Using Artificial Intelligence to Establish Chest X-Ray Image Recognition Model to Assist Crucial Diagnosis in Elder Patients With Dyspnea
Pneumonia and pulmonary edema are the most common causes of acute respiratory failure in emergency and intensive care. Airway maintenance and heart function preservation are two foundations for resuscitation. Laboratory examinations have been utilized for clinicians to early differentiate pneumonia...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2022-06-01
|
| Series: | Frontiers in Medicine |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fmed.2022.893208/full |
| _version_ | 1818232381240246272 |
|---|---|
| author | Liu Liong-Rung Liu Liong-Rung Liu Liong-Rung Chiu Hung-Wen Chiu Hung-Wen Huang Ming-Yuan Huang Ming-Yuan Huang Shu-Tien Huang Shu-Tien Huang Shu-Tien Tsai Ming-Feng Tsai Ming-Feng Tsai Ming-Feng Chang Chia-Yu Chang Kuo-Song |
| author_facet | Liu Liong-Rung Liu Liong-Rung Liu Liong-Rung Chiu Hung-Wen Chiu Hung-Wen Huang Ming-Yuan Huang Ming-Yuan Huang Shu-Tien Huang Shu-Tien Huang Shu-Tien Tsai Ming-Feng Tsai Ming-Feng Tsai Ming-Feng Chang Chia-Yu Chang Kuo-Song |
| author_sort | Liu Liong-Rung |
| collection | DOAJ |
| description | Pneumonia and pulmonary edema are the most common causes of acute respiratory failure in emergency and intensive care. Airway maintenance and heart function preservation are two foundations for resuscitation. Laboratory examinations have been utilized for clinicians to early differentiate pneumonia and pulmonary edema; however, none can provide results as prompt as radiology examinations, such as portable chest X-ray (CXR), which can quickly deliver results without mobilizing patients. However, similar features between pneumonia and pulmonary edema are found in CXR. It remains challenging for Emergency Department (ED) physicians to make immediate decisions as radiologists cannot be on-site all the time and provide support. Thus, Accurate interpretation of images remains challenging in the emergency setting. References have shown that deep convolutional neural networks (CNN) have a high sensitivity in CXR readings. In this retrospective study, we collected the CXR images of patients over 65 hospitalized with pneumonia or pulmonary edema diagnosis between 2016 and 2020. After using the ICD-10 codes to select qualified patient records and removing the duplicated ones, we used keywords to label the image reports found in the electronic medical record (EMR) system. After that, we categorized their CXR images into five categories: positive correlation, negative correlation, no correlation, low correlation, and high correlation. Subcategorization was also performed to better differentiate characteristics. We applied six experiments includes the crop interference and non-interference categories by GoogLeNet and applied three times of validations. In our best model, the F1 scores for pneumonia and pulmonary edema are 0.835 and 0.829, respectively; accuracy rate: 83.2%, Recall rate: 83.2%, positive predictive value: 83.3%, and F1 Score: 0.832. After the validation, the best accuracy rate of our model can reach up to 73%. The model has a high negative predictive value of excluding pulmonary edema, meaning the CXR shows no sign of pulmonary edema. At the time, there was a high positive predictive value in pneumonia. In that way, we could use it as a clinical decision support (CDS) system to rule out pulmonary edema and rule in pneumonia contributing to the critical care of the elderly. |
| first_indexed | 2024-12-12T11:05:23Z |
| format | Article |
| id | doaj.art-202e301c7518493f9c7656f86413b73a |
| institution | Directory Open Access Journal |
| issn | 2296-858X |
| language | English |
| last_indexed | 2024-12-12T11:05:23Z |
| publishDate | 2022-06-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Medicine |
