当前位置: X-MOL 学术Artif. Intell. Med. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Towards automatic encoding of medical procedures using convolutional neural networks and autoencoders.
Artificial Intelligence in Medicine ( IF 6.1 ) Pub Date : 2018-10-29 , DOI: 10.1016/j.artmed.2018.10.001
Yihan Deng 1 , André Sander 2 , Lukas Faulstich 2 , Kerstin Denecke 1
Affiliation  

Classification systems such as ICD-10 for diagnoses or the Swiss Operation Classification System (CHOP) for procedure classification in the clinical treatment are essential for clinical management and information exchange. Traditionally, classification codes are assigned manually or by systems that rely upon concept-based or rule-based classification methods. Such methods can reach their limit easily due to the restricted coverage of handcrafted rules and of the vocabulary in underlying terminological systems. Conventional machine learning approaches normally depend on selected features within a human annotated training set. However, it is quite laborious to obtain a well labeled data set and its generation can easily be influenced by accumulative errors caused by human factors. To overcome this, we will present our processing pipeline for query matching realized through neural networks within the task of medical procedure classification. The pipeline is built upon convolutional neural networks (CNN) and autoencoder with logistic regression. On the task of relevance determination between query and category text, the autoencoder based method has achieved a micro F1 score of 70.29%, while the convolutional based method has reached a micro F1 score of 60.86% with high efficiency. These two algorithms are compared in experiments with different configurations and baselines (SVM, logistic regression) with respect to their suitability for the task of automatic encoding. Advantages and limitations are discussed.



中文翻译:

使用卷积神经网络和自动编码器实现医疗程序的自动编码。

分类系统,例如用于诊断的ICD-10或用于临床治疗过程分类的瑞士手术分类系统(CHOP),对于临床管理和信息交换至关重要。传统上,分类代码是手动分配的,也可以由依赖于基于概念或基于规则的分类方法的系统分配。由于手工术语和基本术语系统中词汇的覆盖范围有限,因此此类方法很容易达到其极限。常规的机器学习方法通​​常取决于人类注释训练集中的选定功能。但是,获得标记正确的数据集非常费力,并且其生成很容易受到人为因素引起的累积错误的影响。为了克服这个问题 我们将介绍在医疗程序分类任务中通过神经网络实现查询匹配的处理管道。该管道基于卷积神经网络(CNN)和具有Logistic回归的自动编码器。在查询和类别文本之间的相关性确定任务上,基于自动编码器的方法已达到70.29%的微F1分数,而基于卷积的方法已达到60.86%的微F1分数,效率很高。在具有不同配置和基线(SVM,逻辑回归)的实验中,比较了这两种算法在自动编码方面的适用性。讨论了优点和限制。该管道基于卷积神经网络(CNN)和具有Logistic回归的自动编码器。在查询和类别文本之间的相关性确定任务上,基于自动编码器的方法已达到70.29%的微F1分数,而基于卷积的方法已达到60.86%的微F1分数,效率很高。在使用不同配置和基线(SVM,逻辑回归)的实验中,比较了这两种算法在自动编码方面的适用性。讨论了优点和限制。该管道基于卷积神经网络(CNN)和具有Logistic回归的自动编码器。在查询和类别文本之间的相关性确定任务上,基于自动编码器的方法已达到70.29%的微F1分数,而基于卷积的方法已达到60.86%的微F1分数,效率很高。在使用不同配置和基线(SVM,逻辑回归)的实验中,比较了这两种算法在自动编码方面的适用性。讨论了优点和限制。效率高达86%。在具有不同配置和基线(SVM,逻辑回归)的实验中,比较了这两种算法在自动编码方面的适用性。讨论了优点和限制。效率高达86%。在具有不同配置和基线(SVM,逻辑回归)的实验中,比较了这两种算法在自动编码方面的适用性。讨论了优点和限制。

更新日期:2018-10-29
down
wechat
bug