Xi Lin
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Xi Lin has a total of 191 co-authors. Here are the top co-authors by number of shared publications:
Huiling Xiang
9 shared publicationsChunyan Li
6 shared publications- Huiling Xiang
Department of Radiology, State Key Laboratory of Oncology in South China, Guangdong Provincial Clinical Research Center for Cancer, Sun Yat-sen University Cancer Center, Guangzhou, China
5 shared publications Jinjing Ou
5 shared publicationsLixian Liu
Department of Ultrasound, Guangdong Second Provincial General Hospital, Guangzhou, China
5 shared publicationsCuiju Yan
4 shared publications
Latest Publications
- 1 November 2026
- Published by Radiological Society of North America (RSNA) in Radiology
- Vol. 321 (2)
Glandular Tissue Component at Preoperative Breast US and Recurrence Risk in Invasive Breast Cancer
High glandular tissue component at preoperative breast US was an independent predictor of ipsilateral or contralateral breast cancer after invasive breast cancer treatment across Korean and Chinese cohorts.
- 1 November 2026
- Published by Radiological Society of North America (RSNA) in Radiology
- Vol. 321 (2)
Glandular Tissue Component at Preoperative Breast US and Recurrence Risk in Invasive Breast Cancer
High glandular tissue component at preoperative breast US was an independent predictor of ipsilateral or contralateral breast cancer after invasive breast cancer treatment across Korean and Chinese cohorts.
- 1 May 2026
- Published by Springer Nature in Nature Biomedical Engineering
Towards generalizable AI in medicine via Generalist–Specialist Collaboration
Generalist foundation models (GFMs) are renowned for their exceptional capability and flexibility in diverse tasks. In the field of medicine, while GFMs exhibit superior generalizability, specialist models excel in precision because of their domain-specific knowledge. Here we show a cooperative framework, Generalist-Specialist Collaboration (GSCo), that synergistically combines a powerful generalist model with lightweight specialists. In this framework, specialists provide expert guidance, such as diagnostic predictions and visually similar clinical cases, as contextual information to the generalist, which then makes a final diagnosis. We developed MedDr, an open-source GFM tailored for medicine, as well as a suite of lightweight specialist models crafted for specific downstream tasks. A comprehensive evaluation on 32 datasets across diverse medical modalities shows that MedDr outperforms state-of-the-art GFMs on downstream datasets. Furthermore, GSCo exceeds GFMs and specialists in medical image diagnosis and report generation. This approach offers an effective and computationally efficient paradigm for deploying GFMs in clinical settings, enhancing scalability and enabling precise analysis across a wide range of scenarios.
- 1 May 2026
- Published by Springer Nature in Nature Biomedical Engineering
Towards generalizable AI in medicine via Generalist–Specialist Collaboration
Generalist foundation models (GFMs) are renowned for their exceptional capability and flexibility in diverse tasks. In the field of medicine, while GFMs exhibit superior generalizability, specialist models excel in precision because of their domain-specific knowledge. Here we show a cooperative framework, Generalist-Specialist Collaboration (GSCo), that synergistically combines a powerful generalist model with lightweight specialists. In this framework, specialists provide expert guidance, such as diagnostic predictions and visually similar clinical cases, as contextual information to the generalist, which then makes a final diagnosis. We developed MedDr, an open-source GFM tailored for medicine, as well as a suite of lightweight specialist models crafted for specific downstream tasks. A comprehensive evaluation on 32 datasets across diverse medical modalities shows that MedDr outperforms state-of-the-art GFMs on downstream datasets. Furthermore, GSCo exceeds GFMs and specialists in medical image diagnosis and report generation. This approach offers an effective and computationally efficient paradigm for deploying GFMs in clinical settings, enhancing scalability and enabling precise analysis across a wide range of scenarios.
