Proactive Meal Detection and Personalized Insulin Dosing System

A next-generation system that integrates wearable devices, machine learning algorithms, and cloud-based analytics to proactively detect un-bolused meals and provide personalized insulin dosing recommendations, revolutionizing diabetes therapy and management.

The original patent, 'Automatic detection of un-bolused meals', addressed the limitations of traditional diabetes management by providing a method for detecting un-bolused meals and reminding users to deliver meal boluses. However, this approach relied on user input and CGM data, which can be incomplete or inaccurate. The new inventive concept solves this problem by introducing proactive meal detection, machine learning-based pattern recognition, and real-time insulin dosing adjustments, providing a more comprehensive and accurate solution for diabetes therapy.

The system comprises a wearable device with a non-invasive glucose sensor, a pattern recognition module, and a haptic feedback module. The wearable device continuously monitors the user's glucose levels and detects patterns indicative of meal consumption using machine learning algorithms. The system then sends personalized alerts to the user to deliver meal boluses. Additionally, the system integrates with a cloud-based platform that analyzes CGM data, insulin dosing regimens, and user behavior to provide real-time guidance on optimal insulin dosing and meal planning. The platform's predictive modeling module forecasts glucose levels and insulin requirements, enabling proactive adjustments to the insulin dosing regimen.

The new claims introduce a paradigm shift in diabetes therapy by integrating wearable devices, machine learning algorithms, and cloud-based analytics to provide proactive meal detection and personalized insulin dosing. This approach is novel and non-obvious compared to the original patent, which relied on user input and CGM data. The inventive step lies in the combination of these technologies to create a comprehensive and accurate solution for diabetes therapy.

Alternative embodiments of the inventive concept could include the use of different types of wearable devices, such as smartwatches or fitness trackers, or the integration of other health data, such as activity levels or nutrition information. Variations could also include different machine learning algorithms or cloud-based analytics platforms.

The inventive concept has significant commercial potential in the diabetes management market, with potential applications in insulin pump therapy, CGM systems, and digital health platforms. The market for diabetes management is growing rapidly, driven by the increasing prevalence of diabetes and the need for more effective and personalized solutions.