AI Enhances Dental Anthropology Research

Mario Modesto Mata, a researcher at the Dental Anthropology Group at the Centro Nacional de Investigación sobre la Evolución Humana (CENIEH), has spearheaded a groundbreaking study utilizing artificial neural networks to reconstruct the number of perikymata, the growth lines in enamel, particularly in worn teeth. This research, published in The Anatomical Record, promises to significantly advance evolutionary studies by overcoming challenges associated with tooth wear.

Modesto Mata underscores the significance of this breakthrough, stating, “Solving this problem is of vital importance, as it would let us increase the number of teeth that are suitable for evolutionary studies, and so lead to more reliable conclusions.” This work is part of the European project Tied2Teeth, led by researcher Leslea Hlusko.

America’s Native Population Traced to Single Asian Migration Wave, According to Dental Anthropologists

Predicting Perikymata with Artificial Intelligence

The study outlines a methodology to predict the number of perikymata in worn teeth using artificial intelligence techniques. By measuring the enamel’s diminishment, expressed as a percentage of crown height lost, researchers can employ artificial neural networks to estimate the missing perikymata count accurately.

Artificial neural networks were specifically trained to predict perikymata count in teeth missing up to 30% of their enamel. Validation of the neural networks demonstrates remarkable precision, with a maximum error of only 3 perikymata in total observed in 86% of cases when 30% of enamel is lost.

Modesto Mata remarks on the precision of the data, highlighting its potential to accurately predict enamel formation time, thus opening avenues for investigating paleobiological questions using neural networks.

Ancient Primate Teeth Suggest Soft-Food Diets Dominated Early Menus

Accessibility and Application

To facilitate widespread adoption of this innovative approach, the researchers have developed user-friendly software called teethR, available as an R package. This software, named for its focus on teeth analysis, requires minimal training in artificial intelligence, making it accessible to researchers with basic knowledge of R. With the package’s functions, predictions can be made swiftly, maximizing the utility of artificial neural networks in dental anthropology research.

The development of teethR signifies a significant step forward in democratizing access to advanced analytical tools, potentially revolutionizing the field of dental anthropology and contributing to more robust evolutionary studies.

Tooth Analysis Reveal Longer Breastfeeding Among Middle Ages Bavarians