DeepDRG: Performance of Artificial Intelligence Model for Real-Time Prediction of Diagnosis-Related Groups.
DeepDRG: Performance of Artificial Intelligence Model for Real-Time Prediction of Diagnosis-Related Groups.
复制标题
DOI：
10.3390/healthcare9121632
复制
发表时间：
2021-11-25
期刊：
影响因子：
--
通讯作者：
Li YJ
中科院分区：
文献类型：
--
作者：
Islam MM;Li GH;Poly TN;Li YJ
如今，使用诊断相关组（DRGs）已经增加了对住院治疗的报销。使用DRGs的总体受益取决于临床编码的准确性，以获得合理的报销。然而，选择适当的代码总是具有挑战性的，需要专业知识。由于工作量大，文件质量差，缺乏计算机辅助，错误的DRG率总是很高。因此，我们开发了深度学习（DL）模型来预测主要诊断，以获得适当的报销并提高医院绩效。使用由81，486名患者（128，105次发作）组成的数据集进行模型训练和测试。患者的年龄、性别、药物、疾病、实验室检查、程序和手术史被用作我们的多类预测模型的输入。门控递归单元（GRU）和人工神经网络（ANN）模型预测200个主要诊断。DL模型的性能通过受试者工作曲线下面积、精确度、召回率和F1评分来衡量。在两种DL模型中，GRU方法在预测初步诊断方面具有最佳性能（AUC：0.99，精确度：83.2%，召回率：66.0%）。然而，ANN模型用于DRG预测的性能达到AUC为0.99，精确度为0.82，召回率为0.57。我们的研究结果表明，DL算法，特别是GRU，可以用来开发DRGs预测模型，以准确地确定主要诊断。DeepDRGs将有助于获得适当的经济激励，使医疗资源得到合理利用，并提高医院绩效。
Nowadays, the use of diagnosis-related groups (DRGs) has been increased to claim reimbursement for inpatient care. The overall benefits of using DRGs depend upon the accuracy of clinical coding to obtain reasonable reimbursement. However, the selection of appropriate codes is always challenging and requires professional expertise. The rate of incorrect DRGs is always high due to the heavy workload, poor quality of documentation, and lack of computer assistance. We therefore developed deep learning (DL) models to predict the primary diagnosis for appropriate reimbursement and improving hospital performance. A dataset consisting of 81,486 patients with 128,105 episodes was used for model training and testing. Patients’ age, sex, drugs, diseases, laboratory tests, procedures, and operation history were used as inputs to our multiclass prediction model. Gated recurrent unit (GRU) and artificial neural network (ANN) models were developed to predict 200 primary diagnoses. The performance of the DL models was measured by the area under the receiver operating curve, precision, recall, and F1 score. Of the two DL models, the GRU method, had the best performance in predicting the primary diagnosis (AUC: 0.99, precision: 83.2%, and recall: 66.0%). However, the performance of ANN model for DRGs prediction achieved AUC of 0.99 with a precision of 0.82 and recall of 0.57. The findings of our study show that DL algorithms, especially GRU, can be used to develop DRGs prediction models for identifying primary diagnosis accurately. DeepDRGs would help to claim appropriate financial incentives, enable proper utilization of medical resources, and improve hospital performance.
登录
查看更多内容
通过电子健康记录中的最小特征来预测肝细胞癌：开发深度学习模型。
影响因子：
2.8
作者：
Liang CW;Yang HC;Islam MM;Nguyen PAA;Feng YT;Hou ZY;Huang CW;Poly TN;Li YJ
通讯作者：
Li YJ
影响因子：
6.1
作者：
Atique, Suleman;Hsieh, Chung-Ho;Syed-Abdul, Shabbir
通讯作者：
Syed-Abdul, Shabbir
影响因子：
3.4
作者：
Colla, Carrie H.;Morden, Nancy E.;Rosenthal, Meredith B.
通讯作者：
Rosenthal, Meredith B.
影响因子：
1.6
作者：
Chen KC;Iqbal U;Nguyen PA;Hsu CH;Huang CL;Hsu YE;Atique S;Islam MM;Li YJ;Jian WS
通讯作者：
Jian WS
影响因子：
13.8
作者：
Kutz, Alexander;Gut, Lara;Mueller, Beat
通讯作者：
Mueller, Beat