Forensic Science Program, Department of Applied Science, Faculty of Science, Prince of Songkla University, Hat Yai, Songkhla 90112, Thailand.
Forensic Sci Int. 2013 Dec 10;233(1-3):288-97. doi: 10.1016/j.forsciint.2013.09.027. Epub 2013 Oct 9.
Recent studies on bloodstains have focused on determining the time since deposition of bloodstains, which can provide useful temporal information to forensic investigations. This study is the first to use smartphone cameras in combination with a truly low-cost illumination system as a tool to estimate the age of bloodstains. Bloodstains were deposited on various substrates and photographed with a smartphone camera. Three smartphones (Samsung Galaxy S Plus, Apple iPhone 4, and Apple iPad 2) were compared. The environmental effects - temperature, humidity, light exposure, and anticoagulant - on the bloodstain age estimation process were explored. The color values from the digital images were extracted and correlated with time since deposition. Magenta had the highest correlation (R(2)=0.966) and was used in subsequent experiments. The Samsung Galaxy S Plus was the most suitable smartphone as its magenta decreased exponentially with increasing time and had highest repeatability (low variation within and between pictures). The quantifiable color change observed is consistent with well-established hemoglobin denaturation process. Using a statistical classification technique called Random Forests™, we could predict bloodstain age accurately up to 42 days with an error rate of 12%. Additionally, the age of forty blind stains were all correctly predicted, and 83% of mock casework samples were correctly classified. No within- and between-person variations were observed (p>0.05), while smartphone camera, temperature, humidity, and substrate color influenced the age determination process in different ways. Our technique provides a cheap, rapid, easy-to-use, and truly portable alternative to more complicated analysis using specialized equipment, e.g. spectroscopy and HPLC. No training is necessary with our method, and we envision a smartphone application that could take user inputs of environmental factors and provide an accurate estimate of bloodstain age.
近期的血痕研究主要集中在确定血痕的沉积时间上，这可以为法医学调查提供有用的时间信息。本研究首次使用智能手机相机结合真正低成本的照明系统作为工具来估计血痕的年龄。在各种基质上沉积血痕，并使用智能手机相机进行拍照。比较了三款智能手机（三星 Galaxy S Plus、苹果 iPhone 4 和苹果 iPad 2）。探索了环境因素（温度、湿度、光照和抗凝剂）对血痕年龄估计过程的影响。从数字图像中提取颜色值，并与沉积时间进行相关分析。品红色具有最高的相关性（R²=0.966），并在后续实验中使用。三星 Galaxy S Plus 是最适合的智能手机，因为其品红色随时间的增加呈指数下降，且具有最高的可重复性（图片内和图片间的变化较小）。观察到的可量化颜色变化与成熟的血红蛋白变性过程一致。使用一种称为随机森林的统计分类技术，我们可以准确预测血痕年龄长达 42 天，误差率为 12%。此外，四十个盲样的年龄都被准确预测，83%的模拟案例样本被正确分类。未观察到个体内和个体间的差异（p＞0.05），而智能手机相机、温度、湿度和基质颜色以不同的方式影响年龄确定过程。我们的技术提供了一种廉价、快速、易于使用且真正便携的替代方案，用于更复杂的分析，例如光谱分析和 HPLC。我们的方法不需要培训，我们设想一个智能手机应用程序，可以接受用户输入的环境因素，并提供血痕年龄的准确估计。