- 19 December 2025
- Published by Elsevier in Academic Radiology
- Vol. 33 (4) , 1429-1440
Glandular Tissue Component on Breast Ultrasound: Association with Survival Outcomes in Women with Invasive Breast Cancer
- 19 December 2025
- Published by Elsevier in Academic Radiology
- Vol. 33 (4) , 1429-1440
Glandular Tissue Component on Breast Ultrasound: Association with Survival Outcomes in Women with Invasive Breast Cancer
- 21 September 2025
- Published by Springer Nature
- p. 450-459
Multimodal Fusion Network with Distribution-Based Tumor-Marker Imputation for Multi-origin Metastatic Cervical Lymphadenopathy Classification
- 21 September 2025
- Published by Springer Nature
- p. 450-459
Multimodal Fusion Network with Distribution-Based Tumor-Marker Imputation for Multi-origin Metastatic Cervical Lymphadenopathy Classification
- 14 May 2025
- Published by Elsevier in Physica Medica
- Vol. 134, 104989
Automated whole-breast ultrasound tumor diagnosis using attention-inception network
- 14 May 2025
- Published by Elsevier in Physica Medica
- Vol. 134, 104989
Automated whole-breast ultrasound tumor diagnosis using attention-inception network
- Open Access
- 12 February 2025
- Published by Elsevier in eClinicalMedicine
- Vol. 81, 103085
Quantitative vascular feature-based multimodality prediction model for multi-origin malignant cervical lymphadenopathy
- 12 February 2025
- Published by Elsevier in eClinicalMedicine
- Vol. 81, 103085
Open AccessQuantitative vascular feature-based multimodality prediction model for multi-origin malignant cervical lymphadenopathy
- Top Cited PapersOpen Access
- 27 March 2024
- Published by Springer Nature in Nature Communications
- Vol. 15 (1) , 1-12
Development and validation of an interpretable model integrating multimodal information for improving ovarian cancer diagnosis
Ovarian cancer, a group of heterogeneous diseases, presents with extensive characteristics with the highest mortality among gynecological malignancies. Accurate and early diagnosis of ovarian cancer is of great significance. Here, we present OvcaFinder, an interpretable model constructed from ultrasound images-based deep learning (DL) predictions, Ovarian–Adnexal Reporting and Data System scores from radiologists, and routine clinical variables. OvcaFinder outperforms the clinical model and the DL model with area under the curves (AUCs) of 0.978, and 0.947 in the internal and external test datasets, respectively. OvcaFinder assistance led to improved AUCs of radiologists and inter-reader agreement. The average AUCs were improved from 0.927 to 0.977 and from 0.904 to 0.941, and the false positive rates were decreased by 13.4% and 8.3% in the internal and external test datasets, respectively. This highlights the potential of OvcaFinder to improve the diagnostic accuracy, and consistency of radiologists in identifying ovarian cancer.
- 27 March 2024
- Published by Springer Nature in Nature Communications
- Vol. 15 (1) , 1-12
Top Cited PapersOpen AccessDevelopment and validation of an interpretable model integrating multimodal information for improving ovarian cancer diagnosis
Ovarian cancer, a group of heterogeneous diseases, presents with extensive characteristics with the highest mortality among gynecological malignancies. Accurate and early diagnosis of ovarian cancer is of great significance. Here, we present OvcaFinder, an interpretable model constructed from ultrasound images-based deep learning (DL) predictions, Ovarian–Adnexal Reporting and Data System scores from radiologists, and routine clinical variables. OvcaFinder outperforms the clinical model and the DL model with area under the curves (AUCs) of 0.978, and 0.947 in the internal and external test datasets, respectively. OvcaFinder assistance led to improved AUCs of radiologists and inter-reader agreement. The average AUCs were improved from 0.927 to 0.977 and from 0.904 to 0.941, and the false positive rates were decreased by 13.4% and 8.3% in the internal and external test datasets, respectively. This highlights the potential of OvcaFinder to improve the diagnostic accuracy, and consistency of radiologists in identifying ovarian cancer.
