Interpretable artificial intelligence-based app assists inexperienced radiologists in diagnosing biliary atresia from sonographic gallbladder images
Abstract Background A previously trained deep learning-based smartphone app provides an artificial intelligence solution to help diagnose biliary atresia from sonographic gallbladder images, but it might be impractical to launch it in real clinical settings. This study aimed to redevelop a new model...
| Main Authors: | , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2024-01-01
|
| Series: | BMC Medicine |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s12916-024-03247-9 |
| _version_ | 1797276479891963904 |
|---|---|
| author | Wenying Zhou Zejun Ye Guangliang Huang Xiaoer Zhang Ming Xu Baoxian Liu Bowen Zhuang Zijian Tang Shan Wang Dan Chen Yunxiang Pan Xiaoyan Xie Ruixuan Wang Luyao Zhou |
| author_facet | Wenying Zhou Zejun Ye Guangliang Huang Xiaoer Zhang Ming Xu Baoxian Liu Bowen Zhuang Zijian Tang Shan Wang Dan Chen Yunxiang Pan Xiaoyan Xie Ruixuan Wang Luyao Zhou |
| author_sort | Wenying Zhou |
| collection | DOAJ |
| description | Abstract Background A previously trained deep learning-based smartphone app provides an artificial intelligence solution to help diagnose biliary atresia from sonographic gallbladder images, but it might be impractical to launch it in real clinical settings. This study aimed to redevelop a new model using original sonographic images and their derived smartphone photos and then test the new model’s performance in assisting radiologists with different experiences to detect biliary atresia in real-world mimic settings. Methods A new model was first trained retrospectively using 3659 original sonographic gallbladder images and their derived 51,226 smartphone photos and tested on 11,410 external validation smartphone photos. Afterward, the new model was tested in 333 prospectively collected sonographic gallbladder videos from 207 infants by 14 inexperienced radiologists (9 juniors and 5 seniors) and 4 experienced pediatric radiologists in real-world mimic settings. Diagnostic performance was expressed as the area under the receiver operating characteristic curve (AUC). Results The new model outperformed the previously published model in diagnosing BA on the external validation set (AUC 0.924 vs 0.908, P = 0.004) with higher consistency (kappa value 0.708 vs 0.609). When tested in real-world mimic settings using 333 sonographic gallbladder videos, the new model performed comparable to experienced pediatric radiologists (average AUC 0.860 vs 0.876) and outperformed junior radiologists (average AUC 0.838 vs 0.773) and senior radiologists (average AUC 0.829 vs 0.749). Furthermore, the new model could aid both junior and senior radiologists to improve their diagnostic performances, with the average AUC increasing from 0.773 to 0.835 for junior radiologists and from 0.749 to 0.805 for senior radiologists. Conclusions The interpretable app-based model showed robust and satisfactory performance in diagnosing biliary atresia, and it could aid radiologists with limited experiences to improve their diagnostic performances in real-world mimic settings. |
| first_indexed | 2024-03-07T15:28:46Z |
| format | Article |
| id | doaj.art-779563d6fc6d431988b55a1b0404a027 |
| institution | Directory Open Access Journal |
| issn | 1741-7015 |
| language | English |
| last_indexed | 2024-03-07T15:28:46Z |
| publishDate | 2024-01-01 |
| publisher | BMC |
| record_format | Article |
| series | BMC Medicine |
| spelling | doaj.art-779563d6fc6d431988b55a1b0404a0272024-03-05T16:32:25ZengBMCBMC Medicine1741-70152024-01-0122111410.1186/s12916-024-03247-9Interpretable artificial intelligence-based app assists inexperienced radiologists in diagnosing biliary atresia from sonographic gallbladder imagesWenying Zhou0Zejun Ye1Guangliang Huang2Xiaoer Zhang3Ming Xu4Baoxian Liu5Bowen Zhuang6Zijian Tang7Shan Wang8Dan Chen9Yunxiang Pan10Xiaoyan Xie11Ruixuan Wang12Luyao Zhou13Department of Medical Ultrasonics, Institute for Diagnostic and Interventional Ultrasound, The First Affiliated Hospital, Sun Yat-Sen UniversitySchool of Computer Science and Engineering, Sun Yat-Sen UniversityDepartment of Medical Ultrasonics, Institute for Diagnostic and Interventional Ultrasound, The First Affiliated Hospital, Sun Yat-Sen UniversityDepartment of Medical Ultrasonics, Institute for Diagnostic and Interventional Ultrasound, The First Affiliated Hospital, Sun Yat-Sen UniversityDepartment of Medical Ultrasonics, Institute for Diagnostic and Interventional Ultrasound, The First Affiliated Hospital, Sun Yat-Sen UniversityDepartment of Medical Ultrasonics, Institute for Diagnostic and Interventional Ultrasound, The First Affiliated Hospital, Sun Yat-Sen UniversityDepartment of Medical Ultrasonics, Institute for Diagnostic and Interventional Ultrasound, The First Affiliated Hospital, Sun Yat-Sen UniversityDepartment of Ultrasound, Shenzhen Children’s HospitalDepartment of Ultrasound, Shenzhen Children’s HospitalDepartment of Ultrasound, Guangdong Women and Children’s HospitalDepartment of Ultrasound, Guangdong Women and Children’s HospitalDepartment of Medical Ultrasonics, Institute for Diagnostic and Interventional Ultrasound, The First Affiliated Hospital, Sun Yat-Sen UniversitySchool of Computer Science and Engineering, Sun Yat-Sen UniversityDepartment of Medical Ultrasonics, Institute for Diagnostic and Interventional Ultrasound, The First Affiliated Hospital, Sun Yat-Sen UniversityAbstract Background A previously trained deep learning-based smartphone app provides an artificial intelligence solution to help diagnose biliary atresia from