Forensic bruise documentation with alternate light sources: Preliminary observation on deceased individuals in a routine illuminated autopsy setting Authors: Sif Korsbæk Topsøe‐Jensen, Carl Johan Wingren, Chiara Villa Published: 2026 ABSTRACT AbstractAlternate light sources (ALS) have been suggested to improve bruise visibility for medicolegal evaluations. Previous studies have recommended darkened conditions; however, this does not reflect real‐life autopsy settings, where multiple examinations are conducted simultaneously and turning off the lights is neither practical nor safe. This study evaluated the use of ALS for documenting bruises under normal lighting conditions without altering the autopsy setup. We photographed 161 bruises (99 recent, 62 non‐recent) using CrimeLite® AUTO (Foster+Freeman) under common white light (CWL), violet ALS with 415 nm long‐pass (VLP), violet ALS with 415 nm band‐pass (VBP), and blue ALS with 445 nm long‐pass (BLP). Bruises were classified as recent or non‐recent based on the presence of yellow coloration. Visibility was scored using a “modified” Bruise Visibility Scale (BVS), and surface area was measured. Observer variability was assessed using Cohen's kappa (κ) and intraclass correlation coefficients (ICC); logistic and Cox regression analyses were performed to evaluate BVS and size across lighting sources. We found that ALS did not enhance bruise visibility compared to CWL. VBP and BLP reduced visibility, while VLP showed occasional improvements but without statistical significance. Surface area measurements indicated that VLP visualizes bruises with larger dimensions than under CWL, whereas BLP and VBP did not. Observer agreement for the four BVS categories yielded κ values ranging from moderate to substantial, while ICCs showed excellent agreement for surface‐area measurements for most light sources. Despite its limitations, the study suggests that ALS, particularly VLP, may have potential utility for bruise documentation under normal lighting conditions.Keywords: alternate light, autopsy, bruise, Bruise Visibility Scale, filters, forensic pathology, illumination, surface area Highlights ALS was evaluated for bruise visualization under standard autopsy lighting conditions. ALS was evaluated for bruise visualization under standard autopsy lighting conditions. Violet ALS with a long‐pass filter visualized bruises as larger compared to common white light but not significant. Violet ALS with a long‐pass filter visualized bruises as larger compared to common white light but not significant. Violet ALS with a bandpass filter was associated with reduced visualization and smaller measured surface area. Violet ALS with a bandpass filter was associated with reduced visualization and smaller measured surface area. Violet ALS with a long‐pass filter resulted in a statistically significant greater bruise‐measured surface area. Violet ALS with a long‐pass filter resulted in a statistically significant greater bruise‐measured surface area. INTRODUCTION Bruises are key indicators of blunt force trauma and can provide crucial forensic evidence in cases of violent abuse and assault. Accurate bruise documentation is of significant medicolegal importance; however, current methods are limited to visual evaluation, written descriptions, and photographs. These approaches may be influenced by lighting conditions and display settings, especially when bruises are assessed using photographs. Such factors can affect both the accuracy of color documentation and individual perception of color–particularly yellow [1, 2]. Researchers have explored the use of an alternate light source (ALS) to improve bruise visualization and potentially reduce limitations associated with the visualization and documentation of bruises [2, 3, 4, 5, 6, 7, 8, 9]. ALS is widely used in forensic investigations to detect biological fluids (e.g., blood, semen) and chemical substances (e.g., paint, fire accelerants) [10, 11]. In practice, ALS involves directing light at specific wavelengths while using filters (e.g., specialized goggles or camera filters/lenses) to enhance contrast [5, 7]. At certain wavelengths, ALS can improve the visibility of substances that are otherwise difficult to detect under common white light (CWL). This principle has also been applied to bruise visualization, where ALS may make bruises more discernible in size and shape and, in some cases, reveal bruises not visible under CWL [5, 12]. ALS operates across the ultraviolet (<400 nm), visible (400–700 nm), and infrared (>700 nm) ranges [5, 7]. When light interacts with skin, it can be absorbed, reflected, scattered, transmitted, or re‐emitted as fluorescence [5, 13, 14]. Hemoglobin