fairfolio
Saved Projects

What If Teeth Had Nano-Armor? Using Artificial Intelligence to Predict the Best Protective Coating

CSEF · 2026 Materials Science (Junior Division)

Praganya Jeyaprassad

Overview

Abstract of the Project The purpose of this project was to investigate whether protective surface coatings reduce acid-induced demineralization on tooth-like substrates and to evaluate artificial intelligence (AI) as an objective tool for assessing surface damage. Eggshells were used as models for tooth enamel because they are composed primarily of calcium carbonate, a mineral which is similar to the type found in enamel called hydroxyapatite. The hypothesis stated that if either a hydrophobic barrier (petroleum jelly) or a hard-polymer barrier (clear nail polish) were applied to a calcium-based surface, then acid exposure from household beverages would result in significantly less surface degradation compared to uncoated samples, due to reduced direct acid contact. Six eggs were divided into three experimental groups: uncoated (control), petroleum jelly coating, and clear nail polish coating. Petroleum jelly was selected as a hydrophobic coating to repel liquid contact, while clear nail polish was selected as a hard-polymer coating to create a rigid protective barrier. Each egg was fully submerged for 24 hours in soda, fruit juice, or coffee to simulate regular dietary acid exposure. Three independent liquid trials with replicated samples were conducted to improve reliability. Surface damage was quantified using a composite scoring system based on texture, crack formation, and discoloration. These values were combined into a Total Damage Score, with higher scores indicating greater erosion. In the soda trials, uncoated samples demonstrated the highest average damage score (6), petroleum jelly coated samples showed moderate protection (3), and clear nail polish coated samples exhibited the lowest damage (1). Similar trends were observed in the coffee trials. Fruit juice resulted in less structural damage overall but produced measurable differences in staining between treatment groups. To enhance objectivity, pre- and post-exposure images were analyzed using a trained AI image classification model (Google Teachable Machine). The model was trained on labeled examples of low, medium, and high damage and then used to classify new images. The AI consistently categorized clear nail polish samples as low damage and uncoated samples as high damage, supporting the quantitative scoring results. The findings support the hypothesis that barrier coatings can significantly reduce acid-related surface degradation in a calcium-based model system. Clear nail polish demonstrated the greatest protective effect among the materials tested. Additionally, AI-based image classification demonstrates that artificial intelligence can be a valuable tool for analyzing experimental image data and identifying damage patterns. These findings have potential applications in dentistry and materials science. Future nano-scale protective coatings may be developed to strengthen enamel, improve dental sealants, and reduce tooth decay. Additionally, artificial intelligence may help researchers rapidly evaluate new dental materials and predict their effectiveness in preventing enamel erosion. Future research could investigate coatings that more closely mimic dental biomaterials, incorporate larger sample sizes, extend exposure durations, and utilize microscopic surface analysis to measure erosion at higher resolution.

Competition history

  • CSEF 2026 Materials Science (Junior Division) · Entry J-13-02

Related projects

Closest projects by meaning, across every fair and year in the corpus.

Browse more like this

Source: California Science & Engineering Fair public projects

Save projects to your library

Sign in with Google to keep track of projects you find interesting, organized into folders. An account also raises your daily allowance for “Has this been done?”, and lets you create a key for the MCP server with a much higher limit than anonymous use. Browsing stays public.

Continue with Google