Morphometric and Gray-Level Texture Analyses of Postmortem Injuries Using ImageJ: An Exploratory Study.

Background The quantitative methods of image analysis can offer objective data on the structure of the injury, its edge architecture, geometrical properties (such as length, width, area, perimeter, and solidity), and gray-level textures. Aim To explore the utility of quantitative morphometric and gray-level texture analyses of postmortem injuries. Materials and methods This pilot study was performed in a cross-sectional manner and used an exploratory design in the Department of Forensic Medicine & Toxicology, King George's Medical University (KGMU), Lucknow. A total of 383 injury images in medico-legal autopsies were studied with the help of the ImageJ software (version 1.53, National Institutes of Health, Bethesda, MD, USA). Parameters such as length, width, area, perimeter, solidity, and gray-level texture were evaluated. The statistical analysis comprised descriptive statistics, comparative analysis, correlation analysis, multicollinearity, exploratory logistic regression, and receiver operating characteristic (ROC) analysis. Results Notable differences were seen for the area, perimeter, solidity, and gray-level standard deviation (SD) between injury groups (p<0.05). There was a notable positive correlation between the area and perimeter (r=0.88). Regression analysis showed statistical significance for the area, SD, and solidity. The exploratory analysis yielded good discrimination power with an area under the curve (AUC) of 0.84 (95% confidence interval (CI): 0.77-0.90). Conclusion The above exploratory investigation shows that image-based techniques such as quantitative morphometry and grayscale texture analysis using the ImageJ software could potentially offer reliable, quantifiable, and objective data about postmortem trauma. Nonetheless, image-based measurements must only be used as a supplementary tool to the traditional forensic evaluation and cannot in any case replace the gross examination and expert opinion of a forensic pathologist. There is a need for large-scale, multicenter investigations with external validation before the said techniques can be considered for forensic use.