自动测量:在线电学研究的LLM辅助科学质量控制
José Guilherme de Almeida1, Nickolas Papanikolaou2
1Champalimaud Foundation, Lisbon, Portugal.
European journal of radiology
|August 14, 2025
概括
大型语言模型 (LLM) 在评估放射学研究质量方面表现有前途,达成与人类放射学家相比的协议. 保护隐私的开放式LLM与商业模式类似,支持研究完整性.
科学领域:
- 医疗成像医学成像
- 人工智能的人工智能
- 无线电学 (Radiomics) 是一种辐射学.
背景情况:
- 放射学研究质量对于临床翻译至关重要,但往往存在方法上的缺陷.
- 为了确保研究的完整性,需要使用标准化的评估工具,如METhodological RadiomICs Score (METRICS).
研究的目的:
- 在使用METRICS框架评估放射学方法质量的过程中,评估大型语言模型 (LLM) 的可靠性.
- 在可重复性设置中,将LLM的表现与人类放射科医生进行比较.
主要方法:
- 一个商业LLM (Gemini Flash 2.0) 和24个开放的LLM被用来评估46篇放射学文章的METRICS分数.
- 在两个可重复性研究 (ADA2025和K2025) 中,评估与放射科医生的评估进行了比较.
- 分析了评价者之间的可靠性 (科恩的卡帕) 和得分相关性 (皮尔森的相关性).
主要成果:
- 商业LLM证明了与人类放射科医生之间的评估者之间的协议,与这两项研究中的人与人之间的协议相当 (平均卡帕≈0.48-0.58).
- 在LLM和人类评分器之间的METRICS得分相关性也与人与人相关性相当 (平均PC ≈0.62-0.68).
- 一个开放的,保护隐私的LLM (Phi4-Reasoning) 与商业的LLM相提并论.
结论:
- 法律学可以有效地协助对放射学研究的标准化质量评估.
- 开源,保护隐私的LLM为商业模型提供了可行的替代方案,以支持放射学研究完整性评估.
更多相关视频
07:15Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
Published on: August 16, 2020
6.9K
13:01Industrialized, Artificial Intelligence-guided Laser Microdissection for Microscaled Proteomic Analysis of the Tumor Microenvironment
Published on: June 3, 2022
3.9K
相关概念视频
Instrument Calibration
Instrument calibration is essential for ensuring that instruments produce accurate and consistent results. It is vital in manufacturing, healthcare, testing laboratories, and scientific research. Calibration processes are specific to each instrument and help enhance data accuracy. Each instrument has a unique calibration process tailored to its design and function to improve data accuracy.
Analytical Balance Calibration
An analytical balance measures mass and requires regular calibration to...
Analytical Balance Calibration
An analytical balance measures mass and requires regular calibration to...
Controlled-Current Coulometry: Overview
Controlled current coulometry, also known as amperostatic coulometry, is a technique used in electrochemical analysis to measure the quantity of a substance through the controlled passage of current. It involves the application of a constant current to an electrochemical cell containing the analyte of interest. As the current flows through the cell, the analyte undergoes a redox reaction at the electrode surface, resulting in a charge transfer. By monitoring the time required for a certain...
Control Systems
Control systems are everywhere in contemporary society, influencing diverse applications from aerospace to automated manufacturing. These systems can be found naturally within biological processes, such as blood sugar regulation and heart rate adjustment in response to stress, as well as in man-made systems like elevators and automated vehicles. A control system is essentially a network of subsystems and processes that collaboratively convert specific inputs into desired outputs.
At the heart...
At the heart...
Electronic Distance Measuring Instruments
Electronic Distance Measuring Instruments (EDMs) are essential tools in modern surveying, offering precise distance measurements by emitting electromagnetic signals and calculating the time required for these signals to travel to a target and return. Two primary types of signals are used in EDMs — light waves and microwaves — each suited to specific environmental and distance requirements. Light-wave-based EDMs utilize either infrared or laser light, providing high accuracy over short distances...
Leveling Equipment
As leveling involves measuring vertical distances relative to a horizontal line of sight, it requires a graduated rod, called a level rod, for vertical measurements and an instrument called a level for a horizontal sight line. A level includes a high-powered telescope with a mechanism for leveling to ensure the line of sight is horizontal when the bubble in the spirit level is centered. Leveling rods, made of wood, metal, or fiberglass, are graduated in feet or meters and commonly used in two-...