| spelling | doaj.art-202e301c7518493f9c7656f86413b73a2022-12-22T00:26:25ZengFrontiers Media S.A.Frontiers in Medicine2296-858X2022-06-01910.3389/fmed.2022.893208893208Using Artificial Intelligence to Establish Chest X-Ray Image Recognition Model to Assist Crucial Diagnosis in Elder Patients With DyspneaLiu Liong-Rung0Liu Liong-Rung1Liu Liong-Rung2Chiu Hung-Wen3Chiu Hung-Wen4Huang Ming-Yuan5Huang Ming-Yuan6Huang Shu-Tien7Huang Shu-Tien8Huang Shu-Tien9Tsai Ming-Feng10Tsai Ming-Feng11Tsai Ming-Feng12Chang Chia-Yu13Chang Kuo-Song14Department of Emergency Medicine, Mackay Memorial Hospital, Taipei, TaiwanDepartment of Medicine, Mackay Medical College, New Taipei City, TaiwanGraduate Institute of Biomedical Informatics, College of Medical Science and Technology, Taipei Medical University, Taipei, TaiwanGraduate Institute of Biomedical Informatics, College of Medical Science and Technology, Taipei Medical University, Taipei, TaiwanClinical Big Data Research Center, Taipei Medical University Hospital, Taipei, TaiwanDepartment of Emergency Medicine, Mackay Memorial Hospital, Taipei, TaiwanDepartment of Medicine, Mackay Medical College, New Taipei City, TaiwanDepartment of Emergency Medicine, Mackay Memorial Hospital, Taipei, TaiwanDepartment of Medicine, Mackay Medical College, New Taipei City, TaiwanGraduate Institute of Biomedical Informatics, College of Medical Science and Technology, Taipei Medical University, Taipei, TaiwanDepartment of Medicine, Mackay Medical College, New Taipei City, TaiwanGraduate Institute of Biomedical Informatics, College of Medical Science and Technology, Taipei Medical University, Taipei, TaiwanDivision of Plastic Surgery, Department of Surgery, Mackay Memorial Hospital, Taipei, TaiwanGraduate Institute of Biomedical Informatics, College of Medical Science and Technology, Taipei Medical University, Taipei, TaiwanDepartment of Emergency Medicine, Mackay Memorial Hospital, Taipei, TaiwanPneumonia and pulmonary edema are the most common causes of acute respiratory failure in emergency and intensive care. Airway maintenance and heart function preservation are two foundations for resuscitation. Laboratory examinations have been utilized for clinicians to early differentiate pneumonia and pulmonary edema; however, none can provide results as prompt as radiology examinations, such as portable chest X-ray (CXR), which can quickly deliver results without mobilizing patients. However, similar features between pneumonia and pulmonary edema are found in CXR. It remains challenging for Emergency Department (ED) physicians to make immediate decisions as radiologists cannot be on-site all the time and provide support. Thus, Accurate interpretation of images remains challenging in the emergency setting. References have shown that deep convolutional neural networks (CNN) have a high sensitivity in CXR readings. In this retrospective study, we collected the CXR images of patients over 65 hospitalized with pneumonia or pulmonary edema diagnosis between 2016 and 2020. After using the ICD-10 codes to select qualified patient records and removing the duplicated ones, we used keywords to label the image reports found in the electronic medical record (EMR) system. After that, we categorized their CXR images into five categories: positive correlation, negative correlation, no correlation, low correlation, and high correlation. Subcategorization was also performed to better differentiate characteristics. We applied six experiments includes the crop interference and non-interference categories by GoogLeNet and applied three times of validations. In our best model, the F1 scores for pneumonia and pulmonary edema are 0.835 and 0.829, respectively; accuracy rate: 83.2%, Recall rate: 83.2%, positive predictive value: 83.3%, and F1 Score: 0.832. After the validation, the best accuracy rate of our model can reach up to 73%. The model has a high negative predictive value of excluding pulmonary edema, meaning the CXR shows no sign of pulmonary edema. At the time, there was a high positive predictive value in pneumonia. In that way, we could use it as a clinical decision support (CDS) system to rule out pulmonary edema and rule in pneumonia contributing to the critical care of the elderly.https://www.frontiersin.org/articles/10.3389/fmed.2022.893208/fullcomputer-aided detection (CAD)artificial intelligencegeriatrics medicinecritical care medicinechest X-ray (CXR) |
