics express 2013. PMID: 23667785 ↩

[23] Schomacker KT, Torri A, Sandison DR, Sheridan RL, Nishioka NS. "Biodistribution of indocyanine green in a porcine burn model: light and fluorescence microscopy." The Journal of trauma 1997. PMID: 9390494 ↩

[24] Wongkietkachorn A, et al. "Indocyanine green dye angiography as an adjunct to assess indeterminate burn wounds: A prospective, multicentered, triple-blinded study." The journal of trauma and acute care surgery 2019. PMID: 30589753 ↩

[25] Zajac JC, et al. "Lighting the Way for Necrosis Excision Through Indocyanine Green Fluorescence-Guided Surgery." Journal of the American College of Surgeons 2022. PMID: 36102554 ↩

[26] Hilgarth MK, et al. "The Diagnostic Value of Indocyanine Green in the Assessment of Depth of Burn Injuries: A Systematic Review." European burn journal 2026. PMID: 41892159 ↩

[27] Hanson-Viana E, Rojas-Ortiz JA, Rendón-Medina MA, Luna-Zepeda BL. "Bacterial fluorescence imaging as a predictor of skin graft integration in burn wounds." Burns : journal of the International Society for Burn Injuries 2024. PMID: 38735804 ↩

[28] Turner E, Kelly C, Zuccaro J, Chakera H, Gus E, Fish JS. "Assessing Pediatric Burn Wound Infection Using a Point-of-Care Fluorescence Imaging Device." Journal of burn care & research : official publication of the American Burn Association 2024. PMID: 38833179 ↩

[29] Jeffery SLA, Hanson-Viana E. "A Scoping Review of Fluorescence Imaging: A Promising New Technology for Bacterial Detection in Burn Wounds." Journal of burn care & research : official publication of the American Burn Association 2026. PMID: 40971539 ↩

[30] Mazhar A, Saggese S, Pollins AC, Cardwell NL, Nanney L, Cuccia DJ. "Noncontact imaging of burn depth and extent in a porcine model using spatial frequency domain imaging." Journal of biomedical optics 2014. PMID: 25147961 ↩

[31] Rowland R, et al. "Burn wound classification model using spatial frequency-domain imaging and machine learning." Journal of biomedical optics 2019. PMID: 31134769 ↩

[32] Ponticorvo A, et al. "Evaluating clinical observation versus Spatial Frequency Domain Imaging (SFDI), Laser Speckle Imaging (LSI) and thermal imaging for the assessment of burn depth." Burns : journal of the International Society for Burn Injuries 2019. PMID: 30327232 ↩

[33] Tan P, Ravulapalli K, Lewis CJ. "A systematic review of advances in the use of spectral imaging in burn depth assessment." Burns : journal of the International Society for Burn Injuries 2025. PMID: 39933419 ↩

[34] Promny D, Aich J, Püski T, Marti Edo A, Reichert B, Billner M. "Evaluation of hyperspectral imaging as a modern aid in clinical assessment of burn wounds of the upper extremity." Burns : journal of the International Society for Burn Injuries 2022. PMID: 34857418 ↩

[35] Promny D, Aich J, Billner M, Reichert B. "First Preliminary Clinical Experiences Using Hyperspectral Imaging for Burn Depth Assessment of Hand Burns." Journal of burn care & research : official publication of the American Burn Association 2022. PMID: 34015099 ↩

[36] Nunez J, et al. "Novel multi-spectral short-wave infrared imaging for assessment of human burn wound depth." Wound repair and regeneration : official publication of the Wound Healing Society [and] the European Tissue Repair Society 2024. PMID: 39323286 ↩

[37] Carter JE, et al. "Refinement of an Artificial Intelligence Algorithm for Enhanced Burn Wound Depth Assessment Using Multispectral Imaging: An Expanded Proof of Concept Study." Journal of burn care & research : official publication of the American Burn Association 2025. PMID: 40452490 ↩

[38] Lindahl F, Tesselaar E, Sjöberg F. "Assessing paediatric scald injuries using Laser Speckle Contrast Imaging." Burns : journal of the International Society for Burn Injuries 2013. PMID: 23092702 ↩