has a peak absorption at 415 nm, and bilirubin at 460 nm [6]. Theoretically, using ALS near 415 nm may enhance visualization of hemoglobin, while wavelengths around 450–460 nm have been proposed to highlight bilirubin, although supporting evidence remains limited [6]. Overall, ALS is more effective compared to CWL in detecting bruises, as shown by a recent systematic review [5]; the wavelengths around 415 nm, combined with long‐pass or bandpass yellow filters, seem to improve bruise visibility across skin tones. However, current research on ALS and bruise visibility has several limitations. Most studies lack blinding of bruise assessors, raising concerns about confirmation bias when evaluating the effects of ALS on bruise visibility. Much of the research has also been conducted by the same group of researchers, which may influence findings. Additionally, research has largely focused on bruises located on extremities, examined a limited range of injury mechanisms, and left unresolved questions about specificity—namely, whether marks revealed by ALS represent true bruises. Interpretation of ALS‐detected bruises remains understudied, and risks of overinterpreting enhanced imagery highlight the need for careful methodological consideration. Furthermore, previous studies have recommended performing ALS assessments under darkened conditions; however, such settings are rarely feasible in routine autopsies. Autopsy rooms often accommodate multiple examinations simultaneously, and turning off the lights would compromise visibility for other procedures, increase the risk of accidents with sharp instruments, and hinder compliance with safety protocols. Therefore, maintaining normal lighting is essential for both workflow efficiency and occupational safety. Various approaches have been used to interpret bruises and evaluate the effectiveness of ALS. Scafide et al. [15] highlight the need for a consistent method of bruise documentation and recommend the use of a Bruise Visibility Scale (BVS) and Absorption Visibility Scale (AVS) as standardized measures of bruise enhancement [15, 16]. Other studies have assessed bruises by measuring surface area (cm2) or diameter [1, 2, 9]. The latter approach is useful not only for research comparing bruise enhancement across lighting sources, but also in practical contexts where bruise size may be clinically or legally relevant, e.g., when comparing the shape and dimensions of a bruise with a suspected weapon or object. The aim of this study was to investigate whether ALS could enhance visualization of bruises on deceased individuals under normal lighting conditions without altering the autopsy setup. Photographs were taken using CWL, Violet ALS with a 415 nm long‐pass filter (VLP), violet ALS with a 415 nm Bandpass filter (VBP), and blue ALS with a 445 nm long‐pass filter (BLP). Bruises were assessed using a “modified” BVS and surface area measurements (cm2). MATERIALS AND METHODS The sample A total of 161 bruises documented on 25 deceased individuals from forensic autopsies were collected from September to November 2024 at the Department of Forensic Medicine, at the University of Copenhagen. Only individuals without visible signs of decomposition were selected. The data were used in an anonymous form. No information regarding the deceased, such as demographic (e.g., sex and age) or medical information, was known. Ethical consent This study was conducted using fully anonymized imaging data. According to applicable regulations and institutional policies, the use of anonymized, non‐identifiable data does not require ethical approval or informed consent, in line with the guidance provided by the Danish Research Ethics Committees (researchethics.dk). Bruise documentation and recording Each body was systematically examined by a forensic pathologist prior to autopsy to identify bruises as part of the routine external examination. The body was washed if there was dirt or staining on the skin. Each subject was assigned an alphabetical identifier (e.g., A, B, C), and each bruise was sequentially numbered starting from 1. Bruises were marked, their numeric codes recorded, and their anatomical locations documented on a standardized body diagram. Each bruise was photographed with a metric ruler including a color scale and the corresponding bruise ID using CrimeLite® AUTO (Foster+Freeman) [17]. The Crime‐lite® AUTO is a portable forensic imaging system that integrates a 20‐megapixel high‐resolution digital camera with multi‐spectral ultraviolet (UV), visible, and infrared illumination. An intuitive 7‐inch touchscreen interface provides live preview and image review, while the full‐spectrum sensitivity of the camera ensures enhanced