Neural network approach to evaluate the physical properties of dentin

Area of Science:

• Biomaterials Science
• Dental Materials
• Computational Biology

Background:

• Diabetic individuals exhibit compromised root canal dentin properties.
• Inorganic trace elements may offer potential benefits for dentin enhancement.
• Artificial Neural Networks (ANN) present a novel approach for material property evaluation.

Purpose of the Study:

• To evaluate the effects of magnesium (Mg), strontium (Sr), and zinc (Zn) on root canal dentin properties in diabetic individuals.
• To assess the efficacy of Artificial Neural Networks (ANN) in analyzing these effects.
• To determine the impact of trace element exposure duration on dentin characteristics.

Main Methods:

• Three hundred extracted human premolars from type II diabetic individuals were used.
• Specimens were treated with Mg, Sr, or Zn solutions for varying durations (0-10 minutes).
• Root fracture resistance (RFR), surface microhardness (SμH), and tubular density (TD) were measured. ANN was employed for data analysis.

Main Results:

• Mg, Sr, and Zn treatments significantly increased RFR and SμH (P<0.05).
• These elements significantly decreased dentin tubular density (TD) (P<0.05).
• No significant differences were found between manual and ANN evaluation methods (P>0.05).

Conclusions:

• Mg, Sr, and Zn show potential for improving RFR and SμH while decreasing TD in root canal dentin of diabetic patients.
• ANN is a reliable tool for evaluating the physical properties of dentin.
• Further research into optimal trace element concentrations and exposure times is warranted.