Ordinal convolutional neural networks for predicting RDoC positive valence psychiatric symptom severity scores.
Ordinal convolutional neural networks for predicting RDoC positive valence psychiatric symptom severity scores.
复制标题
DOI：
10.1016/j.jbi.2017.05.008
复制
发表时间：
2017-11
影响因子：
4.5
通讯作者：
Kavuluru R
中科院分区：
文献类型：
--
作者：
Rios A;Kavuluru R
CEGS N-GRID 2016临床自然语言处理（NLP）共享任务为参与者提供了一组1000个神经精神病学笔记，作为预测精神症状严重程度评分的竞赛的一部分。本文总结了我们参与共享任务第二轨道的方法、结果和经验。经典的文本分类方法通常分为三种问题类型：二值分类、多类分类和多标签分类。在这项工作中，我们研究了文本数据的有序回归问题，其中错误分类会根据基础事实和模型预测在有序尺度上的距离远近而受到不同的惩罚。具体来说，我们展示了我们在N-GRID共享任务中预测研究领域标准（RDoC）阳性效价顺序症状严重程度评分（缺失、轻度、中度和重度）中的条目（方法和结果）。我们提出了一种新颖的卷积神经网络（CNN）模型，用于处理精神病学记录的有序回归任务。总的来说，我们的模型将有序损失函数、CNN和传统特征工程（宽特征）结合到一个端到端学习的单一模型中。考虑到可解释性是非线性模型的一个重要问题，我们采用了一种称为局部可解释模型不可知论解释（LIME）的新方法来识别导致实例特定预测的重要单词。我们进入共享任务的最佳模型在24个团队中排名第三，基于宏观平均绝对误差（MMAE）的归一化得分（100·（1−MMAE））为83.86。自比赛以来，我们将分数（使用基本的合奏）提高到85.55，与获胜的共享任务条目相当。将LIME应用于模型预测，我们通过识别导致特定决策的单词来证明实例特定预测解释的可行性。在本文中，我们提出了一种方法，该方法成功地将宽特征和有序损失函数应用于卷积神经网络，用于有序文本分类，特别是用于预测精神症状严重程度评分。我们的方法在N-GRID共享任务上具有出色的性能，并且可以使用现有的模型不可知方法进行解释。
The CEGS N-GRID 2016 Shared Task in Clinical Natural Language Processing (NLP) provided a set of 1000 neuropsychiatric notes to participants as part of a competition to predict psychiatric symptom severity scores. This paper summarizes our methods, results, and experiences based on our participation in the second track of the shared task. Classical methods of text classification usually fall into one of three problem types: binary, multi-class, and multi-label classification. In this effort, we study ordinal regression problems with text data where misclassifications are penalized differently based on how far apart the ground truth and model predictions are on the ordinal scale. Specifically, we present our entries (methods and results) in the N-GRID shared task in predicting research domain criteria (RDoC) positive valence ordinal symptom severity scores (absent, mild, moderate, and severe) from psychiatric notes. We propose a novel convolutional neural network (CNN) model designed to handle ordinal regression tasks on psychiatric notes. Broadly speaking, our model combines an ordinal loss function, a CNN, and conventional feature engineering (wide features) into a single model which is learned end-to-end. Given interpretability is an important concern with nonlinear models, we apply a recent approach called locally interpretable model-agnostic explanation (LIME) to identify important words that lead to instance specific predictions. Our best model entered into the shared task placed third among 24 teams and scored a macro mean absolute error (MMAE) based normalized score (100 · (1 − M M AE)) of 83.86. Since the competition, we improved our score (using basic ensembling) to 85.55, comparable with the winning shared task entry. Applying LIME to model predictions, we demonstrate the feasibility of instance specific prediction interpretation by identifying words that led to a particular decision. In this paper, we present a method that successfully uses wide features and an ordinal loss function applied to convolutional neural networks for ordinal text classification specifically in predicting psychiatric symptom severity scores. Our approach leads to excellent performance on the N-GRID shared task and is also amenable to interpretability using existing model-agnostic approaches.
登录
查看更多内容
影响因子：
4.5
作者：
Filannino M;Stubbs A;Uzuner Ö
通讯作者：
Uzuner Ö
影响因子：
--
作者：
Bush, K;Kivlahan, DR;Bradley, KA
通讯作者：
Bradley, KA
在线自杀观察论坛上有用评论的分类。
DOI：
10.1145/2975167.2975170
发表时间：
2016-10
期刊：
ACM-BCB ... ... : the ... ACM Conference on Bioinformatics, Computational Biology and Biomedicine. ACM Conference on Bioinformatics, Computational Biology and Biomedicine
影响因子：
--
作者：
Kavuluru R;Williams AG;Ramos-Morales M;Haye L;Holaday T;Cerel J
通讯作者：
Cerel J
DOI：
10.1007/978-3-319-47874-6_22
发表时间：
2016-11
期刊：
Social informatics : 8th International Conference, SocInfo 2016, Bellevue, WA, USA, November 11-14, 2016, Proceedings. Part II. SocInfo (Conference) (8th : 2016 : Bellevue, Wash.)
影响因子：
--
作者：
Han S;Kavuluru R
通讯作者：
Kavuluru R
影响因子：
73.3
作者：
Cuthbert, Bruce N.
通讯作者：
Cuthbert, Bruce N.