| spellingShingle | Liu Liong-Rung Liu Liong-Rung Liu Liong-Rung Chiu Hung-Wen Chiu Hung-Wen Huang Ming-Yuan Huang Ming-Yuan Huang Shu-Tien Huang Shu-Tien Huang Shu-Tien Tsai Ming-Feng Tsai Ming-Feng Tsai Ming-Feng Chang Chia-Yu Chang Kuo-Song Using Artificial Intelligence to Establish Chest X-Ray Image Recognition Model to Assist Crucial Diagnosis in Elder Patients With Dyspnea Frontiers in Medicine computer-aided detection (CAD) artificial intelligence geriatrics medicine critical care medicine chest X-ray (CXR) |
| title | Using Artificial Intelligence to Establish Chest X-Ray Image Recognition Model to Assist Crucial Diagnosis in Elder Patients With Dyspnea |
| title_full | Using Artificial Intelligence to Establish Chest X-Ray Image Recognition Model to Assist Crucial Diagnosis in Elder Patients With Dyspnea |
| title_fullStr | Using Artificial Intelligence to Establish Chest X-Ray Image Recognition Model to Assist Crucial Diagnosis in Elder Patients With Dyspnea |
| title_full_unstemmed | Using Artificial Intelligence to Establish Chest X-Ray Image Recognition Model to Assist Crucial Diagnosis in Elder Patients With Dyspnea |
| title_short | Using Artificial Intelligence to Establish Chest X-Ray Image Recognition Model to Assist Crucial Diagnosis in Elder Patients With Dyspnea |
| title_sort | using artificial intelligence to establish chest x ray image recognition model to assist crucial diagnosis in elder patients with dyspnea |
| topic | computer-aided detection (CAD) artificial intelligence geriatrics medicine critical care medicine chest X-ray (CXR) |
| url | https://www.frontiersin.org/articles/10.3389/fmed.2022.893208/full |
| work_keys_str_mv | AT liuliongrung usingartificialintelligencetoestablishchestxrayimagerecognitionmodeltoassistcrucialdiagnosisinelderpatientswithdyspnea AT liuliongrung usingartificialintelligencetoestablishchestxrayimagerecognitionmodeltoassistcrucialdiagnosisinelderpatientswithdyspnea AT liuliongrung usingartificialintelligencetoestablishchestxrayimagerecognitionmodeltoassistcrucialdiagnosisinelderpatientswithdyspnea AT chiuhungwen usingartificialintelligencetoestablishchestxrayimagerecognitionmodeltoassistcrucialdiagnosisinelderpatientswithdyspnea AT chiuhungwen usingartificialintelligencetoestablishchestxrayimagerecognitionmodeltoassistcrucialdiagnosisinelderpatientswithdyspnea AT huangmingyuan usingartificialintelligencetoestablishchestxrayimagerecognitionmodeltoassistcrucialdiagnosisinelderpatientswithdyspnea AT huangmingyuan usingartificialintelligencetoestablishchestxrayimagerecognitionmodeltoassistcrucialdiagnosisinelderpatientswithdyspnea AT huangshutien usingartificialintelligencetoestablishchestxrayimagerecognitionmodeltoassistcrucialdiagnosisinelderpatientswithdyspnea AT huangshutien usingartificialintelligencetoestablishchestxrayimagerecognitionmodeltoassistcrucialdiagnosisinelderpatientswithdyspnea AT huangshutien usingartificialintelligencetoestablishchestxrayimagerecognitionmodeltoassistcrucialdiagnosisinelderpatientswithdyspnea AT tsaimingfeng usingartificialintelligencetoestablishchestxrayimagerecognitionmodeltoassistcrucialdiagnosisinelderpatientswithdyspnea AT tsaimingfeng usingartificialintelligencetoestablishchestxrayimagerecognitionmodeltoassistcrucialdiagnosisinelderpatientswithdyspnea AT tsaimingfeng usingartificialintelligencetoestablishchestxrayimagerecognitionmodeltoassistcrucialdiagnosisinelderpatientswithdyspnea AT changchiayu usingartificialintelligencetoestablishchestxrayimagerecognitionmodeltoassistcrucialdiagnosisinelderpatientswithdyspnea AT changkuosong usingartificialintelligencetoestablishchestxrayimagerecognitionmodeltoassistcrucialdiagnosisinelderpatientswithdyspnea |