Ontology Sharding for Medication Data Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Health applications face challenges in efficiently updating and managing large health-related ontologies on user devices while maintaining privacy and respecting storage and memory constraints, particularly in handling medication interactions and scheduling, which requires efficient data retrieval and storage of medication information without overwhelming device resources.
Innovation Solution
The implementation of ontology sharding, where the global ontology is partitioned into smaller shards, allowing only required shards to be stored and updated on user devices, with a server-side approach that transmits specific shards for integration into the embedded ontology, reducing memory and processing demands, and the use of user domain concepts for customizable health data management.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If the global ontology is stored completely on user devices, then data accessibility and completeness are improved, but memory storage requirements and processing resources are overwhelmed
Solution Approach 1:
The global ontology is divided into multiple smaller shards, each containing a specific subset of ontology data. User devices store only the relevant shards needed for their specific health application needs, rather than the complete ontology. This segmentation allows devices to maintain data completeness for their requirements while significantly reducing storage requirements.
Solution Approach 2:
Only the necessary portions of the global ontology are extracted and transmitted to user devices. The server identifies and sends specific shards that contain the ontology data required for medication interactions, scheduling, and other health functions, eliminating the need to store unnecessary ontology data on device.
2Measurement precision
If the global ontology is updated frequently on user devices, then data accuracy and currency are improved, but processing resources and update complexity increase
Solution Approach 1:
Updates are applied at the shard level rather than requiring complete ontology re-downloads. When the global ontology changes, only the affected shards are identified, updated on the server, and selectively transmitted to user devices. This granular update approach reduces processing complexity while maintaining data accuracy.
Solution Approach 2:
The server pre-processes and prepares updated shards before transmission to user devices. By organizing updates in advance and identifying which shards need updating before user device interaction, the system minimizes on-device processing requirements and simplifies the update experience for users.
3Speed
If more ontology data is stored on user devices, then data retrieval speed is improved, but storage capacity constraints are exceeded
Solution Approach 1:
The ontology is segmented into shards that are selectively stored on user devices based on their specific health application needs. This ensures that devices store only the minimum necessary data locally for fast retrieval of commonly accessed information, while maintaining the ability to access additional ontology data from the server when needed.
Data Source
AI summary
Aspects of the disclosure relate to medication calendaring. An example method can include a computing device receiving a first interval of days associated with a medication identifier and a second interval of days associated with the medication identifier, wherein the second interval of days sequentially follows the first interval of days. The computing device can determine a medication time interval, the medication time interval comprising the first interval of days and the second interval of days. The computing device can generate a second instance of the medication identifier. The computing device can calendar the second instance of the medication identifier on an electronic calendar via a sliding generation window, the sliding generation window comprising a consecutive set of days of the days of the medication period.