On the use of Aggregation Operators to improve Human Identification using Dental Records

The comparison of dental records is a standardized technique in forensic dentistry used to speed up the identification of individuals in multiple-comparison scenarios. Specifically, the odontogram comparison is a procedure to compute criteria that will be used to perform a ranking. State-of-the-art automatic methods either make use of simple techniques, without utilizing the full potential of the information obtained from a comparison, or their internal behavior is not known due to the lack of peer-reviewed publications. This work aims to design aggregation mechanisms to automatically compare pairs of dental records that can be understood and validated by experts, improving the current methods. To do so, we introduce different aggregation approaches using the state-of-the-art codification, based on seven different criteria. In particular, we study the performance of i) data-driven lexicographical order-based aggregations, ii) well-known fuzzy logic aggregation methods and iii) machine learning techniques as aggregation mechanisms. To validate our proposals, 215 forensic cases from two different populations have been used. The results obtained show how the use of white-box machine learning techniques as aggregation models (average ranking from 2.02 to 2.21) are able to improve the state-of-the-art (average ranking of 3.91) without compromising the explainability and interpretability of the method.

💡 Research Summary

The paper addresses a critical bottleneck in forensic human identification: the automated comparison of dental records (odontograms) in mass‑disaster victim identification (DVI) scenarios. While DNA and fingerprint analyses are highly reliable, they often fail when remains are severely compromised (e.g., burned or incinerated bodies). Teeth, being the hardest tissue in the human body, retain diagnostic information even under extreme conditions, making odontogram comparison a cornerstone of forensic dentistry.

Current state‑of‑the‑art software (CAPMI, WinID, KMD PlassData DVI, etc.) relies on a lexicographic ordering of a small set of criteria (match, possible match, mismatch). This approach uses the criteria only to break ties, discarding much of the information that could improve ranking accuracy. Moreover, the internal scoring mechanisms are proprietary, rendering the systems opaque and non‑compliant with the European AI Act’s ALTAI framework, which demands transparency, human oversight, robustness, fairness, and accountability for high‑risk AI applications such as biometric identification.

To overcome these limitations, the authors propose three families of aggregation operators that combine the seven criteria derived from the Simplified Coding System (SCS) – V (virgin), F (filled), S (special treatment), X (missing), I (implant), P (present but not visible), N (no information). The aggregation mechanisms are:

1. Data‑driven lexicographic order‑based aggregation – statistical analysis of the training set determines the optimal priority order of criteria; once fixed, each criterion receives a learned weight and the final score is a weighted sum.

2. Fuzzy‑logic aggregation – each of the three raw outcomes (match, possible match, mismatch) is mapped to a fuzzy membership value in