System and method for generating battery alarms in infusion devices
Inventors
Fischer, Steven Ward • Vogel, Matthew Stephen • Chen, Ye
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-12589203-B2
Publication Date
2026-03-31
Expiration Date
Abstract
A system and method are disclosed for detecting remaining battery voltage or capacity in an infusion device and generating alarms based on the detection. A battery lifetime extension method includes providing an infusion device that derives its power from a rechargeable battery. The infusion device may derive its power from a rechargeable battery. Furthermore, the infusion device receives, at predetermined intervals of time in real-time sensor data comprising a voltage, a change in the voltage over the predetermined interval of time, an average current, a temperature, and a remaining voltage or capacity reported by a battery gas gauge integrated circuit (“IC”) associated with the rechargeable battery. A customized neural network model utilizes the sensor data to determine an indicia of the actual remaining voltage or capacity of the rechargeable battery in real-time. The indicia may be used to lengthen and/or abate ongoing medical infusion therapy.
Core Innovation
The disclosed problem concerns battery alarms that do not correspond closely to the remaining battery power: the conventional method can issue a low-battery alarm 78 minutes before depletion, compared with an ideal time of 33 minutes. This difference sacrifices battery runtime, and interruptions to battery-powered infusion therapy are also identified. The invention addresses battery-state estimation and alarm timing in infusion and medical devices.
The device uses real-time measurements of a rechargeable battery to generate a feature vector, which a trained neural network applies to determine the actual remaining voltage or capacity. It generates an alarm for low, very low, or depleted battery states according to thresholds. In one medical-device configuration, the device identifies the battery type and selects a trained neural network for that type.
Claims Coverage
The two independent claims cover three main inventive features: neural-network battery estimation and state-specific alarms in an infusion device; battery-type-based neural-network selection in a medical device; and use of the selected network for battery estimation and alarms.
Real-time battery estimation and alarms in an infusion device
The infusion device receives or determines real-time measurements comprising at least three battery measurements and generates a feature vector that also comprises the battery’s full-charge voltage or capacity. A trained neural network determines the actual remaining voltage or capacity, and the processor generates an alarm indicating a low, very low, or depleted battery state according to the respective thresholds.
Battery-type-based neural-network selection
The medical device determines information indicative of the rechargeable battery type and selects, from a plurality of trained neural networks for different rechargeable battery types, the network based on that information.
Battery estimation and state-specific alarms in a medical device
The medical device applies a feature vector containing the full-charge voltage or capacity and at least three real-time battery measurements to the selected trained neural network to determine actual remaining voltage or capacity. It generates an alarm indicating the low, very low, or depleted battery state according to the thresholds.
The claims cover real-time neural-network estimation of rechargeable-battery voltage or capacity and alarms for defined battery states, with the medical-device claim additionally selecting a trained network based on battery type.
Stated Advantages
More reliable battery alarms that more closely match ideal alarm timing.
Longer battery operation by reducing the battery runtime sacrificed by alarms issued too early.
Fewer interruptions to infusion therapy.
Documented Applications
Battery-state monitoring and alarm generation in infusion devices used for infusion therapy.
Medical devices identified in the claims include peristaltic pumps, syringe pumps, and ambulatory pumps.
Interested in licensing this patent?