Juanmei Chen

Ms. Juanmei Chen


Research Focus & Metrics

Academic Career Length

1Year (2026-Present)

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

Representative Publications

View All Publications
19 June 2026 in Diagnostics
Verified
Journal Article
Open access
PDF file
JCJuanmei Chen
Juanmei Chen
ZDZijian Deng
Zijian Deng
YCYong Chen
Yong Chen
RYRuiheng Ye
Ruiheng Ye
JLJiawu Li
Jiawu Li
YHYushuang He
Yushuang He
Background/Objectives: Primary thyroid lymphoma (PTL), including diffuse large B-cell lymphoma (DLBCL) and mucosa-associated lymphoid tissue (MALT) lymphoma, share substantial overlap in ultrasound appearance with Hashimoto’s thyroiditis (HT), making preoperative differentiation challenging. This study aims to develop and validate a deep learning model based on B-mode ultrasound (BMUS) and color Doppler ultrasound (CDUS) for image-level differentiation of DLBCL, MALT lymphoma, and HT. Methods: This retrospective single-center study included 1294 ultrasound images from 290 patients (313 lesions) who underwent preoperative ultrasound examination at West China Hospital between September 2002 and September 2024. All images from the same lesion were assigned to the same data partition, and the dataset was split at the lesion level into training and test sets at an 8:2 ratio. A Frequency-Adaptive WT-ResNet model incorporating wavelet transform convolution and a frequency-adaptive gating mechanism was developed. The primary analysis was performed at the image level. The performance of the model was compared with that of three ultrasound physicians with different levels of experience. Grad-CAM was used for visual interpretation. An exploratory external validation was performed using an independent dataset from Sun Yat-sen Memorial Hospital. Results: In the test set, the model achieved a macro-average AUC of 0.927 (95% CI: 0.889–0.960), with class-specific AUCs of 0.899 for DLBCL, 0.946 for MALT lymphoma, and 0.937 for HT. The macro-average balanced accuracy was 0.866, compared with 0.827 for that of the best-performing senior physician. The exploratory validation set yielded a macro-average AUC of 0.796 (95% CI: 0.686–0.888), with class-specific AUCs of 0.806 for DLBCL, 0.825 for HT, and 0.756 for MALT lymphoma. Grad-CAM showed that the model focused on lesion-internal echotexture and lesion-transition regions with class-dependent patterns. Conclusions: A deep learning model based on BMUS and CDUS showed promising performance for image-level differentiation of DLBCL, MALT lymphoma and HT in a single-center retrospective cohort. The model outperformed three ultrasound physicians and may serve as a potential decision-support tool. However, the exploratory external validation results should be interpreted as preliminary, and larger multicenter cohorts remain necessary to confirm model generalizability.

16 August 2026 in Journal of Clinical Medicine
Verified
Journal Article
Open access
PDF file
JCJuanmei Chen
Juanmei Chen
ARAyibota Ruxian
Ayibota Ruxian
DLDanying Li
Danying Li
YJYong Jiang
Yong Jiang
BMBuyun Ma
Buyun Ma
Background/Objectives: Erdheim–Chester disease (ECD) is a rare non-Langerhans cell histiocytosis characterized by multisystem infiltration of foamy histiocytes, leading to chronic inflammation, fibrosis, and organ dysfunction. Breast involvement in ECD is extremely uncommon, and the sonographic features of ECD involving the breast remain poorly described. Case Presentation: We report the case of a 59-year-old woman with chronic bone pain and multisystem disease who experienced an extended diagnostic course despite undergoing renal biopsy, biopsy of a right elbow lesion, bone marrow examination, and multidisciplinary evaluation. Breast ultrasound revealed bilateral infiltrative hypoechoic lesions involving the breasts and axillae. These were classified as BI-RADS 4C and were highly suspicious for breast malignancy. Subsequently, an ultrasound-guided core needle biopsy was performed on the breast and axillary lesions. Results: Histopathology showed fibroadipose tissue infiltrated by numerous foamy histiocytes, scattered epithelioid cells, and occasional Touton giant cells. Immunohistochemistry showed positivity for CD68, CD163, CD4, and Cyclin D1, partial positivity for OCT2 and CD30, and negativity for S100, CD1a, Langerin, ALK, CK (Pan), and GATA3. The Ki-67 index was approximately 3%. Molecular testing detected the BRAF V600E mutation, supporting the diagnosis of ECD. A review of reported cases showed that breast involvement in ECD lacks specific ultrasound findings and may closely mimic primary breast malignancy. Conclusions: Breast involvement in ECD is rare and may present as bilateral infiltrative hypoechoic lesions with axillary involvement on ultrasound. In patients with chronic bone pain, symmetric osteosclerosis, or multisystem disease, ECD should be considered in the differential diagnosis. Ultrasound-detected superficial lesions may provide accessible biopsy targets, helping to establish a timely diagnosis and reduce diagnostic delay.