[39] Mirdell R, Farnebo S, Sjöberg F, Tesselaar E. "Accuracy of laser speckle contrast imaging in the assessment of pediatric scald wounds." Burns : journal of the International Society for Burn Injuries 2018. PMID: 28797578 ↩

[40] Mirdell R, Farnebo S, Sjöberg F, Tesselaar E. "Using blood flow pulsatility to improve the accuracy of laser speckle contrast imaging in the assessment of burns." Burns : journal of the International Society for Burn Injuries 2020. PMID: 32299641 ↩

[41] Zheng KJ, Middelkoop E, Stoop M, van Zuijlen PPM, Pijpe A. "Validity of laser speckle contrast imaging for the prediction of burn wound healing potential." Burns : journal of the International Society for Burn Injuries 2022. PMID: 34006410 ↩

[42] Mirdell R, Farnebo S, Sjöberg F, Tesselaar E. "Interobserver reliability of laser speckle contrast imaging in the assessment of burns." Burns : journal of the International Society for Burn Injuries 2019. PMID: 31230800 ↩

[43] Elmasry M, Mirdell R, Tesselaar E, Farnebo S, Sjöberg F, Steinvall I. "Laser speckle contrast imaging in children with scalds: Its influence on timing of intervention, duration of healing and care, and costs." Burns : journal of the International Society for Burn Injuries 2019. PMID: 30827850 ↩

[44] Ida T, et al. "Burn depth assessments by photoacoustic imaging and laser Doppler imaging." Wound repair and regeneration : official publication of the Wound Healing Society [and] the European Tissue Repair Society 2016. PMID: 26487320 ↩

[45] Ida T, et al. "Real-time photoacoustic imaging system for burn diagnosis." Journal of biomedical optics 2014. PMID: 25127338 ↩

[46] Liu K, Chen Z, Zhou W, Xing D. "Towards quantitative assessment of burn based on photoacoustic and optical coherence tomography." Journal of biophotonics 2020. PMID: 32609427 ↩

[47] Cross KM, et al. "Clinical utilization of near-infrared spectroscopy devices for burn depth assessment." Wound repair and regeneration : official publication of the Wound Healing Society [and] the European Tissue Repair Society 2007. PMID: 17537120 ↩

[48] Altintas MA, Altintas AA, Knobloch K, Guggenheim M, Zweifel CJ, Vogt PM. "Differentiation of superficial-partial vs. deep-partial thickness burn injuries in vivo by confocal-laser-scanning microscopy." Burns : journal of the International Society for Burn Injuries 2009. PMID: 18691820 ↩

[49] Khani ME, et al. "Accurate and early prediction of the wound healing outcome of burn injuries using the wavelet Shannon entropy of terahertz time-domain waveforms." Journal of biomedical optics 2022. PMID: 36348509 ↩

[50] Osman OB, et al. "Deep neural network classification of in vivo burn injuries with different etiologies using terahertz time-domain spectral imaging." Biomedical optics express 2022. PMID: 35519269 ↩

[51] Heller E, et al. "Terahertz spectral imaging for early assessment of frostbite injuries using the double Debye model and supervised machine learning." Biomedical optics express 2026. PMID: 42110979 ↩

[52] Chang CW, et al. "Deep Learning-Assisted Burn Wound Diagnosis: Diagnostic Model Development Study." JMIR medical informatics 2021. PMID: 34860674 ↩

[53] Xu X, Bu Q, Xie J, Li H, Xu F, Li J. "On-site burn severity assessment using smartphone-captured color burn wound images." Computers in biology and medicine 2024. PMID: 39362001 ↩

[54] Aykut A, Karayıl AR, Yıldırım C, Günsoy E, Tatlı M, Avcı M. "Multimodal large language model versus emergency physicians for burn assessment: a prospective non-inferiority study." Scandinavian journal of trauma, resuscitation and emergency medicine 2026. PMID: 41645231 ↩

