1
|
Jiang T, Chen C, Zhou Y, Cai S, Yan Y, Sui L, Lai M, Song M, Zhu X, Pan Q, Wang H, Chen X, Wang K, Xiong J, Chen L, Xu D. Deep learning-assisted diagnosis of benign and malignant parotid tumors based on ultrasound: a retrospective study. BMC Cancer 2024; 24:510. [PMID: 38654281 PMCID: PMC11036551 DOI: 10.1186/s12885-024-12277-8] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/31/2024] [Accepted: 04/16/2024] [Indexed: 04/25/2024] Open
Abstract
BACKGROUND To develop a deep learning(DL) model utilizing ultrasound images, and evaluate its efficacy in distinguishing between benign and malignant parotid tumors (PTs), as well as its practicality in assisting clinicians with accurate diagnosis. METHODS A total of 2211 ultrasound images of 980 pathologically confirmed PTs (Training set: n = 721; Validation set: n = 82; Internal-test set: n = 89; External-test set: n = 88) from 907 patients were retrospectively included in this study. The optimal model was selected and the diagnostic performance evaluation is conducted by utilizing the area under curve (AUC) of the receiver-operating characteristic(ROC) based on five different DL networks constructed at varying depths. Furthermore, a comparison of different seniority radiologists was made in the presence of the optimal auxiliary diagnosis model. Additionally, the diagnostic confusion matrix of the optimal model was calculated, and an analysis and summary of misjudged cases' characteristics were conducted. RESULTS The Resnet18 demonstrated superior diagnostic performance, with an AUC value of 0.947, accuracy of 88.5%, sensitivity of 78.2%, and specificity of 92.7% in internal-test set, and with an AUC value of 0.925, accuracy of 89.8%, sensitivity of 83.3%, and specificity of 90.6% in external-test set. The PTs were subjectively assessed twice by six radiologists, both with and without the assisted of the model. With the assisted of the model, both junior and senior radiologists demonstrated enhanced diagnostic performance. In the internal-test set, there was an increase in AUC values by 0.062 and 0.082 for junior radiologists respectively, while senior radiologists experienced an improvement of 0.066 and 0.106 in their respective AUC values. CONCLUSIONS The DL model based on ultrasound images demonstrates exceptional capability in distinguishing between benign and malignant PTs, thereby assisting radiologists of varying expertise levels to achieve heightened diagnostic performance, and serve as a noninvasive imaging adjunct diagnostic method for clinical purposes.
Collapse
Affiliation(s)
- Tian Jiang
- Department of Diagnostic Ultrasound Imaging & Interventional Therapy, Zhejiang Cancer Hospital, Hangzhou Institute of Medicine (HIM), Chinese Academy of Sciences, 310022, Hangzhou, Zhejiang, China
- Postgraduate training base Alliance of Wenzhou Medical University (Zhejiang Cancer Hospital), 310022, Hangzhou, Zhejiang, China
- Zhejiang Provincial Research Center for Cancer Intelligent Diagnosis and Molecular Technology, 310022, Hangzhou, Zhejiang, China
| | - Chen Chen
- Department of Diagnostic Ultrasound Imaging & Interventional Therapy, Zhejiang Cancer Hospital, Hangzhou Institute of Medicine (HIM), Chinese Academy of Sciences, 310022, Hangzhou, Zhejiang, China
- Wenling Big Data and Artificial Intelligence Institute in Medicine, 317502, TaiZhou, Zhejiang, China
- Taizhou Key Laboratory of Minimally Invasive Interventional Therapy & Artificial Intelligence, Taizhou Campus of Zhejiang Cancer Hospital (Taizhou Cancer Hospital), 317502, Taizhou, Zhejiang, China
| | - Yahan Zhou
- Wenling Big Data and Artificial Intelligence Institute in Medicine, 317502, TaiZhou, Zhejiang, China
- Taizhou Key Laboratory of Minimally Invasive Interventional Therapy & Artificial Intelligence, Taizhou Campus of Zhejiang Cancer Hospital (Taizhou Cancer Hospital), 317502, Taizhou, Zhejiang, China
| | - Shenzhou Cai
- Wenling Big Data and Artificial Intelligence Institute in Medicine, 317502, TaiZhou, Zhejiang, China
- Taizhou Key Laboratory of Minimally Invasive Interventional Therapy & Artificial Intelligence, Taizhou Campus of Zhejiang Cancer Hospital (Taizhou Cancer Hospital), 317502, Taizhou, Zhejiang, China
| | - Yuqi Yan
- Department of Diagnostic Ultrasound Imaging & Interventional Therapy, Zhejiang Cancer Hospital, Hangzhou Institute of Medicine (HIM), Chinese Academy of Sciences, 310022, Hangzhou, Zhejiang, China
