LLM-Guided Charged-Particle Microscope Operation With Sanity Checks

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Charged-particle microscopes are operationally complex, requiring extensive specialized training, which prevents users from efficiently interacting with them and poses a high barrier-to-entry.

Innovation Solution

Implementing large language models (LLMs) to assist in charged-particle microscope operation by allowing users to interact with intuitive natural language commands, providing real-time image-or-spectrum-conditioned responses for command sanity checks, workflow tutorials, malfunction diagnoses, specimen queries, and GUI customization.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If traditional charged-particle microscope operation interfaces are used, then operational control is precise and reliable, but the device complexity and training requirements increase significantly

Engineering Contradiction:
Improveease of operationVSAvoidoperational complexity
Core Design Contradiction:
Ease of operationVSDevice complexity

Solution Approach 1:

The patent introduces a large language model as an intermediary layer between the user and the charged-particle microscope control system. The LLM translates natural language user intents into precise microscope control commands, maintaining operational reliability while eliminating the need for users to directly navigate complex microscope interfaces and protocols

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent replaces traditional mechanical interaction methods (physical controls, complex software interfaces, manual parameter adjustment) with an AI-based natural language processing system. This substitution allows users to operate the microscope through conversational commands rather than manipulating complex physical or software controls

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Reliability

If comprehensive microscopy training is provided, then operational reliability and precision are improved, but the time and resource investment increases

Engineering Contradiction:
Improveoperational reliabilityVSAvoidtraining time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The system enables users to perform complex microscopy operations without requiring extensive prior training. The large language model provides real-time guidance, automatically adjusts parameters, and executes sophisticated workflows based on simple user requests, allowing the system to serve itself and the user without external expert intervention

Inventive Principle:
Principle #25Self-service

3Measurement precision

If detailed microscopy protocols are implemented, then measurement precision and quality are improved, but the operational complexity increases

Engineering Contradiction:
Improvemicroscopy qualityVSAvoidprotocol complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system has pre-loaded with extensive microscopy knowledge, protocols, and best practices during its training phase. When a user requests an operation, the LLM retrieves and applies the appropriate pre-learned protocols automatically, ensuring high measurement precision without requiring the user to manually configure complex parameters or understand detailed procedures

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS20250348670A1Large language model assistance for charged-particle microscope operation
Publication Date: 2025.11.13 FEI CO
  • US20250348670A1 patent drawing
  • US20250348670A1 patent drawing
  • US20250348670A1 patent drawing

AI summary

Systems/techniques are provided for facilitating large language model assistance for charged-particle microscope operation. In various embodiments, a system can access a natural language instruction associated with a charged-particle microscope, where the natural language instruction can request that the charged-particle microscope undergo a configurable settings adjustment or perform an automated task. In various aspects, the system can cause, in response to the natural language instruction, the charged-particle microscope to capture, according to a default microscopy protocol, an image or an energy spectrum of a specimen that is currently loaded on a stage of the charged-particle microscope. In various instances, the system can execute a large language model on both the natural language instruction and the image or energy spectrum of the specimen, thereby yielding a natural language response that indicates how implementing the natural language instruction would affect the specimen.