[55] Thatcher JE, et al. "Clinical Investigation of a Rapid Non-invasive Multispectral Imaging Device Utilizing an Artificial Intelligence Algorithm for Improved Burn Assessment." Journal of burn care & research : official publication of the American Burn Association 2023. PMID: 37082889 ↩

[56] Yu S, et al. "Feasibility of intra-operative image guidance in burn excision surgery with multispectral imaging and deep learning." Burns : journal of the International Society for Burn Injuries 2024. PMID: 37821282 ↩

[57] Huang S, Dang J, Sheckter CC, Yenikomshian HA, Gillenwater J. "A systematic review of machine learning and automation in burn wound evaluation: A promising but developing frontier." Burns : journal of the International Society for Burn Injuries 2021. PMID: 34419331 ↩

[58] Taib BG, Karwath A, Wensley K, Minku L, Gkoutos GV, Moiemen N. "Artificial intelligence in the management and treatment of burns: A systematic review and meta-analyses." Journal of plastic, reconstructive & aesthetic surgery : JPRAS 2023. PMID: 36571960 ↩

[59] Wilson RH, et al. "Review of machine learning for optical imaging of burn wound severity assessment." Journal of biomedical optics 2024. PMID: 38361506 ↩

[60] Wang Y, Jiang Y, Zhang Y, Liu S. "Diagnostic accuracy and deployment readiness of AI for burn depth assessment across imaging modalities: a PRISMA-DTA systematic review and meta-analysis." International journal of medical informatics 2026. PMID: 42025001 ↩

[61] Holm S, Huss F, Nayyer B, Zdolsek J. "Use of Artificial Intelligence in Burn Assessment: A Scoping Review with a Large Language Model-Generated Decision Tree." European burn journal 2026. PMID: 41562702 ↩

[62] Hop MJ, et al. "Photographic assessment of burn size and depth: reliability and validity." Journal of wound care 2014. PMID: 24633060 ↩

[63] Boissin C, Laflamme L, Wallis L, Fleming J, Hasselberg M. "Photograph-based diagnosis of burns in patients with dark-skin types: the importance of case and assessor characteristics." Burns : journal of the International Society for Burn Injuries 2015. PMID: 25716764 ↩

[64] Wibbenmeyer L, et al. "Video-Enhanced Telemedicine Improves the Care of Acutely Injured Burn Patients in a Rural State." Journal of burn care & research : official publication of the American Burn Association 2016. PMID: 26132049 ↩

[65] den Hollander D, Mars M. "Smart phones make smart referrals: The use of mobile phone technology in burn care - A retrospective case series." Burns : journal of the International Society for Burn Injuries 2017. PMID: 27575675 ↩

[66] Boccara D, et al. "Ongoing Development and Evaluation of a Method of Telemedicine: Burn Care Management With a Smartphone." Journal of burn care & research : official publication of the American Burn Association 2018. PMID: 29789857 ↩

[67] Esberk T, Das K. "Telemedicine for the management of pediatric burn patients admitted to emergency services." Journal of telemedicine and telecare 2023. PMID: 34259585 ↩

[68] Mondor E, Barnabe J, Laguan EMR, Malic C. "Virtual burn care - Friend or foe? A systematic review." Burns : journal of the International Society for Burn Injuries 2024. PMID: 38490837 ↩

[69] García-Díaz A, Vilardell-Roig L, Novillo-Ortiz D, Gacto-Sánchez P, Pereyra-Rodríguez JJ, Saigí-Rubió F. "Utility of Telehealth Platforms Applied to Burns Management: A Systematic Review." International journal of environmental research and public health 2023. PMID: 36833860 ↩

[70] Blom L, Boissin C, Allorto N, Wallis L, Hasselberg M, Laflamme L. "Accuracy of acute burns diagnosis made using smartphones and tablets: a questionnaire-based study among medical experts." BMC emergency medicine 2017. PMID: 29237400 ↩

[71] Barnes J, et al. "The Mersey Burns App: evolving a model of validation." Emergency medicine journal : EMJ 2015. PMID: 25371408 ↩

[72] Colson CD, et al. "EasyTBSA as a method for calculating total body surface area burned: a validation study." Emergency medicine journal : EMJ 2023. PMID: 36639224 ↩

