RATIONALE AND OBJECTIVE:Anaplastic thyroid carcinoma (ATC) may arise from the progression of differentiated thyroid cancers, but the mechanisms are not well understood. Establishing appropriate animal models and conducting ultrasound imaging monitoring can facilitate the study of this dedifferentiation process. MATERIALS AND METHODS:We constructed mouse models of papillary thyroid carcinoma (PTC) and ATC by editing the Braf mutation and Trp53 deletion, and monitored the process using small-animal ultrasound. The you only look once computer model was employed to process ultrasound images, analyze the characteristics of ultrasound images from different tumors and build a lesion identification and diagnostic model. The molecular expression characteristics of PTC with potential for dedifferentiation were analyzed through RNA sequencing. RESULTS:Mice ATCs exhibited a characteristic five-phase growth curve, accompanied by earlier lung metastasis. The mice thyroid tumor lesion recognition model was capable of identifying different types of tumor lesions and simulating their progression process with an accuracy of 0.765. Some PTCs showed lower Thyroid Differentiation Scores and were associated with high expression of genes such as Ptprn2, Tnfsf18 and Cdkn2a. CONCLUSION:This study validated the hypothesis that Trp53 deletion and Braf mutation promote the progression of PTC to ATC through ultrasound monitoring and computer modeling. Certain PTCs with the potential for dedifferentiation exhibit a specific molecular expression profile, which may represent potential therapeutic targets.

OBJECTIVES:This study aimed to develop and validate an ultrasound-based IHC4-associated radiomic model for predicting late recurrence in ER-positive breast cancer and to evaluate its potential to support risk stratification and guide decisions on extended endocrine therapy. METHODS:In this retrospective multicenter study, patients were divided into a training cohort, an internal validation cohort, and two external validation cohorts. Radiomic features associated with the immunohistochemical four-marker (IHC4) score were selected to construct a support vector machine (SVM)-based IHC4-associated radiomic model for generating an IHC4-associated radiomic score. Model performance was evaluated using the area under the receiver operating characteristic curve (AUC). Late distant recurrence (DR) was defined as the primary endpoint, and progression-free survival (PFS) was defined as the secondary endpoint. The prognostic value of the score was further evaluated using Cox regression analysis. RESULTS:A total of 523 patients were included in this study. Seven IHC4-associated radiomic features were selected to construct the IHC4-associated radiomic model. The model demonstrated consistent performance across cohorts, with AUCs of 0.81, 0.84, 0.79, and 0.81 in the training, internal validation, and two external validation cohorts, respectively. The IHC4-associated radiomic score stratified late DR risk across the training, internal validation, and external validation cohorts. In the secondary exploratory PFS analysis, the score remained associated with PFS in multivariable Cox analysis (HR = 4.246, 95% CI: 1.749-10.307, P = .001). CONCLUSIONS:The IHC4-associated radiomic model may serve as a noninvasive biomarker for predicting prognosis in ER-positive breast cancer.

Recently, numerous deep learning models have been proposed for breast cancer diagnosis using multimodal multi-view ultrasound images. However, their performance could be highly affected by overlooking interactions between different modalities and views. Moreover, existing methods struggle to handle cases where certain modalities or views are missing, which limits their clinical applications. To address these issues, we propose a novel Alignment and Imputation Network (AINet) by integrating 1) alignment and imputation pre-training, and 2) hierarchical fusion fine-tuning. Specifically, in the pre-training stage, cross-modal contrastive learning is employed to align features across different modalities, for effectively capturing inter-modal interactions. To simulate missing modality (view) scenarios, we randomly mask out features and then impute them by leveraging inter-modal and inter-view relationships. Following the clinical diagnosis procedure, the subsequent fine-tuning stage further incorporates modality-level and view-level fusion in a hierarchical manner. The proposed AINet is developed and evaluated on three datasets, comprising 15,223 subjects in total. Experimental results demonstrate that AINet significantly outperforms state-of-the-art methods, particularly in handling missing modalities (views). This highlights its robustness and potential for real-world clinical applications.

