知识图增强的变压器用于诊断 中国医学的一代
Xin-Yu Wang1, Tao Yang1,2,3, Xiao-Yuan Gao1
1School of Artificial Intelligence and Information Technology, Nanjing University of Chinese Medicine, Nanjing, 210023, China.
Chinese journal of integrative medicine
|January 14, 2024
概括
这项研究引入了一种用于中医诊断的新型文本生成模型,克服了传统分类方法的局限性. 知识图增强型变压器 (KGET) 模型显示出卓越的性能和广泛的应用前景.
科学领域:
- 人工智能的人工智能
- 自然语言处理自然语言处理.
- 传统中国医药 传统中国医药
背景情况:
- 传统中医 (CM) 诊断现代化面临基于分类的智能模型的挑战,特别是过度或类似类别的问题.
- 自然语言处理 (NLP) 的进步已经成熟了文本生成技术,为CM诊断提供了潜在的替代方法.
- 现有的CM智能诊断模型与诊断信息的细微和复杂性质作斗争.
研究的目的:
- 通过将诊断问题重新定义为文本生成任务,建立一个CM诊断生成模型.
- 为了提高语义上下文特征的学习能力,使用双向长期短期记忆 (BILSTM) 与变压器骨干.
- 将医学知识引入知识图增强型变压器 (KGET) 模型,以提高推断能力.
主要方法:
- 开发了知识图增强变压器 (KGET) 模型,集成BILSTM和医学知识图.
- 在566个CM案例文本上训练和评估KGET,将其与LSTM序列对序列 (LSTM-seq2seq),双向和自动回归变压器 (BART) 以及中国预训练不平衡变压器 (CPT) 模型进行比较.
- 进行了废弃实验,以评估特定模型组件 (BILSTM,先前知识) 对性能的影响.
主要成果:
- 该KGET模型实现了卓越的性能指标:双语评估研究 (BLEU) 的45.85,回忆导向的评估研究1 (ROUGE1) 的73.93,ROUGE2的54.59,和7.12.的编辑距离.
- 在BLEU,ROUGE1和ROUGE2分数中,KGET的表现优于LSTM-seq2seq,BART和CPT模型,同时显示较低的编辑距离.
- 废除研究证实,包括BILSTM和先前的医学知识显著提高了KGET的性能,与实际的CM诊断保持一致.
结论:
- 文本生成技术是CM诊断建模的可行和有效方法,成功地解决了传统分类方法的局限性.
- KGET模型为智能CM诊断提供了一个有前途的解决方案,避免了与类别过载和相似性相关的问题.
- CM诊断文本生成技术对未来医疗AI的进步和传统医学的现代化具有重大潜力.
相关概念视频
01:18Voltmeter
A voltmeter is an electrical device that measures the potential difference or voltage between two points. It is connected in parallel with the circuit element it is measuring. A parallel connection is used because elements in parallel experience the same potential difference. The voltmeter is represented by the symbol "V ".
An ideal voltmeter would have infinite resistance, so connecting it between two points in a circuit would not alter any of the currents. Real voltmeters always have finite...
An ideal voltmeter would have infinite resistance, so connecting it between two points in a circuit would not alter any of the currents. Real voltmeters always have finite...
01:25Instrumentation Amplifier
An electrocardiography (ECG) machine is an essential piece of medical equipment used to monitor the electrical activity of the heart. It operates by detecting small electrical changes on the skin that result from the depolarization of the heart muscle during each heartbeat. However, these signals are in the microvolt range and can be easily overwhelmed by noise or interference.
To overcome this challenge, an ECG machine utilizes an instrumentation amplifier. This specialized amplifier is...
To overcome this challenge, an ECG machine utilizes an instrumentation amplifier. This specialized amplifier is...
01:23Voltage Doubler Circuit
A voltage doubler circuit integrates two main components: a clamping section and a rectifier section. The clamping section consists of a capacitor (C1) and a diode (D1), whereas the rectifier section is equipped with another diode (D2) and capacitor (C2). This circuit produces an output voltage with twice the amplitude of the sinusoidal input voltage.
01:23Instrument Transformers
Instrument transformers, comprising voltage transformers (VTs) and current transformers (CTs), play crucial roles in power substations by providing isolated replicas of current or voltage for measurement and protection purposes. Voltage transformers reduce the primary voltage to levels suitable for relay operation and measurement, while current transformers scale down the primary current. The primary winding of a current transformer often consists of a single turn, achieved by threading the...
01:24Automated Microbial Diagnostics
Automated diagnostic analyzers have transformed clinical microbiology by providing rapid and reliable methods for pathogen identification and antibiotic susceptibility testing. Among these systems, the Vitek 2 is widely used because it automates the traditionally labor-intensive processes of microbial identification (ID) and antibiotic susceptibility testing (AST), delivering standardized and timely results that are essential for effective patient care.Microbial Identification with ID CardsThe...