A New Method to Prevent Unintentional Child Poisoning

Area of Science:

• Biomedical Engineering
• Child Safety Technology
• Human-Computer Interaction

Background:

• Unintentional child poisoning is a significant global health concern, exacerbated by increased access to medications and supplements.
• Traditional child-resistant pill bottles present usability challenges for adults and can be overcome by persistent children.
• Biometric authentication for pill bottles is often complex and impractical for everyday use.

Purpose of the Study:

• To develop and evaluate a novel smart pill bottle system for enhanced child safety.
• To investigate the efficacy of touch capacitance-based user identification for deterring child access to medications.
• To reduce the incidence of unintentional child poisoning through real-time alerts and deterrents.

Main Methods:

• Development of a prototype smart pill bottle incorporating touch capacitance sensors.
• Implementation of machine learning algorithms, specifically Support Vector Machine (SVM) and Neural Network (NN), for user identification.
• Pilot testing with adult and child participants, analyzing bottle-opening events to assess detection accuracy.

Main Results:

• The optimized Neural Network (NN) model demonstrated high accuracy in distinguishing between adult and child users.
• Zero instances of children being incorrectly identified as adults were recorded.
• Four instances of adults being incorrectly identified as children occurred during pilot testing with 232 bottle-opening events.

Conclusions:

• Smart pill bottles utilizing touch capacitance offer a promising solution for preventing child access to medications.
• The developed system can reliably detect children attempting to open pill bottles, thereby mitigating poisoning risks.
• This technology has the potential to significantly enhance child safety and reduce accidental ingestion incidents.