Deep learning based ultrasound analysis facilitates precise distinction between parotid pleomorphic adenoma and Warthin tumor
BackgroundPleomorphic adenoma (PA), often with the benign-like imaging appearances similar to Warthin tumor (WT), however, is a potentially malignant tumor with a high recurrence rate. It is worse that pathological fine-needle aspiration cytology (FNAC) is difficult to distinguish PA and WT for inex...
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2024-02-01
|
| Series: | Frontiers in Oncology |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fonc.2024.1337631/full |
| _version_ | 1797294183761838080 |
|---|---|
| author | Xi-hui Liu Xi-hui Liu Yi-yi Miao Lang Qian Lang Qian Zhao-ting Shi Zhao-ting Shi Yu Wang Jiong-long Su Cai Chang Cai Chang Jia-ying Chen Jian-gang Chen Jia-wei Li Jia-wei Li |
| author_facet | Xi-hui Liu Xi-hui Liu Yi-yi Miao Lang Qian Lang Qian Zhao-ting Shi Zhao-ting Shi Yu Wang Jiong-long Su Cai Chang Cai Chang Jia-ying Chen Jian-gang Chen Jia-wei Li Jia-wei Li |
| author_sort | Xi-hui Liu |
| collection | DOAJ |
| description | BackgroundPleomorphic adenoma (PA), often with the benign-like imaging appearances similar to Warthin tumor (WT), however, is a potentially malignant tumor with a high recurrence rate. It is worse that pathological fine-needle aspiration cytology (FNAC) is difficult to distinguish PA and WT for inexperienced pathologists. This study employed deep learning (DL) technology, which effectively utilized ultrasound images, to provide a reliable approach for discriminating PA from WT.Methods488 surgically confirmed patients, including 266 with PA and 222 with WT, were enrolled in this study. Two experienced ultrasound physicians independently evaluated all images to differentiate between PA and WT. The diagnostic performance of preoperative FNAC was also evaluated. During the DL study, all ultrasound images were randomly divided into training (70%), validation (20%), and test (10%) sets. Furthermore, ultrasound images that could not be diagnosed by FNAC were also randomly allocated to training (60%), validation (20%), and test (20%) sets. Five DL models were developed to classify ultrasound images as PA or WT. The robustness of these models was assessed using five-fold cross-validation. The Gradient-weighted Class Activation Mapping (Grad-CAM) technique was employed to visualize the region of interest in the DL models.ResultsIn Grad-CAM analysis, the DL models accurately identified the mass as the region of interest. The area under the receiver operating characteristic curve (AUROC) of the two ultrasound physicians were 0.351 and 0.598, and FNAC achieved an AUROC of only 0.721. Meanwhile, for DL models, the AUROC value for discriminating between PA and WT in the test set was from 0.828 to 0.908. ResNet50 demonstrated the optimal performance with an AUROC of 0.908, an accuracy of 0.833, a sensitivity of 0.736, and a specificity of 0.904. In the test set of cases that FNAC failed to provide a diagnosis, DenseNet121 demonstrated the optimal performance with an AUROC of 0.897, an accuracy of 0.806, a sensitivity of 0.789, and a specificity of 0.824.ConclusionFor the discrimination of PA and WT, DL models are superior to ultrasound and FNAC, thereby facilitating surgeons in making informed decisions regarding the most appropriate surgical approach. |
| first_indexed | 2024-03-07T21:26:34Z |
| format | Article |
| id | doaj.art-18fc9072645c4ca9bc518ac2073ce072 |
| institution | Directory Open Access Journal |
| issn | 2234-943X |
| language | English |
| last_indexed | 2024-03-07T21:26:34Z |
| publishDate | 2024-02-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Oncology |
