Machine learning utilising spectral derivative data improves cellular health classification through hyperspectral infra-red spectroscopy.

The objective differentiation of facets of cellular metabolism is important for several clinical applications, including accurate definition of tumour boundaries and targeted wound debridement. To this end, spectral biomarkers to differentiate live and necrotic/apoptotic cells have been defined using in vitro methods. The delineation of different cellular states using spectroscopic methods is difficult due to the complex nature of these biological processes. Sophisticated, objective classification methods will therefore be important for such differentiation. In this study, spectral data from healthy/traumatised cell samples using hyperspectral imaging between 2500-3500 nm were collected using a portable prototype device. Machine learning algorithms, in the form of clustering, have been performed on a variety of pre-processing data types including 'raw' unprocessed, smoothed resampling, background subtracted and spectral derivative. The resulting clusters were utilised as a diagnostic tool for the assessment of cellular health and quantified using both sensitivity and specificity to compare the different analysis methods. The raw data exhibited differences for one of the three different trauma types applied, although unable to accurately cluster all the traumatised samples due to signal contamination from the chemical insult. The background subtracted and smoothed data sets reduced the accuracy further, due to the apparent removal of key spectral features which exhibit cellular health. However, the spectral derivative data-types significantly improved the accuracy of clustering compared to other data types, with both sensitivity and specificity for the background subtracted data set being >94% highlighting its utility to account for unknown signal contamination while maintaining important cellular spectral features.

Blast Wave Exposure to the Extremities Causes Endothelial Activation and Damage.

Extremity injury is a significant burden to those injured in explosive incidents and local ischaemia can result in poor functionality in salvaged limbs. This study examined whether blast injury to a limb resulted in a change in endothelial phenotype leading to changes to the surrounding tissue.The hind limbs of terminally anaesthetized rabbits were subjected to one of four blast exposures (high, medium, low, or no blast). Blood samples were analyzed for circulating endothelial cells pre-injury and at 1, 6, and 11 h postinjury as well as analysis for endothelial activation pre-injury and at 1, 6, and 12  h postinjury. Post-mortem tissue (12  h post-injury) was analysed for both protein and mRNA expression and also for histopathology. The high blast group had significantly elevated levels of circulating endothelial cells 6  h postinjury. This group also had significantly elevated tissue mRNA expression of IL-6, E-selection, TNF-α, HIF-1, thrombomodulin, and PDGF. There was a significant correlation between blast dose and the degree of tissue pathology (hemorrhage, neutrophil infiltrate, and oedema) with the worst scores in the high blast group. This study has demonstrated that blast injury can activate the endothelium and in some cases cause damage that in turn leads to pathological changes in the surrounding tissue. For the casualty injured by an explosion the damaging effects of hemorrhage and shock could be exacerbated by blast injury and vice versa so that even low levels of blast become damaging, all of which could affect tissue functionality and long-term outcomes.

A pre-clinical evaluation of silver, iodine and Manuka honey based dressings in a model of traumatic extremity wounds contaminated with Staphylococcus aureus.

Prevention of extremity war wound infection remains a clinical challenge. Staphylococcus aureus is the most common pathogen in delayed infection. We hypothesised that choice of wound dressings may affect bacterial burden over 7 days reflecting the current practice of delayed primary closure of wounds within this timeframe. A randomised controlled trial of 3 commercially available dressings (Inadine(®) (Johnson & Johnson, NJ, USA), Acticoat(®) (Smith & Nephew, Hull, UK), Activon Tulle (Advancis Medical, Nottingham, UK)) was conducted in a rabbit model of contaminated forelimb muscle injury. A positive control group treated with antibiotics was included. Groups were compared to a saline soaked gauze control. The primary outcome was a statistically significant reduction (p < 0.05) in tissue S. aureus at 7 days post-injury. Secondary outcome measurements included bacteraemias, observational data, whole blood determination, ELISA for plasma biomarkers, PCR array analysis of wound healing gene expression and muscle/lymph node histopathology. Antibiotic, Inadine and Acticoat groups had statistically significant lower bacterial counts (mean 7.13 [95% CI 0.00-96.31]×10(2); 1.66 [0.94-2.58]×10(5); 8.86 [0.00-53.35]×10(4)cfu/g, respectively) and Activon Tulle group had significantly higher counts (2.82 [0.98-5.61]×10(6)cfu/g) than saline soaked gauze control (7.58 [1.65-17.83]×10(5)cfu/g). There were no bacteraemias or significant differences in observational data or whole blood determination. There were no significant differences in muscle/loss or pathology and lymph node cross-sectional area or morphology. There were some significant differences between treatment groups in the plasma cytokines IL-4, TNFα and MCP-1 in comparison to the control. PCR array data demonstrated more general changes in gene expression in the muscle tissue from the Activon Tulle group than the Inadine or Acticoat dressings with a limited number of genes showing significantly altered expression compared to control. This study has demonstrated that both Acticoat(®) and Inadine(®) dressings can reduce the bacteria burden in a heavily contaminated soft tissue wound and so they may offer utility in the clinical setting particularly where surgical treatment is delayed.