Fail-safe systems for the wearable artificial endocrine pancreas

Area of Science:

• Biomedical Engineering
• Medical Devices
• Endocrinology

Background:

• Long-term clinical use of artificial endocrine pancreas requires robust safety mechanisms.
• Existing wearable artificial pancreas systems incorporate hardware and software alarms.

Purpose of the Study:

• To evaluate the efficacy of a fail-safe system for a wearable artificial endocrine pancreas.
• To assess the system's ability to ensure long-term glycemic control in diabetics.

Main Methods:

• Incorporation of hardware alarms for battery and pump issues.
• Implementation of software alarms for converter status, insulin levels, and glycemic parameters.
• Development of a software noise filter to mitigate sensor interferences.
• Detection of sensor abnormalities via analog-to-digital converter (AD converter) overflow/underflow.

Main Results:

• Hardware alarms effectively addressed battery and pump failures.
• Software alarms successfully detected conditions like hypoglycemia, hyperglycemia, and abnormal insulin delivery.
• The noise filter proved effective against artificial and exercise-induced sensor noise.
• Sudden sensor lead disconnections or short circuits were identified as AD converter overflow/underflow events.
• Gradual sensor output changes were accurately signaled by hyperglycemia or hypoglycemia alarms.

Conclusions:

• The integrated fail-safe system is essential for the wearable artificial endocrine pancreas.
• The system demonstrates utility in providing safe and effective long-term glycemic management for diabetic patients.
• Fail-safe mechanisms enhance the reliability and clinical applicability of artificial pancreas technology.