Enhanced character-level deep convolutional neural networks for cardiovascular disease prediction
Abstract Background Electronic medical records contain a variety of valuable medical information for patients. So, when we are able to recognize and extract risk factors for disease from EMRs of patients with cardiovascular disease (CVD), and are able to use them to predict CVD, we have the ability...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2020-07-01
|
| Series: | BMC Medical Informatics and Decision Making |
| Subjects: | |
| Online Access: | http://link.springer.com/article/10.1186/s12911-020-1118-z |
| _version_ | 1818491517564616704 |
|---|---|
| author | Zhichang Zhang Yanlong Qiu Xiaoli Yang Minyu Zhang |
| author_facet | Zhichang Zhang Yanlong Qiu Xiaoli Yang Minyu Zhang |
| author_sort | Zhichang Zhang |
| collection | DOAJ |
| description | Abstract Background Electronic medical records contain a variety of valuable medical information for patients. So, when we are able to recognize and extract risk factors for disease from EMRs of patients with cardiovascular disease (CVD), and are able to use them to predict CVD, we have the ability to automatically process clinical texts, resulting in an improved accuracy of supporting doctors for the clinical diagnosis of CVD. In the case where CVD is becoming more worldwide, predictive CVD based on EMRs has been studied by many researchers to address this important aspect of improving diagnostic efficiency. Methods This paper proposes an Enhanced Character-level Deep Convolutional Neural Networks (EnDCNN) model for cardiovascular disease prediction. Results On the manually annotated Chinese EMRs corpus, our risk factor identification extraction model achieved 0.9073 of F-score, our prediction model achieved 0.9516 of F-score, and the prediction result is better than the most previous methods. Conclusions The character-level model based on text region embedding can well map risk factors and their labels as a unit into a vector, and downsampling plays a crucial role in improving the training efficiency of deep CNN. What’s more, the shortcut connections with pre-activation used in our model architecture implements dimension-matching free in training. |
| first_indexed | 2024-12-10T17:32:05Z |
| format | Article |
| id | doaj.art-9a56492b3ccf4f5aac5d5f1fc532f96b |
| institution | Directory Open Access Journal |
| issn | 1472-6947 |
| language | English |
| last_indexed | 2024-12-10T17:32:05Z |
| publishDate | 2020-07-01 |
| publisher | BMC |
| record_format | Article |
| series | BMC Medical Informatics and Decision Making |
| spelling | doaj.art-9a56492b3ccf4f5aac5d5f1fc532f96b2022-12-22T01:39:39ZengBMCBMC Medical Informatics and Decision Making1472-69472020-07-0120S311010.1186/s12911-020-1118-zEnhanced character-level deep convolutional neural networks for cardiovascular disease predictionZhichang Zhang0Yanlong Qiu1Xiaoli Yang2Minyu Zhang3College of Computer Science and Engineering, Northwest Normal UniversityCollege of Computer Science and Engineering, Northwest Normal UniversityCollege of Computer Science and Engineering, Northwest Normal UniversityCollege of Computer Science and Engineering, Northwest Normal UniversityAbstract Background Electronic medical records contain a variety of valuable medical information for patients. So, when we are able to recognize and extract risk factors for disease from EMRs of patients with cardiovascular disease (CVD), and are able to use them to predict CVD, we have the ability to automatically process clinical texts, resulting in an improved accuracy of supporting doctors for the clinical diagnosis of CVD. In the case where CVD is becoming more worldwide, predictive CVD based on EMRs has been studied by many researchers to address this important aspect of improving diagnostic efficiency. Methods This paper proposes an Enhanced Character-level Deep Convolutional Neural Networks (EnDCNN) model for cardiovascular disease prediction. Results On the manually annotated Chinese EMRs corpus, our risk factor identification extraction model achieved 0.9073 of F-score, our prediction model achieved 0.9516 of F-score, and the prediction result is better than the most previous methods. Conclusions The character-level model based on text region embedding can well map risk factors and their labels as a unit into a vector, and downsampling plays a crucial role in improving the training efficiency of deep CNN. What’s more, the shortcut connections with pre-activation used in our model architecture implements dimension-matching free in training.http://link.springer.com/article/10.1186/s12911-020-1118-zChinese electronic medical recordCVD risk factors extractionCVD predictionDownsamplingPre-activationDimension-matching free |
| spellingShingle | Zhichang Zhang Yanlong Qiu Xiaoli Yang Minyu Zhang Enhanced character-level deep convolutional neural networks for cardiovascular disease prediction BMC Medical Informatics and Decision Making Chinese electronic medical record CVD risk factors extraction CVD prediction Downsampling Pre-activation Dimension-matching free |
| title | Enhanced character-level deep convolutional neural networks for cardiovascular disease prediction |
| title_full | Enhanced character-level deep convolutional neural networks for cardiovascular disease prediction |
| title_fullStr | Enhanced character-level deep convolutional neural networks for cardiovascular disease prediction |
| title_full_unstemmed | Enhanced character-level deep convolutional neural networks for cardiovascular disease prediction |
| title_short | Enhanced character-level deep convolutional neural networks for cardiovascular disease prediction |
| title_sort | enhanced character level deep convolutional neural networks for cardiovascular disease prediction |
| topic | Chinese electronic medical record CVD risk factors extraction CVD prediction Downsampling Pre-activation Dimension-matching free |
| url | http://link.springer.com/article/10.1186/s12911-020-1118-z |
| work_keys_str_mv | AT zhichangzhang enhancedcharacterleveldeepconvolutionalneuralnetworksforcardiovasculardiseaseprediction AT yanlongqiu enhancedcharacterleveldeepconvolutionalneuralnetworksforcardiovasculardiseaseprediction AT xiaoliyang enhancedcharacterleveldeepconvolutionalneuralnetworksforcardiovasculardiseaseprediction AT minyuzhang enhancedcharacterleveldeepconvolutionalneuralnetworksforcardiovasculardiseaseprediction |