International Statistical Classification of Diseases and Related Health Problems
PulseAugur coverage of International Statistical Classification of Diseases and Related Health Problems — every cluster mentioning International Statistical Classification of Diseases and Related Health Problems across labs, papers, and developer communities, ranked by signal.
2 天有情绪数据
-
LLM以86.6%的准确率自动化精神疾病诊断分类
研究人员开发了一个自动化系统,利用自然语言处理(NLP)和机器学习(ML)对精神疾病诊断进行分类。该研究在一个包含超过145,000份西班牙语精神病描述的数据集上,评估了包括e5_large、BioLORD和Llama-3-8B等经典模型和大型语言模型(LLM)在内的各种文本表示方法。研究结果表明,基于Transformer的嵌入方法显著优于传统方法,经过微调的e5_large模型达到了0.866的最高F1分数。这项工作强调了将LLM…
-
多版本训练提高罕见ICD编码预测准确性
研究人员开发了一种多版本训练方法,以提高自动化临床编码的准确性,特别是对于罕见的医疗编码。通过整合不同版本的国际疾病分类(ICD)数据,如ICD-9和ICD-10,该模型表现出显著的性能提升。该方法解决了编码系统不断演变以及罕见编码预测中的长尾问题,从而以更少的模型参数实现了整体指标的改善。
-
LLMs trained with Span-Centric Learning improve ICD coding accuracy and efficiency
Researchers have developed a new training framework called Span-Centric Learning (SCL) to improve the accuracy of Large Language Models (LLMs) in assigning International Classification of Diseases (ICD) codes to clinica…