detection and documentation of evidence that may be invisible under CWL. The following lights and filters were used: CWL, VLP, violet ALS with a 415 nm Bandpass filter (VBP), and BLP. The photographs were taken in the autopsy room under the standard lighting, which consists of LED ceiling panels. Color assessment and age classification A senior forensic pathologist performed a color assessment based on CWL photographs to estimate bruise age. Bruises without any yellow or brownish tint were classified as “recent”. Bruises showing a yellow or brownish discoloration—whether slight or predominant—were categorized, or bruises dominated by yellow or brown tones were considered “non‐recent” [18]. Bruise Visibility Scale The bruises were categorized using a modified BVS, which ranges from 1 (barely visible) to 5 (highly visible) following the indication of the study of Scafide et al. [15]. Figure 1 shows representative images of bruises under each lighting source, together with their corresponding BVS scores. This figure served as a reference throughout the scoring process to ensure consistent evaluation of the photographs. BVS scores were recorded using whole numbers only. A modified Bruise Visibility Scale (BVS), which ranges from 1 (barely visible) to 5 (highly visible) following the indication of the study of Scafide et al. [13]. Representative images of bruises under each lighting condition, together with their corresponding BVS scores, are shown. Each set of images, i.e., photographs of the same bruise taken with the four different light sources, was analyzed at separate times to prevent the evaluator from being influenced by previous observations of the same bruise. Surface area of bruises Bruise size analysis was performed in ImageJ [19]. To calibrate the scale, a 5 cm segment of the metric ruler visible in each image was measured at three different locations using the line tool. The average of the number of pixels per 5 cm of the 3 measurements was used to set the scale in ImageJ. For each bruise, the outline was manually traced along its visible borders to define the region of interest (ROI) using a WACOM® screen CINTIQ 21UX2 [20] using a pen mouse. ImageJ then automatically calculated the surface area of the bruise. Each bruise was measured independently three times, and the mean of these three measurements was recorded as the final area value. This procedure was repeated for all bruises under each of the four lighting conditions. Please refer to Data S1 Measurement of surface area of bruise for a step‐by‐step procedure. The overall approach was inspired by the image quantification methodology described by Trefan et al. [2]. Statistical analysis Data were analyzed using IBM SPSS for Windows, version 29.0.2 [21]. Bruise Visibility Scale A Shapiro–Wilk test was conducted to ascertain data normality. Descriptive statistics were used to calculate frequency and distribution. Logistic regression was performed to test for association between bruise visibility and light sources. Cox regression analysis was conducted to examine the association between light sources and BVS, stratified by individual bruise. For Cox regression analysis, BVS scores were dichotomized to meet the requirement of a binary event variable: with BVS 3–5 defined as the event and BVS 1–2 as the reference group. The null hypothesis was rejected at p < 0.05. Surface area A Shapiro–Wilk test was conducted to ascertain data normality. A related samples Wilcoxon Signed Rank Test was performed to determine whether there is a significant difference in bruise size when comparing the different light sources to CWL. Intra‐ and inter‐observer variability To assess BVS variability, a subset of 15 bruise photographs for each light source was evaluated by three observers: a medical student (observer 1), a forensic anthropologist (observer 2), and a senior forensic pathologist (observer 3). Weighted Cohen's kappa (κ) was used to assess intra‐ and inter‐observer variability in the test, as well as agreement between datasets, i.e., different light sources. According to the guidelines proposed by Landis and Koch [22], a kappa value of <0.2 was classified as slight agreement, 0.21–0.40 as fair, 0.41–0.60 as moderate, 0.61–0.80 as substantial, and >0.80 as near‐complete agreement [22]. To assess the surface area variability, a subset of 11 bruise photographs for each light source was evaluated by observer 1 and observer 3. The intra‐class correlation (ICC) test (two‐way random, single measures, absolute agreement) was used to assess the level of reliability between measurements. Results of the ICC tests were interpreted following the guidelines of Cicchetti [23]: poor reliability for ICC