- Open Access
- 29 February 2024
- Published by Springer Nature in Cancer Imaging
- Vol. 24 (1) , 1-11
Development and validation of ultrasound-based radiomics model to predict germline BRCA mutations in patients with breast cancer
Background: Identifying breast cancer (BC) patients with germline breast cancer susceptibility gene (gBRCA) mutation is important. The current criteria for germline testing for BC remain controversial. This study aimed to develop a nomogram incorporating ultrasound radiomic features and clinicopathological factors to predict gBRCA mutations in patients with BC. Materials and methods: In this retrospective study, 497 women with BC who underwent gBRCA genetic testing from March 2013 to May 2022 were included, including 348 for training (84 with and 264 without a gBRCA mutation) and 149 for validation(36 patients with and 113 without a gBRCA mutation). Factors associated with gBRCA mutations were identified to establish a clinicopathological model. Radiomics features were extracted from the intratumoral and peritumoral regions (3 mm and 5 mm) of each image. The least absolute shrinkage and selection operator regression algorithm was used to select the features and logistic regression analysis was used to construct three imaging models. Finally, a nomogram that combined clinicopathological and radiomics features was developed. The models were evaluated based on the area under the receiver operating characteristic curve (AUC), calibration, and clinical usefulness. Results: Age at diagnosis, family history of BC, personal history of other BRCA-related cancers, and human epidermal growth factor receptor 2 status were independent predictors of the clinicopathological model. The AUC of the imaging radiomics model combining intratumoral and peritumoral 3 mm areas in the validation set was 0.783 (95% confidence interval [CI]: 0.702—0.862), which showed the best performance among three imaging models. The nomogram yielded better performance than the clinicopathological model in validation sets (AUC: 0.824 [0.755—0.894] versus 0.659 [0.563—0.755], p = 0.007). Conclusion: The nomogram based on ultrasound images and clinicopathological factors performs well in predicting gBRCA mutations in BC patients and may help to improve clinical decisions about genetic testing.
- 29 February 2024
- Published by Springer Nature in Cancer Imaging
- Vol. 24 (1) , 1-11
Open AccessDevelopment and validation of ultrasound-based radiomics model to predict germline BRCA mutations in patients with breast cancer
Background: Identifying breast cancer (BC) patients with germline breast cancer susceptibility gene (gBRCA) mutation is important. The current criteria for germline testing for BC remain controversial. This study aimed to develop a nomogram incorporating ultrasound radiomic features and clinicopathological factors to predict gBRCA mutations in patients with BC. Materials and methods: In this retrospective study, 497 women with BC who underwent gBRCA genetic testing from March 2013 to May 2022 were included, including 348 for training (84 with and 264 without a gBRCA mutation) and 149 for validation(36 patients with and 113 without a gBRCA mutation). Factors associated with gBRCA mutations were identified to establish a clinicopathological model. Radiomics features were extracted from the intratumoral and peritumoral regions (3 mm and 5 mm) of each image. The least absolute shrinkage and selection operator regression algorithm was used to select the features and logistic regression analysis was used to construct three imaging models. Finally, a nomogram that combined clinicopathological and radiomics features was developed. The models were evaluated based on the area under the receiver operating characteristic curve (AUC), calibration, and clinical usefulness. Results: Age at diagnosis, family history of BC, personal history of other BRCA-related cancers, and human epidermal growth factor receptor 2 status were independent predictors of the clinicopathological model. The AUC of the imaging radiomics model combining intratumoral and peritumoral 3 mm areas in the validation set was 0.783 (95% confidence interval [CI]: 0.702—0.862), which showed the best performance among three imaging models. The nomogram yielded better performance than the clinicopathological model in validation sets (AUC: 0.824 [0.755—0.894] versus 0.659 [0.563—0.755], p = 0.007). Conclusion: The nomogram based on ultrasound images and clinicopathological factors performs well in predicting gBRCA mutations in BC patients and may help to improve clinical decisions about genetic testing.