- Postgraduate training base Alliance of Wenzhou Medical University (Zhejiang Cancer Hospital), 310022, Hangzhou, Zhejiang, China
- Wenling Big Data and Artificial Intelligence Institute in Medicine, 317502, TaiZhou, Zhejiang, China
- Taizhou Key Laboratory of Minimally Invasive Interventional Therapy & Artificial Intelligence, Taizhou Campus of Zhejiang Cancer Hospital (Taizhou Cancer Hospital), 317502, Taizhou, Zhejiang, China
| | - Lin Sui
- Department of Diagnostic Ultrasound Imaging & Interventional Therapy, Zhejiang Cancer Hospital, Hangzhou Institute of Medicine (HIM), Chinese Academy of Sciences, 310022, Hangzhou, Zhejiang, China
- Postgraduate training base Alliance of Wenzhou Medical University (Zhejiang Cancer Hospital), 310022, Hangzhou, Zhejiang, China
- Wenling Big Data and Artificial Intelligence Institute in Medicine, 317502, TaiZhou, Zhejiang, China
- Taizhou Key Laboratory of Minimally Invasive Interventional Therapy & Artificial Intelligence, Taizhou Campus of Zhejiang Cancer Hospital (Taizhou Cancer Hospital), 317502, Taizhou, Zhejiang, China
| | - Min Lai
- Department of Diagnostic Ultrasound Imaging & Interventional Therapy, Zhejiang Cancer Hospital, Hangzhou Institute of Medicine (HIM), Chinese Academy of Sciences, 310022, Hangzhou, Zhejiang, China
- Zhejiang Provincial Research Center for Cancer Intelligent Diagnosis and Molecular Technology, 310022, Hangzhou, Zhejiang, China
- Second Clinical College, Zhejiang University of Traditional Chinese Medicine, 310022, Hangzhou, Zhejiang, China
| | - Mei Song
- Department of Diagnostic Ultrasound Imaging & Interventional Therapy, Zhejiang Cancer Hospital, Hangzhou Institute of Medicine (HIM), Chinese Academy of Sciences, 310022, Hangzhou, Zhejiang, China
- Zhejiang Provincial Research Center for Cancer Intelligent Diagnosis and Molecular Technology, 310022, Hangzhou, Zhejiang, China
| | - Xi Zhu
- Department of Diagnostic Ultrasound Imaging & Interventional Therapy, Zhejiang Cancer Hospital, Hangzhou Institute of Medicine (HIM), Chinese Academy of Sciences, 310022, Hangzhou, Zhejiang, China
- Wenling Big Data and Artificial Intelligence Institute in Medicine, 317502, TaiZhou, Zhejiang, China
- Taizhou Key Laboratory of Minimally Invasive Interventional Therapy & Artificial Intelligence, Taizhou Campus of Zhejiang Cancer Hospital (Taizhou Cancer Hospital), 317502, Taizhou, Zhejiang, China
| | - Qianmeng Pan
- Taizhou Key Laboratory of Minimally Invasive Interventional Therapy & Artificial Intelligence, Taizhou Campus of Zhejiang Cancer Hospital (Taizhou Cancer Hospital), 317502, Taizhou, Zhejiang, China
| | - Hui Wang
- Taizhou Key Laboratory of Minimally Invasive Interventional Therapy & Artificial Intelligence, Taizhou Campus of Zhejiang Cancer Hospital (Taizhou Cancer Hospital), 317502, Taizhou, Zhejiang, China
| | - Xiayi Chen
- Wenling Big Data and Artificial Intelligence Institute in Medicine, 317502, TaiZhou, Zhejiang, China
- Taizhou Key Laboratory of Minimally Invasive Interventional Therapy & Artificial Intelligence, Taizhou Campus of Zhejiang Cancer Hospital (Taizhou Cancer Hospital), 317502, Taizhou, Zhejiang, China
| | - Kai Wang
- Dongyang Hospital Affiliated to Wenzhou Medical University, 322100, Jinhua, Zhejiang, China
| | - Jing Xiong
- Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, 518000, Shenzhen, Guangdong, China
| | - Liyu Chen
- Department of Diagnostic Ultrasound Imaging & Interventional Therapy, Zhejiang Cancer Hospital, Hangzhou Institute of Medicine (HIM), Chinese Academy of Sciences, 310022, Hangzhou, Zhejiang, China.
- Zhejiang Provincial Research Center for Cancer Intelligent Diagnosis and Molecular Technology, 310022, Hangzhou, Zhejiang, China.
| | - Dong Xu
- Department of Diagnostic Ultrasound Imaging & Interventional Therapy, Zhejiang Cancer Hospital, Hangzhou Institute of Medicine (HIM), Chinese Academy of Sciences, 310022, Hangzhou, Zhejiang, China.
- Postgraduate training base Alliance of Wenzhou Medical University (Zhejiang Cancer Hospital), 310022, Hangzhou, Zhejiang, China.
- Zhejiang Provincial Research Center for Cancer Intelligent Diagnosis and Molecular Technology, 310022, Hangzhou, Zhejiang, China.
- Wenling Big Data and Artificial Intelligence Institute in Medicine, 317502, TaiZhou, Zhejiang, China.
