Assessment of bruise age on dark-skinned individuals using tristimulus colorimetry

Quantifying the Degree of Bruise Visibility Observed Under White Light and an Alternate Light Source

BACKGROUND

Documentation of injuries associated with abuse and assault has pivotal impacts on clinical and legal outcomes. Before this study, no reliable and valid tools to consistently document the clinical visibility of bruises existed. The purpose of this study was to systematically evaluate reliability and validity of the Bruise Visibility Scale for documenting bruises visualized in normal (white) light and the Absorption Visibility Scale for documenting bruises visualized using an alternate light source (ALS).

METHODS

Bruises were induced using a paintball on the upper arms of 157 participants stratified into six skin color categories. Bruises were visualized 21 times over 4 weeks under white light and 10 ALS wavelength/goggle color combinations. Bruise size was measured using a metric ruler; bruise color was measured using a spectrophotometer. Interrater reliability was calculated using kappa and intraclass correlations coefficients. Construct validity was evaluated using generalized linear mixed modeling of associations between bruise size and color with both visibility scales.

RESULTS

Interrater agreement for bruise detection was over 90% for all but two ALS wavelength/goggle combinations. Kappa values indicated adequate interrater agreement under white light (κ = 0.76) and ALS (κ = 0.78). The visibility scale intraclass correlation coefficients were .91 for normal light and .93 for ALS. Statistical modeling showed greater bruise size was associated with higher visibility using either scale, and greater contrast in color or lightness was associated with higher Bruise Visibility Scale values.

IMPLICATIONS FOR PRACTICE

Both visibility scales showed satisfactory reliability and validity. Forensic nurses can use the scales to consistently document bruises.

Bruise dating using deep learning

The bruise dating can have important medicolegal implications in family violence and violence against women cases. However, studies show that the medical specialist has 50% accuracy in classifying a bruise by age, mainly due to the variability of the images and the color of the bruise. This research proposes a model, based on deep convolutional neural networks, for bruise dating using only images, by age ranges, ranging from 0-2 days to 17-30 days, and images of healthy skin. A 2140 experimental bruise photograph dataset was constructed, for which a data capture protocol and a preprocessing procedure are proposed. Similarly, 20 classification models were trained with the Inception V3, Resnet50, MobileNet, and MnasNet architectures, where combinations of learning transfer, cross-validation, and data augmentation were used. Numerical experiments show that classification models based on MnasNet have better results, reaching 97.00% precision and sensitivity, and 99.50% specificity, exceeding 40% precision reported in the literature. Also, it was observed that the precision of the model decreases with the age of the bruise.

Validation of alternate light sources for detection of bruises in non-embalmed and embalmed cadavers

Bruising is frequently documented in cases of violence for use as forensic evidence. However, bruises can be overlooked if they are not visible to the naked eye. Alternate light sources such as ultraviolet, narrow band, and infrared have been used in an attempt to reveal the presence of bruising that is not otherwise apparent. However, there is a significant gap in knowledge surrounding this technique as it has not been validated against histology to confirm that bruising is genuinely being enhanced. A recent study evaluated the ability of alternate light sources to enhance visibility of bruises using a pigskin model. However, histological confirmation of bruising in humans using these light sources has not yet been performed. In this study, embalmed and non-embalmed human cadavers were used. Bodies were surveyed with alternate light sources, and enhanced regions that were unapparent under white light were photographed with the alternate light sources and sampled for histological assessment. Immunohistochemical staining for the red blood cell surface protein glycophorin was used determine if the enhanced area was a bruise (defined by the presence of extravasated erythrocytes). Photographs of areas confirmed to be bruises were analyzed using the program Fiji to measure enhancement, which was defined as an increase in the measured transverse diameter. In the non-embalmed and the embalmed cadavers violet alternate light produced the greatest enhancement of histologically confirmed bruises, followed by blue (both p < 0.0001). Regions that were not confirmed as bruises also enhanced, indicating that light sources may not be specific. This suggests that the use of light sources to enhance the visibility of bruising should be undertaken with caution and further studies are required.

Reliability of tristimulus colourimetry in the assessment of cutaneous bruise colour

BACKGROUND

Bruising is one of the most common types of injury clinicians observe among victims of violence and other trauma patients. However, research has shown commonly used qualitative description of cutaneous bruise colour via the naked eye is subjective and unreliable. No published work has formally evaluated the reliability of tristimulus colourimetry as an alternative for assessing bruise colour, despite its clinical and research applications in accurately assessing skin colour. The purpose of this study was to systematically evaluate the test-retest and inter-observer reliability of tristimulus colourimetry in the assessment of cutaneous bruise colour.

METHODS

Two researchers obtained repeated tristimulus colourimetry measures of cutaneous bruises with participants of diverse skin colour. Measures were obtained using the Minolta CR-400 Chomameter. Commission Internationale d'Eclairage (CIE) L*a*b* colour space was used. Data was analysed using intraclass correlation coefficients (ICC), Cronbach's alpha, and minimal detectable change (MDC) on all three L*a*b* values.

RESULTS

The colorimeter demonstrated excellent test-retest or intra-rater reliability (L* ICC=0.999; a* ICC=0.973; b* ICC=0.892) and inter-rater reliability (L* ICC=0.997; a* ICC=0.976; b* ICC=0.982).

CONCLUSIONS

With consistent placement, the tristimulus colourimetry is reliable for the objective assessment and documentation of cutaneous bruise colour for purposes of clinical practice and research. Recommendations for use in practice/research are provided.

A novel, comprehensive, and reproducible porcine model for determining the timing of bruises in forensic pathology

PURPOSE

Calculating the timing of bruises is crucial in forensic pathology but is a challenging discipline in both human and veterinary medicine. A mechanical device for inflicting bruises in pigs was developed and validated, and the pathological reactions in the bruises were studied over time in order to identify gross and histological parameters that may be useful in determining the age of a bruise.

METHODS

The mechanical device was able to apply a single reproducible stroke with a plastic tube that was equivalent to being struck by a man. In each of 10 anesthetized pigs, four strokes that resulted in bruises were inflicted on the back. In addition, 2 control pigs were included in the study. The pigs were euthanized consecutively from 1 to 10 h after the infliction of bruises. Following gross evaluation, skin, and muscle tissues were sampled for histology.

RESULTS

Grossly, the bruises appeared uniform and identical to the tramline bruises seen in humans and pigs subjected to blunt trauma. Histologically, the number of neutrophils in the subcutis, the number of macrophages in the muscle tissue, and the localization of neutrophils and macrophages in muscle tissue showed a time-dependent response. Combining these parameters, bruises could be grouped as being either less than 4 h old or between 4 and 10 h of age. Gross lesions and changes in the epidermis and dermis were inconclusive with respect to time determination.

CONCLUSIONS

The model was reproducible and resembled forensic cases of bruises in pigs and humans. Therefore, the histological parameters are suitable for age determination of bruises in pigs and likely also in humans.

The use of photographs to record variation in bruising response in humans

There is considerable value in developing tools capable of accurately and reliably determining when bruises were inflicted in humans. Previous work has focused on the visual changes observed in a bruise as the injury develops and heals. However, due to variables such as how and where on the body the bruise was inflicted, differing tissue compositions at the injured skin site between individuals and inter- and intra-observer variation; a technique sufficiently robust for use in a clinical or medicolegal setting has not yet been identified. In this study we present a series of photographs taken under controlled conditions illustrating standardised bruises induced on participants using a weight dropping mechanism. We show that variation in the appearance of bruises over time across individuals is large and, although photography may be a suitable technique for the recording of injuries, it is not sufficiently reliable for determining the age of a bruise.