Xu Liu

Xu Liu

Auto-generated profile, based on publicly available information

Xu Liu

ENT institute and Department of Otorhinolaryngology, Eye & ENT Hospital, Fudan University, Shanghai, China

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
16 May 2022 in The Laryngoscope
Verified
Journal Article
Open access
XLXu Liu
Xu Liu
YPYucheng Pan
Yucheng Pan
XZXin Zhang
Xin Zhang
YSYongfang Sha
Yongfang Sha
SWShihui Wang
Shihui Wang
JLJianping Liu
Jianping Liu
Objective: To design a deep learning model based on multimodal magnetic resonance image (MRI) sequences for automatic parotid neoplasm classification, and to improve the diagnostic decision‐making in clinical settings.Methods: First, multimodal MRI sequences were collected from 266 patients with parotid neoplasms, and an artificial intelligence (AI)‐based deep learning model was designed from scratch, combining the image classification network of Resnet and the Transformer network of Natural language processing. Second, the effectiveness of the deep learning model was improved through the multi‐modality fusion of MRI sequences, and the fusion strategy of various MRI sequences was optimized. In addition, we compared the effectiveness of the model in the parotid neoplasm classification with experienced radiologists.Results: The deep learning model delivered reliable outcomes in differentiating benign and malignant parotid neoplasms. The model, which was trained by the fusion of T2‐weighted, postcontrast T1‐weighted, and diffusion‐weighted imaging (b = 1000 s/mm2), produced the best result, with an accuracy score of 0.85, an area under the receiver operator characteristic (ROC) curve of 0.96, a sensitivity score of 0.90, and a specificity score of 0.84. In addition, the multi‐modal paradigm exhibited reliable outcomes in diagnosing the pleomorphic adenoma and the Warthin tumor, but not in the identification of the basal cell adenoma.Conclusion: An accurate and efficient AI based classification model was produced to classify parotid neoplasms, resulting from the fusion of multimodal MRI sequences. The effectiveness certainly outperformed the model with single MRI images or single MRI sequences as input, and potentially, experienced radiologists.Level of Evidence: 3 Laryngoscope, 133:327–335, 2023

5 April 2023 in Laryngoscope Investigative Otolaryngology
Verified
Journal Article
Open access
MXMengzhao Xun
Mengzhao Xun
XLXu Liu
Xu Liu
YSYongfang Sha
Yongfang Sha
XZXin Zhang
Xin Zhang
JLJian Ping Liu
Jian Ping Liu
Objective: The purpose of this meta‐analysis was to compare the efficiency of high‐resolution computed tomography (HRCT) and diffusion‐weighted magnetic resonance imaging (DWI) in guiding the diagnosis of middle ear cholesteatoma in clinical practice.Materials and methods: Cochrane Library, Medline, Embase, PubMed, and Web of Science were searched for studies that evaluated the sensitivity and specificity of HRCT or DWI in detecting middle ear cholesteatoma. A random‐effects model was used to calculate and summarize the pooled estimates of sensitivity, specificity, and diagnostic odds ratios. Postoperative pathological results were considered as the diagnostic gold standard for middle ear cholesteatoma.Results: Fourteen published articles (860 patients) met the inclusion criteria. The sensitivity and specificity of DWI when diagnosing cholesteatoma (regardless of type) were 0.88 (95% confidence interval [CI], 0.80–0.93) and 0.93 (95% CI, 0.86–0.97), respectively, while those of HRCT were 0.68 (95% CI, 0.57–0.77) and 0.78 (95% CI, 0.60–0.90), respectively. Notably, the sensitivity and specificity levels of DWI were similar to those of HRCT (p = .1178 for sensitivity, p = .2144 for specificity; pair‐sampled t tests). The sensitivity and specificity of DWI or HRCT for the diagnosis of primary cholesteatoma were 0.78 (95% CI, 0.65–0.88) and 0.84 (95% CI, 0.69–0.93), respectively, while that for recurrent cholesteatoma were 0.93 (95% CI, 0.61–0.99) and 0.94 (95% CI, 0.82–0.98), respectively.Conclusion: DWI and HRCT have similar levels of high sensitivity and specificity in detecting various cholesteatomas. Also, the diagnostic efficiency of HRCT or DWI for recurrent cholesteatoma is identical to that of primary cholesteatoma. Therefore, HRCT may be used in clinical settings to reduce the use of DWI and save clinical resources.Lay summary: Data on the use of diffusion‐weighted magnetic resonance imaging and high‐resolution computed tomography in the diagnosis of cholesteatoma were obtained through a literature search. They were analyzed to guide the clinical diagnosis and treatment of cholesteatoma.Level of evidence: NA

17 January 2025 in European Archives of Oto-Rhino-Laryngology
Verified
Journal Article
Closed access
PDF file
XLXu Liu
Xu Liu
YHYue-Lin Hsieh
Yue-Lin Hsieh
YWYunfeng Wang
Yunfeng Wang
WWWuqing Wang
Wuqing Wang

23 August 2025 in European Archives of Oto-Rhino-Laryngology
Verified
Journal Article
Closed access
PDF file
YHYue-Lin Hsieh
Yue-Lin Hsieh
XLXu Liu
Xu Liu
JSJae-Jin Song
Jae-Jin Song
DEDavid J. Eisenman
David J. Eisenman
WWWuqing Wang
Wuqing Wang