Dual Model Medical Invoices Recognition
Hospitals need to invest a lot of manpower to manually input the contents of medical invoices (nearly 300,000,000 medical invoices a year) into the medical system. In order to help the hospital save money and stabilize work efficiency, this paper designed a system to complete the complicated work us...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2019-10-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/19/20/4370 |
| _version_ | 1818008185348292608 |
|---|---|
| author | Fei Yi Yi-Fei Zhao Guan-Qun Sheng Kai Xie Chang Wen Xin-Gong Tang Xuan Qi |
| author_facet | Fei Yi Yi-Fei Zhao Guan-Qun Sheng Kai Xie Chang Wen Xin-Gong Tang Xuan Qi |
| author_sort | Fei Yi |
| collection | DOAJ |
| description | Hospitals need to invest a lot of manpower to manually input the contents of medical invoices (nearly 300,000,000 medical invoices a year) into the medical system. In order to help the hospital save money and stabilize work efficiency, this paper designed a system to complete the complicated work using a Gaussian blur and smoothing–convolutional neural network combined with a recurrent neural network (GBS-CR) method. Gaussian blur and smoothing (GBS) is a novel preprocessing method that can fix the breakpoint font in medical invoices. The combination of convolutional neural network (CNN) and recurrent neural network (RNN) was used to raise the recognition rate of the breakpoint font in medical invoices. RNN was designed to be the semantic revision module. In the aspect of image preprocessing, Gaussian blur and smoothing were used to fix the breakpoint font. In the period of making the self-built dataset, a certain proportion of the breakpoint font (the font of breakpoint is 3, the original font is 7) was added, in this paper, so as to optimize the Alexnet–Adam–CNN (AA-CNN) model, which is more suitable for the recognition of the breakpoint font than the traditional CNN model. In terms of the identification methods, we not only adopted the optimized AA-CNN for identification, but also combined RNN to carry out the semantic revisions of the identified results of CNN, meanwhile further improving the recognition rate of the medical invoices. The experimental results show that compared with the state-of-art invoice recognition method, the method presented in this paper has an average increase of 10 to 15 percentage points in recognition rate. |
| first_indexed | 2024-04-14T05:25:56Z |
| format | Article |
| id | doaj.art-0a711fa7e7ca4403b033ba393b7e4051 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-04-14T05:25:56Z |
| publishDate | 2019-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-0a711fa7e7ca4403b033ba393b7e40512022-12-22T02:09:59ZengMDPI AGSensors1424-82202019-10-011920437010.3390/s19204370s19204370Dual Model Medical Invoices RecognitionFei Yi0Yi-Fei Zhao1Guan-Qun Sheng2Kai Xie3Chang Wen4Xin-Gong Tang5Xuan Qi6Key Laboratory of Exploration Technologies for Oil and Gas Resources, Yangtze University, Ministry of Education, Wuhan 430100, ChinaSchool of Electronic Information, Yangtze University, Jingzhou 434023, ChinaKey Laboratory of Exploration Technologies for Oil and Gas Resources, Yangtze University, Ministry of Education, Wuhan 430100, ChinaSchool of Electronic Information, Yangtze University, Jingzhou 434023, ChinaSchool of Computer Science, Yangtze University, Jingzhou 434023, ChinaKey Laboratory of Exploration Technologies for Oil and Gas Resources, Yangtze University, Ministry of Education, Wuhan 430100, ChinaSchool of Petroleum Engineering, China University of Petroleum, Beijing 102249, ChinaHospitals need to invest a lot of manpower to manually input the contents of medical invoices (nearly 300,000,000 medical invoices a year) into the medical system. In order to help the hospital save money and stabilize work efficiency, this paper designed a system to complete the complicated work using a Gaussian blur and smoothing–convolutional neural network combined with a recurrent neural network (GBS-CR) method. Gaussian blur and smoothing (GBS) is a novel preprocessing method that can fix the breakpoint font in medical invoices. The combination of convolutional neural network (CNN) and recurrent neural network (RNN) was used to raise the recognition rate of the breakpoint font in medical invoices. RNN was designed to be the semantic revision module. In the aspect of image preprocessing, Gaussian blur and smoothing were used to fix the breakpoint font. In the period of making the self-built dataset, a certain proportion of the breakpoint font (the font of breakpoint is 3, the original font is 7) was added, in this paper, so as to optimize the Alexnet–Adam–CNN (AA-CNN) model, which is more suitable for the recognition of the breakpoint font than the traditional CNN model. In terms of the identification methods, we not only adopted the optimized AA-CNN for identification, but also combined RNN to carry out the semantic revisions of the identified results of CNN, meanwhile further improving the recognition rate of the medical invoices. The experimental results show that compared with the state-of-art invoice recognition method, the method presented in this paper has an average increase of 10 to 15 percentage points in recognition rate.https://www.mdpi.com/1424-8220/19/20/4370medical invoicesbreakpoint fontcnnrnnsemantic revisions |
| spellingShingle | Fei Yi Yi-Fei Zhao Guan-Qun Sheng Kai Xie Chang Wen Xin-Gong Tang Xuan Qi Dual Model Medical Invoices Recognition Sensors medical invoices breakpoint font cnn rnn semantic revisions |
| title | Dual Model Medical Invoices Recognition |
| title_full | Dual Model Medical Invoices Recognition |
| title_fullStr | Dual Model Medical Invoices Recognition |
| title_full_unstemmed | Dual Model Medical Invoices Recognition |
| title_short | Dual Model Medical Invoices Recognition |
| title_sort | dual model medical invoices recognition |
| topic | medical invoices breakpoint font cnn rnn semantic revisions |
| url | https://www.mdpi.com/1424-8220/19/20/4370 |
| work_keys_str_mv | AT feiyi dualmodelmedicalinvoicesrecognition AT yifeizhao dualmodelmedicalinvoicesrecognition AT guanqunsheng dualmodelmedicalinvoicesrecognition AT kaixie dualmodelmedicalinvoicesrecognition AT changwen dualmodelmedicalinvoicesrecognition AT xingongtang dualmodelmedicalinvoicesrecognition AT xuanqi dualmodelmedicalinvoicesrecognition |