相关实验视频

Updated: Jul 5, 2025

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Published on: December 6, 2024

在放射学报告中检测语音识别错误的生成大型语言模型

Reuben A Schmidt¹, Jarrel C Y Seah¹, Ke Cao¹

¹From the Department of Medical Imaging, Western Health, Footscray, Australia (R.A.S., L.L., W.L.); Alfred Health, Harrison.ai, Monash University, Clayton, Australia (J.C.Y.S.); Department of Surgery, Western Precinct, University of Melbourne, Melbourne, Australia (K.C., J.Y.); and Department of Surgery, Western Health, Melbourne, Australia (J.Y.).

Radiology. Artificial intelligence

|January 24, 2024

概括

生成型大语言模型 (LLM) 可以检测放射学报告中的语音识别错误. GPT-4显示出高精度,表明在医学成像报告中自动错误检测的潜力.

科学领域:

医疗信息学医疗信息学
人工智能的人工智能
放射学放射学是一门学科.

背景情况:

放射学报告中的语音识别错误可能会影响患者的护理.
手动错误检测是耗时的,容易出现人为错误.

研究的目的:

评估生成型大语言模型 (LLM) 在放射学报告中检测语音识别错误的有效性.
为了比较各种LLM,包括GPT-4,在识别这些错误的性能.

主要方法:

放射科医生对3233份CT和MRI报告的数据集进行了语音识别错误的分析.
五个生成的LLM (GPT-3.5-turbo,GPT-4,文本-davinci-003,Llama-v2-70B-chat,Bard) 进行了测试,以检测错误.
快速工程被用来优化LLM性能,手动检测作为参考标准.

主要成果:

GPT-4实现了高精度,临床显著错误的F1得分为86.9%,非临床显著错误的F1得分为94.3%.
其他LLM的表现各不相同,其中Bard的准确性最低.
长期报告和居民指令等因素与错误率的增加有关.

结论:

先进的生成LLM,特别是GPT-4,在放射学报告中自动检测语音识别错误方面显示出显著的潜力.
这项技术可以提高放射学报告质量保证的准确性和效率.
进一步的研究可能会探索将LLM集成到临床工作流程中,以实时标记错误.

关键词:

这就是为什么CTCTCTCTCTCT 大型语言模型这就是为什么MRI是MRI.机器学习机器学习自然语言处理自然语言处理.放射学报告放射学报告演讲演讲演讲没有监督的学习学习.

更多相关视频

Ultrasound Images of the Tongue: A Tutorial for Assessment and Remediation of Speech Sound Errors

Ultrasound Images of the Tongue: A Tutorial for Assessment and Remediation of Speech Sound Errors

Published on: January 3, 2017

Foreign Accent and Forensic Speaker Identification in Voice Lineups: The Influence of Acoustic Features Based on Prosody

Foreign Accent and Forensic Speaker Identification in Voice Lineups: The Influence of Acoustic Features Based on Prosody

Published on: September 27, 2024

相关概念视频

Leaky Scanning

Leaky Scanning

During most eukaryotic translation processes, the small 40S ribosome subunit scans an mRNA from its 5' end until it encounters the first start AUG codon. The large 60S ribosomal subunit then joins the smaller one to initiate protein synthesis. The location of the translation initiation is largely determined by the nucleotides near the start codon as there may be multiple translation initiation sites present on the mRNA. Marilyn Kozak discovered that the sequence RCCAUGG (where R...

Nonsense-mediated mRNA Decay

Nonsense-mediated mRNA Decay

The Upf proteins that carry out nonsense-mediated decay (NMD) are found in all eukaryotic organisms, including humans. Each protein has an individual role, but they need to work in collaboration. Upf1 is an ATP-dependent RNA helicase that unwinds the RNA helix. Because Upf1 can unwind any RNA, Upf2 and Upf3 are required to help Upf1 discriminate between nonsense and normal mRNAs.
Usually, Upf3 binds to an Exon Junction Complex (EJC) at mRNA splice sites. If a ribosome fully translates the mRNA,...

Positron Emission Tomography

Positron Emission Tomography

Positron emission tomography (PET) is a medical imaging technique involving radiopharmaceuticals — substances that emit short-lived radiation. Although the first PET scanner was introduced in 1961, it took 15 more years before radiopharmaceuticals were combined with the technique and revolutionized its potential.
One of the main requirements of a PET scan is a positron-emitting radioisotope, which is produced in a cyclotron and then attached to a substance used by the part of the body...