使用合成数据对放射学报告进行上下文意识的句子分类:开发和验证研究.
Tomohiro Kikuchi1, Yosuke Yamagishi2, Kohei Yamamoto1
1Jichi Medical University, 3311-1, Yakushiji, Shimotsuke, Tochigi, JP.
Journal of medical Internet research
|March 1, 2026
概括
这项研究表明,合成数据可以训练人工智能模型来分类日本放射学报告,减少隐私风险和手工工作. 这些模型准确地识别了积极的发现,即使在真实世界的数据验证.
科学领域:
- 放射学中的人工智能
- 医疗数据的自然语言处理.
- 为医疗保健提供综合数据生成
背景情况:
- 放射学报告的自动结构对于数据利用和AI开发至关重要.
- 手动注释是劳动密集型的,真正的临床数据会给隐私带来风险,特别是在日语等非英语语言.
- 合成数据提供了一个保护隐私的替代方案,但它对复杂的临床细微差别的有效性尚未得到证实.
研究的目的:
- 用完全合成的培训管道为日本放射学报告开发一个上下文意识的句子分类模型.
- 在模型开发过程中消除对现实世界的临床数据的依赖.
- 评估模型在各种多机构现实世界报告中的通用性.
主要方法:
- 使用GPT-4.1生成了3,104份合成日本放射学报告,并用GPT-4.1-mini.注释它们.
- 精心调整的轻量级本地LLM (Qwen3,Llama 3.2) 和使用LoRA的日本文本分类模型 (BERT,JMedRoBERTa,ModernBERT).
- 在J-MID的280份真实世界报告中验证了模型,并由放射科医生确认了标签.
主要成果:
- 模型在合成数据上实现了高性能 (准确度:0.938-0.951,宏F1:0.924-0.940).
- 在真实数据上,性能下降了 (准确度:0.783-0.813,宏F1:0.761-0.790),但积极发现的积极预测值 (PPV_1) 仍然很高 (例如,0.952).
- 基于LLM的方法在外部数据集上显示了轻微的解析错误 (0.55%-7.48%).
结论:
- 使用合成数据为日本放射学报告开发上下文感知句子分类模型是可行的.
- 模型成功地捕获了关键的临床术语和识别积极发现的模式,如稳定的PPV_1.1所示.
- 这种合成数据方法减少了注释负担和隐私风险,支持AI在医学成像方面的发展.
相关概念视频
01:30Radiological Investigation I: X-ray and CT
1.4K
Radiological investigations, including X-rays and computed tomography (CT) scans, are critical for diagnosing and evaluating various medical conditions. These imaging techniques provide valuable insights into the body's internal structures, aiding in the detection of abnormalities, assessment of disease progression, and development of treatment strategies. This article delves into two primary radiological investigations, chest X-rays and CT scans, outlining their purpose, procedures, and...
1.4K
01:30Radiological Investigation II: MRI and Ventilation Perfusion Scan
744
Description
Magnetic Resonance Imaging (MRI) and Ventilation Perfusion Scans are two radiological investigations that offer detailed diagnostic images of the body, particularly lung structures.
MRI
MRI uses magnetic fields and radiofrequency signals to distinguish between normal and abnormal tissues. This technology provides a more detailed diagnostic image than CT scans, enabling it to characterize pulmonary nodules, stage bronchogenic carcinoma, and evaluate inflammatory activity in...
Magnetic Resonance Imaging (MRI) and Ventilation Perfusion Scans are two radiological investigations that offer detailed diagnostic images of the body, particularly lung structures.
MRI
MRI uses magnetic fields and radiofrequency signals to distinguish between normal and abnormal tissues. This technology provides a more detailed diagnostic image than CT scans, enabling it to characterize pulmonary nodules, stage bronchogenic carcinoma, and evaluate inflammatory activity in...
744
01:13Radiological Investigation III: Pulmonary Angiogram and PET Scan
535
Radiological investigations are paramount in the diagnosis and management of various pulmonary diseases. Two essential investigations are the Pulmonary Angiogram and the Positron Emission Tomography (PET) Scan.
Pulmonary Angiogram
A Pulmonary Angiogram is an invasive procedure involving injecting a contrast medium through a catheter threaded into the pulmonary artery or the right side of the heart to visualize the pulmonary vasculature. Computed Tomography (CT) scans have mainly replaced this...
Pulmonary Angiogram
A Pulmonary Angiogram is an invasive procedure involving injecting a contrast medium through a catheter threaded into the pulmonary artery or the right side of the heart to visualize the pulmonary vasculature. Computed Tomography (CT) scans have mainly replaced this...
535
01:14Imaging Studies I: CT and MRI
1.0K
Introduction: MRI and CT scans are crucial advancements in medical imaging techniques, playing a vital role in diagnosing conditions related to the gastrointestinal (GI) system. Each scan serves distinct purposes, targets specific areas, and requires unique nursing duties.
Description of the Procedures
Computed Tomography (CT) scan:
Computed Tomography (CT) scans use X-ray technology to generate detailed images of bones, organs, and tissues. During the scan, the patient lies on a moving table...
Description of the Procedures
Computed Tomography (CT) scan:
Computed Tomography (CT) scans use X-ray technology to generate detailed images of bones, organs, and tissues. During the scan, the patient lies on a moving table...
1.0K
01:30Classification of Signals
1.5K
In signal processing, signals are classified based on various characteristics: continuous-time versus discrete-time, periodic versus aperiodic, analog versus digital, and causal versus noncausal. Each category highlights distinct properties crucial for understanding and manipulating signals.
A continuous-time signal holds a value at every instant in time, representing information seamlessly. In contrast, a discrete-time signal holds values only at specific moments, often denoted as x(n), where...
A continuous-time signal holds a value at every instant in time, representing information seamlessly. In contrast, a discrete-time signal holds values only at specific moments, often denoted as x(n), where...
1.5K
01:20Imaging Studies for Cardiovascular System III: X-Ray
541
The most common cardiovascular diagnostic test is an X-ray. It produces images of the heart, blood vessels, and adjacent structures.
Definition and Purpose
An X-ray, or radiograph, is a non-invasive method that uses ionizing radiation to take images of internal structures. It is mainly used in cardiac imaging to examine the heart, lungs, and major blood vessels, aiming to identify abnormalities in the heart's size, shape, and position, such as heart failure, congenital defects, and vascular...
Definition and Purpose
An X-ray, or radiograph, is a non-invasive method that uses ionizing radiation to take images of internal structures. It is mainly used in cardiac imaging to examine the heart, lungs, and major blood vessels, aiming to identify abnormalities in the heart's size, shape, and position, such as heart failure, congenital defects, and vascular...
541