values <0.40, fair for values between 0.40 and 0.59, good for values between 0.60 and 0.74, and excellent for values >0.75. RESULTS The analysis included a total of 161 bruises that were photographed during the study. Of these, 99 were classified as “recent” and 62 as “non‐recent”. The observer variability tests revealed that, for BVS, the intra‐observer κ for observer 1 was 0.72, indicating substantial agreement. Inter‐observer agreement was 0.61 between observers 1 and 2, and 0.70 between observers 1 and 3, both reflecting substantial agreement. For surface area, the ICC results indicate excellent agreement for both intra‐ and inter‐observer reliability (0.98 and 0.83, respectively). When analyzed by light source, all lights show excellent agreement except VLP, which has poor inter‐observer reliability (ICC = 0.138). Bruise Visibility Scale The distribution of bruises across BVS categories for each light source and filter is presented in Table 1. TABLE 1BVS assessment distribution in number and percentage.BVS 1BVS 2BVS 3BVS 4BVS 5CWL23 (14.3%)49 (30.4%)31 (19.3%)35 (21.7%)23 (14.3%)VLP15 (9.3%)48 (29.8%)41 (25.5%)34 (21.1%)23 (14.3%)BLP31 (19.3%)64 (39.8%)31 (22%)22 (13.7%)13 (8.1%)VBP81 (50.3%)40 (24.8%)21 (13.0%)9 (5.6%)10 (6.2%) BVS assessment distribution in number and percentage. It seems that VLP slightly enhances bruise visibility compared to CWL, BLP reduces visibility compared to CWL, and VBP is in most cases barely detectable. Figures 1 and 2 show two examples of a bruise photographed with the different light sources. Example of bruise photographed with CWL (top right), VLP (top left), BLP (bottom right), and VBP (bottom left). VBP visualizes bruises with a larger area than in CWL, but with less contrast to the surrounding tissue compared to the other filters. The bruise is barely detectable. Logistic regression analyses are reported in Table 2. Odds ratio values below 1 (BLP and VBP) indicate lower odds compared to CWL, meaning these filters were associated with decreased visibility. For VBP, this reduction was statistically significant (p < 0.001). Although VLP provided better visual detection of bruises, the difference was not statistically significant. TABLE 2Logistic regression results BVS. p‐valueOdds ratio95% C.I. for odds ratioLowerUpperCWLRef.VLP0.5771.1690.6762.020BLP0.0320.5370.3040.949VBP<0.0010.2870.1540.536 Logistic regression results BVS. When examining the association between BVS and ALS stratified for age classification (Table 3), we found similar results. Both recent and non‐recent bruises appeared less visible when using VBP and BLP. VLP appeared to enhance bruise visualization; however, the difference was not statistically significant. TABLE 3Logistic regression of BVS stratified by bruise age. p‐valueOdds ratio95% C.I. for odds ratioLowerUpper1: recent bruisesCWLRef.VLP0.5771.1690.6762.020BLP0.0320.5370.3040.949VBP<0.0010.2870.1540.5362: not recent bruisesCWLRef.VLP0.2741.6280.6803.898BLP0.1040.5240.2401.142VBP<0.0010.2010.0910.442 Logistic regression of BVS stratified by bruise age. We examined the change in bruise visibility using Cox regression models stratified by age classification (Table 4). VBP was strongly associated with reduced bruise visibility compared to CWL, indicating that this light source significantly reduced bruise visibility in recent bruises and almost significantly in non‐recent bruises. BLP showed a reduction in visibility, but the difference was not statistically significant. VLP did not differ significantly from CWL. VLP and BLP showed no significant differences in non‐recent bruises. TABLE 4Cox regression models of BVS stratified by bruise age classification. p‐valueOdds ratio95% C.I. for odds ratioLowerUpper1: recentCWLRef.VLP0.6861.0850.7301.613BLP0.0930.6810.4341.067VBP0.0010.4260.2520.7182: not recentCWLRef.VLP0.5961.1190.7381.697BLP0.3390.8000.5071.264VBP0.0060.4760.2800.811 Cox regression models of BVS stratified by bruise age classification. Surface area Comparisons of bruise size under different lights revealed notable variations as can be seen in Figure 3. Below, it gives the percentage and raw count of observations in which one light produced a larger bruise than the other, for each pair: Example of bruises photographed with CWL (top right), VLP (top left), BLP (bottom right), and VBP (bottom left). VBP visualizes bruises with a larger area than in CWL but with less contrast to the surrounding tissue compared to the other filters. CWL vs. VLP: VLP larger in 66% (94) of 161 observations; CWL larger in 34% (48). CWL vs. BLP: BLP larger in 55% (78) of 161 observations; CWL larger in 45% (64). CWL vs. VBP: nearly equal—CWL larger in 49% (69) and VBP in 51% (73) of 161 