Štepanovský Michal, Buk Zdeněk, Pilmann Kotěrová Anežka, Brůžek Jaroslav, Bejdová Šárka, Techataweewan Nawaporn, Velemínská Jana
Faculty of Information Technology, Czech Technical University in Prague, Thakurova 9, Prague 160 00, Czech Republic.
Department of Anthropology and Human Genetics, Faculty of Science, Charles University, Vinicna 7, Prague 128 43, Czech Republic.
Forensic Sci Int. 2024 Dec;365:112272. doi: 10.1016/j.forsciint.2024.112272. Epub 2024 Oct 28.
Age-at-death estimation is usually done manually by experts. As such, manual estimation is subjective and greatly depends on the past experience and proficiency of the expert. This becomes even more critical if experts need to evaluate individuals with unknown population affinity or with affinity that they are not familiar with. The purpose of this study is to design a novel age-at-death estimation method allowing for automatic evaluation on computers, thus eliminating the human factor.
We used a traditional machine-learning approach with explicit feature extraction. First, we identified and described the features that are relevant for age-at-death estimation. Then, we created a multi-linear regression model combining these features. Finally, we analysed the model performance in terms of Mean Absolute Error (MAE), Mean Bias Error (MBE), Slope of Residuals (SoR) and Root Mean Squared Error (RMSE).
The main result of this study is a population-independent method of estimating an individual's age-at-death using the acetabulum of the pelvis. Apart from data acquisition, the whole procedure of pre-processing, feature extraction and age estimation is fully automated and implemented as a computer program. This program is a part of a freely available web-based software tool called CoxAGE3D, which is available at https://coxage3d.fit.cvut.cz/. Based on our dataset, the MAE of the presented method is about 10.7 years. In addition, five population-specific models for Thai, Lithuanian, Portuguese, Greek and Swiss populations are also given. The MAEs for these populations are 9.6, 9.8, 10.8, 10.5 and 9.2 years, respectively. Our age-at-death estimation method is suitable for individuals with unknown population affinity and provides acceptable accuracy. The age estimation error cannot be completely eliminated, because it is a consequence of the variability of the ageing process of different individuals not only across different populations but also within a certain population.
死亡年龄估计通常由专家手动完成。因此，手动估计具有主观性，并且很大程度上取决于专家过去的经验和熟练程度。如果专家需要评估与未知人群有亲缘关系或与他们不熟悉的亲缘关系的个体，这一点就变得更加关键。本研究的目的是设计一种新颖的死亡年龄估计方法，以便在计算机上进行自动评估，从而消除人为因素。
我们使用了一种具有显式特征提取的传统机器学习方法。首先，我们识别并描述了与死亡年龄估计相关的特征。然后，我们创建了一个结合这些特征的多元线性回归模型。最后，我们根据平均绝对误差（MAE）、平均偏差误差（MBE）、残差斜率（SoR）和均方根误差（RMSE）分析了模型性能。
本研究的主要成果是一种使用骨盆髋臼估计个体死亡年龄的与人群无关的方法。除了数据采集外，预处理、特征提取和年龄估计的整个过程都是完全自动化的，并作为一个计算机程序实现。该程序是一个名为CoxAGE3D的免费基于网络的软件工具的一部分，可在https://coxage3d.fit.cvut.cz/获取。基于我们的数据集，所提出方法的MAE约为10.7岁。此外，还给出了针对泰国、立陶宛、葡萄牙、希腊和瑞士人群的五个特定人群模型。这些人群的MAE分别为9.6、9.8、10.8、10.5和9.2岁。我们的死亡年龄估计方法适用于与未知人群有亲缘关系的个体，并提供了可接受的准确性。年龄估计误差无法完全消除，因为这是不同个体衰老过程变异性的结果，不仅在不同人群之间存在差异，而且在特定人群内部也存在差异。
