X-Ray Imaging Prompt Selection for Reliable LLM Output
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing techniques for generating prompts for large language models (LLMs) lack evaluation by third parties, leading to suboptimal outputs due to varying user skills in creating effective prompts.
Innovation Solution
An X-ray image diagnostic apparatus and medical information processing apparatus utilize a prompt database that associates prompts with user instructions, evaluates their conformity and relevance, and displays suitable prompts based on evaluation values, ensuring desired outputs from the generative model.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If users directly input prompts to generative models without third-party evaluation, then the system operation is simple, but the output quality and user satisfaction deteriorate due to varying user skills in creating effective prompts
Solution Approach 1:
The patent introduces a prompt database as an intermediary between users and generative models. The database stores pre-evaluated prompts with high conformity scores, acting as a mediator that translates user intents into optimized prompts without requiring users to have prompt engineering skills. This resolves the contradiction by improving output quality through third-party evaluated prompts while keeping the user interface simple.
Solution Approach 2:
The patent applies preliminary action by pre-evaluating and storing prompts in the database before actual use. Prompts are evaluated for conformity with user intents and stored with their evaluation scores, so that when users need prompts, they can directly select from pre-evaluated options rather than creating and evaluating prompts in real-time. This improves reliability while maintaining operational simplicity.
2Adaptability or versatility
If multiple systems each deploy their own generative models, then each system can operate independently, but hardware resource usage and costs increase significantly
Solution Approach 1:
The patent implements universality by creating a shared prompt database that can be accessed by multiple different systems and generative models. Instead of each system maintaining separate prompt repositories and models, a universal database serves multiple functions across different medical imaging systems, reducing redundant hardware resources while maintaining system independence through network access.
Solution Approach 2:
The patent applies copying by allowing multiple systems to access and use the same prompt database through network connections. Rather than duplicating the entire generative model infrastructure, systems can copy or access prompt data remotely, significantly reducing hardware requirements while maintaining operational autonomy.
Data Source
AI summary
An X-ray image diagnostic apparatus or a medical information processing apparatus according to an embodiment includes an X-ray irradiator that outputs an X-ray and an X-ray detector that detects the X-ray output by the X-ray irradiator and transmitted through a subject. The X-ray image diagnostic apparatus or the medical information processing apparatus that generates an X-ray image based on detection data obtained by detection by the X-ray detector includes processing circuitry. The processing circuitry receives a first instruction, receive an instruction to add the X-ray image, acquires the degree of conformity between the first instruction and a second instruction based on a database in which a plurality of the second instructions and a plurality of prompts are associated with each other, acquires an evaluation value obtained by evaluating the plurality of prompts identified by the acquired degree of conformity, and displays one or a plurality of prompts to which the X-ray image is added selected from the plurality of prompts based on the evaluation value.