observations. These results indicate that lighting conditions can influence apparent bruise size: VLP tends to visualize larger bruises than CWL, BLP often visualizes larger bruises than CWL, and VBP and CWL are roughly equivalent. Table 5 shows the result of the related‐samples Wilcoxon Signed Rank test (Table 5). VLP showed a positive and statistically significant effect, indicating a significant difference between VLP and CWL, both in the overall sample and when stratified by age groups (recent and non‐recent). In contrast, the differences for BLP and VBP were not statistically significant. TABLE 5Wilcoxon Signed Rank Test of bruise area for the entire sample and stratified by age classification.Test statisticStandard errorStandardized test statisticAsymptomatic sig. (2–2 sided test)entire sampleCWLRef.VLP6904.5491.0543.723<0.001*BLP4394.5491.053−1.3890.165VBP4881491.054−0.3980.6911: recentCWLRef.VLP2668.5252.6592.2780.023*BLP1508.5252.659−2.3130.21VBP1690252.659−1.5950.1112: not recentCWLRef.VLP89097.4993.0970.002*BLP66697.4990.7380.46VBP70797.4551.2210.22 Wilcoxon Signed Rank Test of bruise area for the entire sample and stratified by age classification. DISCUSSION We examined whether ALSs enhance bruise visibility in deceased individuals under a normal lighting condition to best fit with the routine workflow of autopsies. Our sample of 161 bruises was photographed using CWL, VLP, BLP, and VBP and examined for visibility and surface area. We found that the ALS camera did not enhance bruise visibility compared to CWL. VBP and BLP tended to reduce visibility compared to CWL, while VLP occasionally appeared to improve visibility. When comparing the bruise surface area among the light sources, we found that VLP visualized bruises as significantly larger than CWL, whereas BLP and VBP did not. We used a different design than previous studies [3, 6, 9, 24]: images were taken under normal room lighting rather than in darkened conditions, and we studied deceased individuals rather than living subjects. These differences limit direct comparability and may explain discrepant findings. Prior work has reported increased bruise visibility with VLP versus CWL. Downing et al. [3] found that 415 nm (VLP and VBP) and 445 nm (BLP) significantly improved visualization in living subjects imaged in darkened rooms, and Olds et al. [9] reported that VLP outperformed CWL and BLP in deceased subjects. Although our visibility results were not statistically significant, there was a tendency toward higher scores with VLP compared with CWL. The lack of significance may reflect reduced ALS effectiveness in a non‐darkened environment and the limited sample size. Analysis of bruise surface area, however, showed a statistically significant difference: VLP increased measured bruise area versus CWL. This suggests that ALS can enhance the apparent extent of bruises under normal lighting and that objective measures (e.g., surface area) may detect subtle differences that subjective visibility scales miss. Our initial thought was to use the BVS and AVS developed by Scafide et al. [15] and recommended by Tyr et al. [5]. However, we ultimately decided to create a “modified” BVS scale in which we assembled reference images showing representative bruises under each lighting condition together with their corresponding BVS scores. This was done to ensure consistent evaluation of each photograph. The BVS evaluates only the presence and visibility of a bruise and does not provide information about bruise size or the number of bruises. As a result, we argue that the full potential of ALS is not revealed when relying solely on BVS. Because bruise size is also an important factor in assessment, we additionally measured bruise surface area. Our findings show a statistically significant increase in surface area when using VLP compared to CWL, suggesting that VLP enhances the visible extent of bruises. In contrast, BLP and VBP did not consistently influence bruise size. Inter‐ and intra‐observer variability analyses indicated excellent agreement overall; however, VLP showed excellent intra‐observer agreement but poor inter‐observer agreement between the two observers. More observers should be added to confirm the results. However, it remains uncertain whether the increased bruise size observed with VLP reflects a true anatomical difference, such as deeper or more extensive hemorrhage, or whether it results from optical enhancement due to light‐tissue interactions. This distinction is critical, particularly in medico‐legal contexts, where the perceived size and severity of a bruise may influence the medicolegal assessment. Moreover, although ALS may aid in visualizing injuries, it is unclear whether