sonographic gallbladder images, but it might be impractical to launch it in real clinical settings. This study aimed to redevelop a new model using original sonographic images and their derived smartphone photos and then test the new model’s performance in assisting radiologists with different experiences to detect biliary atresia in real-world mimic settings. Methods A new model was first trained retrospectively using 3659 original sonographic gallbladder images and their derived 51,226 smartphone photos and tested on 11,410 external validation smartphone photos. Afterward, the new model was tested in 333 prospectively collected sonographic gallbladder videos from 207 infants by 14 inexperienced radiologists (9 juniors and 5 seniors) and 4 experienced pediatric radiologists in real-world mimic settings. Diagnostic performance was expressed as the area under the receiver operating characteristic curve (AUC). Results The new model outperformed the previously published model in diagnosing BA on the external validation set (AUC 0.924 vs 0.908, P = 0.004) with higher consistency (kappa value 0.708 vs 0.609). When tested in real-world mimic settings using 333 sonographic gallbladder videos, the new model performed comparable to experienced pediatric radiologists (average AUC 0.860 vs 0.876) and outperformed junior radiologists (average AUC 0.838 vs 0.773) and senior radiologists (average AUC 0.829 vs 0.749). Furthermore, the new model could aid both junior and senior radiologists to improve their diagnostic performances, with the average AUC increasing from 0.773 to 0.835 for junior radiologists and from 0.749 to 0.805 for senior radiologists. Conclusions The interpretable app-based model showed robust and satisfactory performance in diagnosing biliary atresia, and it could aid radiologists with limited experiences to improve their diagnostic performances in real-world mimic settings.https://doi.org/10.1186/s12916-024-03247-9Biliary atresiaDeep learningSmartphone appGallbladderUltrasonography |
| spellingShingle | Wenying Zhou Zejun Ye Guangliang Huang Xiaoer Zhang Ming Xu Baoxian Liu Bowen Zhuang Zijian Tang Shan Wang Dan Chen Yunxiang Pan Xiaoyan Xie Ruixuan Wang Luyao Zhou Interpretable artificial intelligence-based app assists inexperienced radiologists in diagnosing biliary atresia from sonographic gallbladder images BMC Medicine Biliary atresia Deep learning Smartphone app Gallbladder Ultrasonography |
| title | Interpretable artificial intelligence-based app assists inexperienced radiologists in diagnosing biliary atresia from sonographic gallbladder images |
| title_full | Interpretable artificial intelligence-based app assists inexperienced radiologists in diagnosing biliary atresia from sonographic gallbladder images |
| title_fullStr | Interpretable artificial intelligence-based app assists inexperienced radiologists in diagnosing biliary atresia from sonographic gallbladder images |
| title_full_unstemmed | Interpretable artificial intelligence-based app assists inexperienced radiologists in diagnosing biliary atresia from sonographic gallbladder images |
| title_short | Interpretable artificial intelligence-based app assists inexperienced radiologists in diagnosing biliary atresia from sonographic gallbladder images |
| title_sort | interpretable artificial intelligence based app assists inexperienced radiologists in diagnosing biliary atresia from sonographic gallbladder images |
| topic | Biliary atresia Deep learning Smartphone app Gallbladder Ultrasonography |
| url | https://doi.org/10.1186/s12916-024-03247-9 |
| work_keys_str_mv | AT wenyingzhou interpretableartificialintelligencebasedappassistsinexperiencedradiologistsindiagnosingbiliaryatresiafromsonographicgallbladderimages AT zejunye interpretableartificialintelligencebasedappassistsinexperiencedradiologistsindiagnosingbiliaryatresiafromsonographicgallbladderimages AT guanglianghuang interpretableartificialintelligencebasedappassistsinexperiencedradiologistsindiagnosingbiliaryatresiafromsonographicgallbladderimages AT xiaoerzhang interpretableartificialintelligencebasedappassistsinexperiencedradiologistsindiagnosingbiliaryatresiafromsonographicgallbladderimages AT mingxu interpretableartificialintelligencebasedappassistsinexperiencedradiologistsindiagnosingbiliaryatresiafromsonographicgallbladderimages AT baoxianliu interpretableartificialintelligencebasedappassistsinexperiencedradiologistsindiagnosingbiliaryatresiafromsonographicgallbladderimages AT bowenzhuang interpretableartificialintelligencebasedappassistsinexperiencedradiologistsindiagnosingbiliaryatresiafromsonographicgallbladderimages AT zijiantang interpretableartificialintelligencebasedappassistsinexperiencedradiologistsindiagnosingbiliaryatresiafromsonographicgallbladderimages AT shanwang interpretableartificialintelligencebasedappassistsinexperiencedradiologistsindiagnosingbiliaryatresiafromsonographicgallbladderimages AT danchen interpretableartificialintelligencebasedappassistsinexperiencedradiologistsindiagnosingbiliaryatresiafromsonographicgallbladderimages AT yunxiangpan interpretableartificialintelligencebasedappassistsinexperiencedradiologistsindiagnosingbiliaryatresiafromsonographicgallbladderimages AT xiaoyanxie interpretableartificialintelligencebasedappassistsinexperiencedradiologistsindiagnosingbiliaryatresiafromsonographicgallbladderimages AT ruixuanwang interpretableartificialintelligencebasedappassistsinexperiencedradiologistsindiagnosingbiliaryatresiafromsonographicgallbladderimages AT luyaozhou interpretableartificialintelligencebasedappassistsinexperiencedradiologistsindiagnosingbiliaryatresiafromsonographicgallbladderimages |