| spelling | doaj.art-18fc9072645c4ca9bc518ac2073ce0722024-02-27T04:25:04ZengFrontiers Media S.A.Frontiers in Oncology2234-943X2024-02-011410.3389/fonc.2024.13376311337631Deep learning based ultrasound analysis facilitates precise distinction between parotid pleomorphic adenoma and Warthin tumorXi-hui Liu0Xi-hui Liu1Yi-yi Miao2Lang Qian3Lang Qian4Zhao-ting Shi5Zhao-ting Shi6Yu Wang7Jiong-long Su8Cai Chang9Cai Chang10Jia-ying Chen11Jian-gang Chen12Jia-wei Li13Jia-wei Li14Department of Medical Ultrasound, Fudan University Shanghai Cancer Center, Shanghai, ChinaDepartment of Oncology, Shanghai Medical College, Fudan University, Shanghai, ChinaSchool of AI and Advanced Computing, XJTLU Entrepreneur College (Taicang), Xi’an Jiaotong-Liverpool University, Suzhou, ChinaDepartment of Medical Ultrasound, Fudan University Shanghai Cancer Center, Shanghai, ChinaDepartment of Oncology, Shanghai Medical College, Fudan University, Shanghai, ChinaDepartment of Medical Ultrasound, Fudan University Shanghai Cancer Center, Shanghai, ChinaDepartment of Oncology, Shanghai Medical College, Fudan University, Shanghai, ChinaDepartment of Oral Pathology, Shanghai Ninth People’s Hospital, Shanghai Jiaotong University, Shanghai, ChinaSchool of AI and Advanced Computing, XJTLU Entrepreneur College (Taicang), Xi’an Jiaotong-Liverpool University, Suzhou, ChinaDepartment of Medical Ultrasound, Fudan University Shanghai Cancer Center, Shanghai, ChinaDepartment of Oncology, Shanghai Medical College, Fudan University, Shanghai, ChinaDepartment of Neck Surgery, Fudan University Shanghai Cancer Center, Shanghai, ChinaShanghai Key Laboratory of Multidimensional Information Processing, East China Normal University, Shanghai, ChinaDepartment of Medical Ultrasound, Fudan University Shanghai Cancer Center, Shanghai, ChinaDepartment of Oncology, Shanghai Medical College, Fudan University, Shanghai, ChinaBackgroundPleomorphic adenoma (PA), often with the benign-like imaging appearances similar to Warthin tumor (WT), however, is a potentially malignant tumor with a high recurrence rate. It is worse that pathological fine-needle aspiration cytology (FNAC) is difficult to distinguish PA and WT for inexperienced pathologists. This study employed deep learning (DL) technology, which effectively utilized ultrasound images, to provide a reliable approach for discriminating PA from WT.Methods488 surgically confirmed patients, including 266 with PA and 222 with WT, were enrolled in this study. Two experienced ultrasound physicians independently evaluated all images to differentiate between PA and WT. The diagnostic performance of preoperative FNAC was also evaluated. During the DL study, all ultrasound images were randomly divided into training (70%), validation (20%), and test (10%) sets. Furthermore, ultrasound images that could not be diagnosed by FNAC were also randomly allocated to training (60%), validation (20%), and test (20%) sets. Five DL models were developed to classify ultrasound images as PA or WT. The robustness of these models was assessed using five-fold cross-validation. The Gradient-weighted Class Activation Mapping (Grad-CAM) technique was employed to visualize the region of interest in the DL models.ResultsIn Grad-CAM analysis, the DL models accurately identified the mass as the region of interest. The area under the receiver operating characteristic curve (AUROC) of the two ultrasound physicians were 0.351 and 0.598, and FNAC achieved an AUROC of only 0.721. Meanwhile, for DL models, the AUROC value for discriminating between PA and WT in the test set was from 0.828 to 0.908. ResNet50 demonstrated the optimal performance with an AUROC of 0.908, an accuracy of 0.833, a sensitivity of 0.736, and a specificity of 0.904. In the test set of cases that FNAC failed to provide a diagnosis, DenseNet121 demonstrated the optimal performance with an AUROC of 0.897, an accuracy of 0.806, a sensitivity of 0.789, and a specificity of 0.824.ConclusionFor the discrimination of PA and WT, DL models are superior to ultrasound and FNAC, thereby facilitating surgeons in making informed decisions regarding the most appropriate surgical approach.https://www.frontiersin.org/articles/10.3389/fonc.2024.1337631/fulldeep learningpleomorphic adenomaWarthin tumorultrasounddiagnosis |