- Taizhou Key Laboratory of Minimally Invasive Interventional Therapy & Artificial Intelligence, Taizhou Campus of Zhejiang Cancer Hospital (Taizhou Cancer Hospital), 317502, Taizhou, Zhejiang, China.
| |
Collapse
|
2
|
Shi L, Wu D, Yang X, Yan C, Huang P. Contrast-Enhanced Ultrasound and Strain Elastography for Differentiating Benign and Malignant Parotid Tumors. ULTRASCHALL IN DER MEDIZIN (STUTTGART, GERMANY : 1980) 2023; 44:419-427. [PMID: 36731495 PMCID: PMC10629480 DOI: 10.1055/a-1866-4633] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 02/03/2022] [Accepted: 05/21/2022] [Indexed: 06/18/2023]
Abstract
OBJECTIVES Preoperative differentiation between benign parotid tumors (BPT) and malignant parotid tumors (MPT) is crucial for treatment decisions. The purpose of this study was to investigate the benefits of combining contrast-enhanced ultrasound (CEUS) and strain elastography (SE) for preoperative differentiation between BPT and MPT. METHODS A total of 115 patients with BPT (n=72) or MPT (n=43) who underwent ultrasound (US), SE, and CEUS were enrolled. US and CEUS features and the elasticity score were evaluated. Receiver operating characteristic curve (ROC) analysis was used to assess the diagnostic performance of SE, CEUS, and SE + CEUS with respect to identifying MPT from BPT. RESULTS Solitary presentation, larger diameter, irregular shape, ill-defined margin, heterogeneous echogenicity, and calcification on US and higher elasticity score on SE had a significant association with malignancy. MPT also presented an unclear margin, larger size after enhancement, and "fast-in and fast-out" pattern on CEUS. The combination of SE and CEUS was effective for differentiating MPT from BPT (AUC: 0.88, 0.80-0.95), with a sensitivity of 86.0%, specificity of 88.9%, and accuracy of 87.8%, which were significantly higher than the values for SE (AUC: 0.75, 0.66-0.85) and CEUS (AUC: 0.82, 0.73-0.91) alone. CONCLUSION The combination of CEUS and SE is valuable for distinguishing MPT from BPT.
Collapse
Affiliation(s)
- Liuhong Shi
- Department of Ultrasound in Medicine, Zhejiang University School of Medicine Second Affiliated Hospital, Hangzhou, China
| | - Dingting Wu
- Nutrition Division, The Fourth Affiliated Hospital Zhejiang University School of Medicine, Yiwu, China
| | - Xu Yang
- Pathology, Zhejiang University School of Medicine First Affiliated Hospital, Hangzhou, China
| | - Caoxin Yan
- Department of Ultrasound in Medicine, Zhejiang University School of Medicine Second Affiliated Hospital, Hangzhou, China
| | - Pintong Huang
- Department of Ultrasound in Medicine, Zhejiang University School of Medicine Second Affiliated Hospital, Hangzhou, China
| |
Collapse
|
5
|
Wang Y, Nie F, Wang P. Clinical Diagnostic Application of Contrast-Enhanced Ultrasound in Focal Lesions of the Salivary Glands. JOURNAL OF ULTRASOUND IN MEDICINE : OFFICIAL JOURNAL OF THE AMERICAN INSTITUTE OF ULTRASOUND IN MEDICINE 2022; 41:2535-2546. [PMID: 35043446 DOI: 10.1002/jum.15943] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Received: 11/03/2021] [Revised: 11/27/2021] [Accepted: 12/29/2021] [Indexed: 06/14/2023]
Abstract
OBJECTIVES To evaluate the clinical diagnostic value of contrast-enhanced ultrasound (CEUS) for focal benign and malignant lesions of the salivary glands, as well as for common benign lesions. METHODS A total of 91 patients with focal lesions of the salivary glands were included in this study. In this study, CEUS was used to study the differential diagnosis of focal benign and malignant lesions of the salivary gland and the most common benign tumors, that is, pleomorphic adenoma (PA) and adenolymphoma. RESULTS The differences between focal benign and malignant lesions in the salivary glands were statistically significant (P < .05) in terms of qualitative CEUS indicators, enhancement pattern, enhancement homogeneity, enhancement margin, and enhanced lesion size, whereas the differences were not statistically significant (P > .05) in terms of wash-in and wash-out pattern, enhancement degree. Blurred margins and increased size of the lesion after enhancement are two CEUS features independently associated with focal malignant lesions of the salivary gland. The differences between salivary gland PA and adenolymphoma were statistically significant (P < .05) in terms of wash-in pattern, enhancement degree, enhancement homogeneity, and enhancement pattern, but not in terms of wash-out pattern, enhancement margin, and enhanced lesion size (P > .05). CONCLUSIONS As an economical, convenient, and safe imaging method, CEUS has important clinical value in distinguishing benign and malignant salivary glands.
Collapse
Affiliation(s)
- Yanqing Wang
- Department of Ultrasound, Lanzhou University Second Hospital, Lanzhou, China
| | - Fang Nie
- Department of Ultrasound, Lanzhou University Second Hospital, Lanzhou, China
| | - Peihua Wang
- Department of Ultrasound, Lanzhou University Second Hospital, Lanzhou, China
| |
Collapse
|