the improvement represents enhanced detection or merely altered perception. Not all increases in apparent bruise size necessarily translate into improved diagnostic accuracy. Future studies should verify each finding to confirm that it is a bruise and determine its true extent—for example, in postmortem populations by making a small incision to confirm the presence of extravasated blood. In addition, photographing the entire body may help identify bruises that are not visible in targeted images. Strengths, limitations and future research Throughout the study, only the medical student photographed the bruises to ensure consistency, since familiarity with the camera was necessary. This reduced variability during image acquisition. Future studies should examine whether photographers influence the quality of the images. Our results show high reliability for the BVS. We included a forensic anthropologist as a proxy non‐expert observer to assess how the BVS performs when used by someone without specialized forensic pathology training. Future studies should include more observers with varying levels of expertise in forensic and clinical forensic pathology. To improve accuracy, a schematic overview of the BVS for each lighting condition was created (Figure 1). There are several limitations in our study. Firstly, we were unable to control for sex, age, and skin color in the regression models. These factors have been found to influence bruise physiology and were accounted for in the studies by Scafide et al. [25, 26]; therefore, their absence in our analysis may have affected the findings. For surface area measurements, intra‐observer agreement is excellent, except for VLP. Future studies should focus on developing a more automated detection method to determine the exact boundaries of bruises, as this remains a subjective process. Another limitation, highlighted in the introduction, is that we could not test specificity, i.e., whether the areas of enlargement revealed by VLP represented actual bruise tissue rather than imaging artifacts. In addition, the age classification of bruises based on color assessment was performed by a single evaluator using photographs, with the presence of yellow used to distinguish recent from non‐recent bruises. This is a limitation, as previous studies have demonstrated that bruise age estimation is subjective and may be influenced by the observer [27, 28] or display settings [29] and because yellow may not appear reliably [18]. To minimize variation, all evaluations were performed by a single observer using the same screen to ensure consistency. Nevertheless, future studies should involve multiple evaluators and establish consensus when estimating bruise age, preferably through direct examination during autopsy rather than from photographs. Finally, future studies should consider photographing bruises with the ALS camera both under darkened conditions and under normal conditions to validate the results of our study. CONCLUSION This study tested different ALS lights and filters for documenting bruises in deceased individuals under normal lighting conditions, without altering the autopsy setup. We found that the ALS camera did not enhance bruise visibility compared to CWL. VBP and BLP tended to reduce visibility compared with CWL, while VLP occasionally appeared to improve visibility; however, this effect was not statistically significant, likely due to the small sample size and the inability to account for confounding factors such as sex, age, and skin tone of the subjects. In contrast, surface area measurements indicated that VLP visualized bruises as significantly larger than CWL, whereas BLP and VBP did not. Overall, ALS lighting mainly affected apparent bruise size, with VLP consistently producing larger measurements than the other settings. Further studies are needed to assess the actual extent of bruises beneath the skin and to address this study's limitations. Because of these limitations, our results cannot be generalized. However, they represent a first step toward evaluating ALS under routine autopsy conditions without modifying the standard workflow or lighting. CONFLICT OF INTEREST STATEMENT The authors have no conflict of interest to declare. Supporting information Data S1. ACKNOWLEDGMENTS We thank the Editor‐in‐Chief, Associate Editor, and reviewer for their careful reading and constructive feedback, which substantially improved the clarity and quality of this manuscript. DATA AVAILABILITY STATEMENT The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions. REFERENCES Associated Data Supplementary Materials Data S1. Data Availability Statement The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.