| spellingShingle | Xi-hui Liu Xi-hui Liu Yi-yi Miao Lang Qian Lang Qian Zhao-ting Shi Zhao-ting Shi Yu Wang Jiong-long Su Cai Chang Cai Chang Jia-ying Chen Jian-gang Chen Jia-wei Li Jia-wei Li Deep learning based ultrasound analysis facilitates precise distinction between parotid pleomorphic adenoma and Warthin tumor Frontiers in Oncology deep learning pleomorphic adenoma Warthin tumor ultrasound diagnosis |
| title | Deep learning based ultrasound analysis facilitates precise distinction between parotid pleomorphic adenoma and Warthin tumor |
| title_full | Deep learning based ultrasound analysis facilitates precise distinction between parotid pleomorphic adenoma and Warthin tumor |
| title_fullStr | Deep learning based ultrasound analysis facilitates precise distinction between parotid pleomorphic adenoma and Warthin tumor |
| title_full_unstemmed | Deep learning based ultrasound analysis facilitates precise distinction between parotid pleomorphic adenoma and Warthin tumor |
| title_short | Deep learning based ultrasound analysis facilitates precise distinction between parotid pleomorphic adenoma and Warthin tumor |
| title_sort | deep learning based ultrasound analysis facilitates precise distinction between parotid pleomorphic adenoma and warthin tumor |
| topic | deep learning pleomorphic adenoma Warthin tumor ultrasound diagnosis |
| url | https://www.frontiersin.org/articles/10.3389/fonc.2024.1337631/full |
| work_keys_str_mv | AT xihuiliu deeplearningbasedultrasoundanalysisfacilitatesprecisedistinctionbetweenparotidpleomorphicadenomaandwarthintumor AT xihuiliu deeplearningbasedultrasoundanalysisfacilitatesprecisedistinctionbetweenparotidpleomorphicadenomaandwarthintumor AT yiyimiao deeplearningbasedultrasoundanalysisfacilitatesprecisedistinctionbetweenparotidpleomorphicadenomaandwarthintumor AT langqian deeplearningbasedultrasoundanalysisfacilitatesprecisedistinctionbetweenparotidpleomorphicadenomaandwarthintumor AT langqian deeplearningbasedultrasoundanalysisfacilitatesprecisedistinctionbetweenparotidpleomorphicadenomaandwarthintumor AT zhaotingshi deeplearningbasedultrasoundanalysisfacilitatesprecisedistinctionbetweenparotidpleomorphicadenomaandwarthintumor AT zhaotingshi deeplearningbasedultrasoundanalysisfacilitatesprecisedistinctionbetweenparotidpleomorphicadenomaandwarthintumor AT yuwang deeplearningbasedultrasoundanalysisfacilitatesprecisedistinctionbetweenparotidpleomorphicadenomaandwarthintumor AT jionglongsu deeplearningbasedultrasoundanalysisfacilitatesprecisedistinctionbetweenparotidpleomorphicadenomaandwarthintumor AT caichang deeplearningbasedultrasoundanalysisfacilitatesprecisedistinctionbetweenparotidpleomorphicadenomaandwarthintumor AT caichang deeplearningbasedultrasoundanalysisfacilitatesprecisedistinctionbetweenparotidpleomorphicadenomaandwarthintumor AT jiayingchen deeplearningbasedultrasoundanalysisfacilitatesprecisedistinctionbetweenparotidpleomorphicadenomaandwarthintumor AT jiangangchen deeplearningbasedultrasoundanalysisfacilitatesprecisedistinctionbetweenparotidpleomorphicadenomaandwarthintumor AT jiaweili deeplearningbasedultrasoundanalysisfacilitatesprecisedistinctionbetweenparotidpleomorphicadenomaandwarthintumor AT jiaweili deeplearningbasedultrasoundanalysisfacilitatesprecisedistinctionbetweenparotidpleomorphicadenomaandwarthintumor |