Artificial Intelligence in Forensic Odontology: A Descriptive Bibliometric Analysis

When a body is found in a state where traditional identification is impossible—perhaps due to severe decomposition or a mass disaster—forensic odontologists step in. These are dental experts who use the unique architecture of a person's teeth, jaws, and dental work to solve the mystery of who they were. For decades, this work has relied on comparing physical evidence, like a jawbone or a dental X-ray, against old medical records. It is a meticulous, human-driven process that can be slow and prone to error when the volume of evidence is high or the remains are difficult to examine. Recently, a new tool has entered the laboratory: artificial intelligence. This is not a single machine, but a collection of computer programs designed to learn from data, recognize patterns, and make decisions with a speed and consistency that humans cannot match. In the context of forensics, these programs can scan thousands of dental images in moments, looking for the specific shapes and structures that link a set of remains to a name.

A team of researchers set out to map how this technology is changing the field. They did not run new experiments on teeth or bodies; instead, they performed a large-scale review of the scientific literature itself. By gathering every relevant study published over the last fifteen years, they created a picture of the entire landscape of research. They looked at who was writing, where the work was coming from, and what specific questions the scientists were trying to answer. Their goal was to understand the current state of artificial intelligence in forensic dentistry, to see if the field was just beginning to wake up or if it had already matured into a standard practice.

The results of this review reveal a story of sudden, explosive growth. For the first nine years of the period they studied, from 2011 through 2019, the field was quiet. Only three articles were published in that entire time, suggesting that the idea of using computers to identify people through their teeth was still a distant concept. Then, starting in 2020, the activity surged. The number of published studies jumped dramatically, reaching a peak in 2025 with forty-one articles in a single year. This sharp rise indicates that the scientific community has recently recognized the potential of these tools, moving from theoretical curiosity to active investigation. The researchers found that the most popular application for these computer programs is the deep neural network, a type of advanced algorithm that mimics the way the human brain processes information. These networks are currently being used most frequently to estimate a person's age and determine their sex based on dental images, tasks that are fundamental to narrowing down a list of possible identities.

The analysis also highlighted where this knowledge is being generated. While India has produced the highest number of citations, indicating a strong influence on the field, the country with the most research output is not explicitly ranked by article count in the summary, though India, Indonesia, and China are the leading nations. Within the academic world, a few institutions stand out as the primary drivers of this work, with Universitas Airlangga in Indonesia leading the way in the number of articles published. When the researchers looked at which studies had the most influence, measured by how often other scientists cited them, they found that the most impactful work often focused on digital dental radiography. This suggests that the ability to analyze X-rays with high precision is currently the most valuable application of the technology. However, the review also noted a significant gap. Despite the rapid increase in publications, the field of artificial intelligence in forensic dentistry remains relatively small compared to other areas of dental research. It is still an emerging discipline, with plenty of room for new discoveries and for the technology to become a routine part of forensic investigations.

The data showed that while the number of papers is growing, the relationship between how many times a paper is cited and how long it has been published is not straightforward. Some older papers have not received as many citations as newer ones, indicating that the field is evolving quickly and that the most recent work is driving the conversation. The researchers concluded that while the tools are becoming more sophisticated and the volume of research is expanding, the field is not yet fully established. The path forward will require continued development of methods and international cooperation to ensure that these powerful computer systems can be reliably used in real-world legal and humanitarian situations. For now, the story is one of a field that has found its footing and is rapidly accelerating, promising to make the difficult task of human identification more accurate and efficient than ever before.