- 1 September 2023
- Published by Radiological Society of North America (RSNA) in Radiology: Artificial Intelligence
- Vol. 5 (5) , e220185
Deep Learning-assisted Diagnosis of Breast Lesions on US Images: A Multivendor, Multicenter Study
To evaluate the diagnostic performance of a deep learning (DL) model for breast US across four hospitals and assess its value to readers with different levels of experience. In this retrospective study, a dual attention–based convolutional neural network was built and validated to discriminate malignant tumors from benign tumors by using B-mode and color Doppler US images (n = 45 909, March 2011–August 2018), acquired with 42 types of US machines, of 9895 pathologic analysis–confirmed breast lesions in 8797 patients (27 men and 8770 women; mean age, 47 years ± 12 [SD]). With and without assistance from the DL model, three novice readers with less than 5 years of US experience and two experienced readers with 8 and 18 years of US experience, respectively, interpreted 1024 randomly selected lesions. Differences in the areas under the receiver operating characteristic curves (AUCs) were tested using the DeLong test. The DL model using both B-mode and color Doppler US images demonstrated expert-level performance at the lesion level, with an AUC of 0.94 (95% CI: 0.92, 0.95) for the internal set. In external datasets, the AUCs were 0.92 (95% CI: 0.90, 0.94) for hospital 1, 0.91 (95% CI: 0.89, 0.94) for hospital 2, and 0.96 (95% CI: 0.94, 0.98) for hospital 3. DL assistance led to improved AUCs (P < .001) for one experienced and three novice radiologists and improved interobserver agreement. The average false-positive rate was reduced by 7.6% (P = .08). The DL model may help radiologists, especially novice readers, improve accuracy and interobserver agreement of breast tumor diagnosis using US. Keywords: Ultrasound, Breast, Diagnosis, Breast Cancer, Deep Learning, Ultrasonography Supplemental material is available for this article. © RSNA, 2023
- 1 September 2023
- Published by Radiological Society of North America (RSNA) in Radiology: Artificial Intelligence
- Vol. 5 (5) , e220185
Deep Learning-assisted Diagnosis of Breast Lesions on US Images: A Multivendor, Multicenter Study
To evaluate the diagnostic performance of a deep learning (DL) model for breast US across four hospitals and assess its value to readers with different levels of experience. In this retrospective study, a dual attention–based convolutional neural network was built and validated to discriminate malignant tumors from benign tumors by using B-mode and color Doppler US images (n = 45 909, March 2011–August 2018), acquired with 42 types of US machines, of 9895 pathologic analysis–confirmed breast lesions in 8797 patients (27 men and 8770 women; mean age, 47 years ± 12 [SD]). With and without assistance from the DL model, three novice readers with less than 5 years of US experience and two experienced readers with 8 and 18 years of US experience, respectively, interpreted 1024 randomly selected lesions. Differences in the areas under the receiver operating characteristic curves (AUCs) were tested using the DeLong test. The DL model using both B-mode and color Doppler US images demonstrated expert-level performance at the lesion level, with an AUC of 0.94 (95% CI: 0.92, 0.95) for the internal set. In external datasets, the AUCs were 0.92 (95% CI: 0.90, 0.94) for hospital 1, 0.91 (95% CI: 0.89, 0.94) for hospital 2, and 0.96 (95% CI: 0.94, 0.98) for hospital 3. DL assistance led to improved AUCs (P < .001) for one experienced and three novice radiologists and improved interobserver agreement. The average false-positive rate was reduced by 7.6% (P = .08). The DL model may help radiologists, especially novice readers, improve accuracy and interobserver agreement of breast tumor diagnosis using US. Keywords: Ultrasound, Breast, Diagnosis, Breast Cancer, Deep Learning, Ultrasonography Supplemental material is available for this article. © RSNA, 2023