[73] Sritharan T, et al. "Temporal trends in burn size estimation and the impact of the NSW Trauma App on estimation accuracy." Burns : journal of the International Society for Burn Injuries 2023. PMID: 36878736 ↩

[74] Tiwari S, Pradhan A, Koh E, Solanki NS, Wagstaff MJD, Concannon E. "Comparative accuracy and inter-rater reliability of digital tools for burn surface area estimation: A systematic review." Journal of plastic, reconstructive & aesthetic surgery : JPRAS 2026. PMID: 41966915 ↩

[75] Dirnberger J, Giretzlehner M, Ruhmer M, Haller H, Rodemund C. "Modelling human burn injuries in a three-dimensional virtual environment." Studies in health technology and informatics 2003. PMID: 15455863 ↩

[76] Stockton KA, McMillan CM, Storey KJ, David MC, Kimble RM. "3D photography is as accurate as digital planimetry tracing in determining burn wound area." Burns : journal of the International Society for Burn Injuries 2015. PMID: 24877886 ↩

[77] Sheng WB, Zeng D, Wan Y, Yao L, Tang HT, Xia ZF. "BurnCalc assessment study of computer-aided individual three-dimensional burn area calculation." Journal of translational medicine 2014. PMID: 25204349 ↩

[78] Rashaan ZM, et al. "Three-dimensional imaging: a novel, valid, and reliable technique for measuring wound surface area." Skin research and technology : official journal of International Society for Bioengineering and the Skin (ISBS) [and] International Society for Digital Imaging of Skin (ISDIS) [and] International Society for Skin Imaging (ISSI) 2016. PMID: 26853829 ↩

[79] Sun J, et al. "Convolutional neural network models for automatic diagnosis and graduation in skin frostbite." International wound journal 2023. PMID: 36054618 ↩

[80] Masters T, et al. "Microangiography to Monitor Treatment Outcomes Following Severe Frostbite Injury to the Hands." Journal of burn care & research : official publication of the American Burn Association 2018. PMID: 28328661 ↩

[81] van de Warenburg MS, Wieringa M, Ulrich DJO, Hummelink S, Vehmeijer-Heeman MLAW. "The use of the FLIR ONE Pro thermal camera for paediatric burn wound depth assessment: A pilot study." Burns : journal of the International Society for Burn Injuries 2026. PMID: 41780415 ↩

[82] Farhan N, Hassan Z, Ali MAM, Alqalaf Z, Rasul RE, Jeffery S. "How Can Technology Improve Burn Wound Care: A Review of Wound Imaging Technologies and Their Application in Burns-UK Experience." Diagnostics (Basel, Switzerland) 2025. PMID: 40941764 ↩

[83] Phan T, et al. "Quantifying the confounding effect of pigmentation on measured skin tissue optical properties: a comparison of colorimetry with spatial frequency domain imaging." Journal of biomedical optics 2022. PMID: 35324096 ↩

[84] Phan T, et al. "Characterizing reduced scattering coefficient of normal human skin across different anatomic locations and Fitzpatrick skin types using spatial frequency domain imaging." Journal of biomedical optics 2021. PMID: 33569936 ↩

[85] Khorsandi J, et al. "Artificial Intelligence in Burn and Wound Care: Image Analysis, Prediction, and Clinical Integration." Journal of burn care & research : official publication of the American Burn Association 2026. PMID: 42251750 ↩

[86] Calin MA, Parasca SV, Savastru R, Manea D. "Characterization of burns using hyperspectral imaging technique - a preliminary study." Burns : journal of the International Society for Burn Injuries 2015. PMID: 24997530 ↩

[87] Rahman S, et al. "A Comprehensive Scoping Review on the Use of Point-Of-Care Infrared Thermography Devices for Assessing Various Wound Types." International wound journal 2025. PMID: 40819851 ↩

[88] Shabbir R, et al. "Artificial Intelligence in the Management and Treatment of Pediatric Burns: A Narrative Review of Current Applications and Future Directions." Cureus 2025. PMID: 41477388 ↩