Attention based automated radiology report generation using CNN and LSTM.
The automated generation of radiology reports provides X-rays and has tremendous potential to enhance the clinical diagnosis of diseases in patients. A new research direction is gaining increasing attention that involves the use of hybrid approaches based on natural language processing and computer...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Public Library of Science (PLoS)
2022-01-01
|
| Series: | PLoS ONE |
| Online Access: | https://doi.org/10.1371/journal.pone.0262209 |
| _version_ | 1819319668443185152 |
|---|---|
| author | Mehreen Sirshar Muhammad Faheem Khalil Paracha Muhammad Usman Akram Norah Saleh Alghamdi Syeda Zainab Yousuf Zaidi Tatheer Fatima |
| author_facet | Mehreen Sirshar Muhammad Faheem Khalil Paracha Muhammad Usman Akram Norah Saleh Alghamdi Syeda Zainab Yousuf Zaidi Tatheer Fatima |
| author_sort | Mehreen Sirshar |
| collection | DOAJ |
| description | The automated generation of radiology reports provides X-rays and has tremendous potential to enhance the clinical diagnosis of diseases in patients. A new research direction is gaining increasing attention that involves the use of hybrid approaches based on natural language processing and computer vision techniques to create auto medical report generation systems. The auto report generator, producing radiology reports, will significantly reduce the burden on doctors and assist them in writing manual reports. Because the sensitivity of chest X-ray (CXR) findings provided by existing techniques not adequately accurate, producing comprehensive explanations for medical photographs remains a difficult task. A novel approach to address this issue was proposed, based on the continuous integration of convolutional neural networks and long short-term memory for detecting diseases, followed by the attention mechanism for sequence generation based on these diseases. Experimental results obtained by using the Indiana University CXR and MIMIC-CXR datasets showed that the proposed model attained the current state-of-the-art efficiency as opposed to other solutions of the baseline. BLEU-1, BLEU-2, BLEU-3, and BLEU-4 were used as the evaluation metrics. |
| first_indexed | 2024-12-24T11:07:21Z |
| format | Article |
| id | doaj.art-89d5695ce9f8451e84ff5183b96bf133 |
| institution | Directory Open Access Journal |
| issn | 1932-6203 |
| language | English |
| last_indexed | 2024-12-24T11:07:21Z |
| publishDate | 2022-01-01 |
| publisher | Public Library of Science (PLoS) |
| record_format | Article |
| series | PLoS ONE |
| spelling | doaj.art-89d5695ce9f8451e84ff5183b96bf1332022-12-21T16:58:36ZengPublic Library of Science (PLoS)PLoS ONE1932-62032022-01-01171e026220910.1371/journal.pone.0262209Attention based automated radiology report generation using CNN and LSTM.Mehreen SirsharMuhammad Faheem Khalil ParachaMuhammad Usman AkramNorah Saleh AlghamdiSyeda Zainab Yousuf ZaidiTatheer FatimaThe automated generation of radiology reports provides X-rays and has tremendous potential to enhance the clinical diagnosis of diseases in patients. A new research direction is gaining increasing attention that involves the use of hybrid approaches based on natural language processing and computer vision techniques to create auto medical report generation systems. The auto report generator, producing radiology reports, will significantly reduce the burden on doctors and assist them in writing manual reports. Because the sensitivity of chest X-ray (CXR) findings provided by existing techniques not adequately accurate, producing comprehensive explanations for medical photographs remains a difficult task. A novel approach to address this issue was proposed, based on the continuous integration of convolutional neural networks and long short-term memory for detecting diseases, followed by the attention mechanism for sequence generation based on these diseases. Experimental results obtained by using the Indiana University CXR and MIMIC-CXR datasets showed that the proposed model attained the current state-of-the-art efficiency as opposed to other solutions of the baseline. BLEU-1, BLEU-2, BLEU-3, and BLEU-4 were used as the evaluation metrics.https://doi.org/10.1371/journal.pone.0262209 |
| spellingShingle | Mehreen Sirshar Muhammad Faheem Khalil Paracha Muhammad Usman Akram Norah Saleh Alghamdi Syeda Zainab Yousuf Zaidi Tatheer Fatima Attention based automated radiology report generation using CNN and LSTM. PLoS ONE |
| title | Attention based automated radiology report generation using CNN and LSTM. |
| title_full | Attention based automated radiology report generation using CNN and LSTM. |
| title_fullStr | Attention based automated radiology report generation using CNN and LSTM. |
| title_full_unstemmed | Attention based automated radiology report generation using CNN and LSTM. |
| title_short | Attention based automated radiology report generation using CNN and LSTM. |
| title_sort | attention based automated radiology report generation using cnn and lstm |
| url | https://doi.org/10.1371/journal.pone.0262209 |
| work_keys_str_mv | AT mehreensirshar attentionbasedautomatedradiologyreportgenerationusingcnnandlstm AT muhammadfaheemkhalilparacha attentionbasedautomatedradiologyreportgenerationusingcnnandlstm AT muhammadusmanakram attentionbasedautomatedradiologyreportgenerationusingcnnandlstm AT norahsalehalghamdi attentionbasedautomatedradiologyreportgenerationusingcnnandlstm AT syedazainabyousufzaidi attentionbasedautomatedradiologyreportgenerationusingcnnandlstm AT tatheerfatima attentionbasedautomatedradiologyreportgenerationusingcnnandlstm |