Objectives This study aimed to develop and validate an ultrasound‐based IHC4‐associated radiomic model for predicting late recurrence in ER‐positive breast cancer and to evaluate its potential to support risk stratification and guide decisions on extended endocrine therapy. Methods In this retrospective multicenter study, patients were divided into a training cohort, an internal validation cohort, and two external validation cohorts. Radiomic features associated with the immunohistochemical four‐marker (IHC4) score were selected to construct a support vector machine (SVM)‐based IHC4‐associated radiomic model for generating an IHC4‐associated radiomic score. Model performance was evaluated using the area under the receiver operating characteristic curve (AUC). Late distant recurrence (DR) was defined as the primary endpoint, and progression‐free survival (PFS) was defined as the secondary endpoint. The prognostic value of the score was further evaluated using Cox regression analysis. Results A total of 523 patients were included in this study. Seven IHC4‐associated radiomic features were selected to construct the IHC4‐associated radiomic model. The model demonstrated consistent performance across cohorts, with AUCs of 0.81, 0.84, 0.79, and 0.81 in the training, internal validation, and two external validation cohorts, respectively. The IHC4‐associated radiomic score stratified late DR risk across the training, internal validation, and external validation cohorts. In the secondary exploratory PFS analysis, the score remained associated with PFS in multivariable Cox analysis (HR = 4.246, 95% CI: 1.749–10.307, P = .001). Conclusions The IHC4‐associated radiomic model may serve as a noninvasive biomarker for predicting prognosis in ER‐positive breast cancer.

In clinical practice, triple-negative breast cancer (TNBC) patients with varying levels of lipid metabolism exhibit differences in tumor shear-wave elastography (SWE) stiffness and prognosis, but this association with unclear mechanism. In this study, a clinical cohort from FUSCC (n = 147) demonstrated that both elevated BMI and higher SWE stiffness were significantly associated with poorer long-term prognosis in TNBC patients, and these associations were further validated in multi-TNBC animal models. Our findings emphasize the role of SWE stiffness in capturing BMI-related alterations in the tumor mechanical microenvironment. Based on integrated lipidomic and transcriptomic analyses, we demonstrated that diacylglycerol (DAG) serves as a critical lipid molecule promoting elevated SWE stiffness and malignant progression. Mechanistically, DAG upregulates TGF-β1 expression through PKC-mediated enhancement of CREB1 phosphorylation in multiple TNBC cell lines, directly promoting TNBC progression and activating cancer-associated fibroblasts. This creates a self-sustaining feedback loop that accelerates malignancy. Finally, we confirmed that the DAG/PKC/CREB1/TGF-β1 signaling axis profoundly regulates SWE imaging stiffness in TNBC models, with further validation in clinical samples. Our study establishes SWE stiffness as a non-invasive imaging biomarker for the activation of this specific pro-metastatic pathway, providing a mechanistic basis for interpreting SWE features through a biological lens and paving the way for its application in prognosis prediction and tailored therapeutic strategies for high-risk TNBC patients.

In the clinical setting, the efficacy of single-agent immune checkpoint inhibitors (ICIs) in triple-negative breast cancer (TNBC) remains suboptimal. Therefore, there is a pressing need to develop predictive biomarkers to identify non-responders. Considering that cancer-associated fibroblasts (CAFs) represent an integral component of the tumor microenvironment that affects the stiffness of solid tumors on shear-wave elastography (SWE) imaging, wound healing CAFs (WH CAFs) were identified in highly heterogeneous TNBC. This subtype highly expressed vitronectin (VTN) and constituted the majority of CAFs. Moreover, WH CAFs were negatively correlated with CD8+ T cell infiltration levels and influenced tumor proliferation in the Eo771 mouse model. Furthermore, multi-omics analysis validated its role in immunosuppression. In order to non-invasively classify patients as responders or non-responders to ICI monotherapy, a deep learning model was constructed to classify the level of WH CAFs based on SWE imaging. As anticipated, this model effectively distinguished the level of WH CAFs in tumors. Based on the classification of the level of WH CAFs, while tumors with a high level of WH CAFs were found to exhibit a poor response to anti programmed cell death protein 1 (PD-1) monotherapy, they were responsive to the combination of anti-PD-1 and erdafitinib, a selective fibroblast growth factor receptor (FGFR) inhibitor. Overall, these findings establish a reference for a novel non-invasive method for predicting ICI efficacy to guide the selection of TNBC patients for precision treatment in clinical settings.

OBJECTIVES:Obesity is closely associated with the occurrence and progression of breast cancer. While body mass index (BMI) is widely used to diagnose obesity, it has certain limitations. Subcutaneous fat thickness (SFT) also serves as an indicator of body composition. However, studies on breast SFT are scarce. This study aims to investigate the relationship between BMI, breast SFT, and ultrasound features of breast cancer, as well as their associations with tumor proliferation and invasiveness. METHODS:This study retrospectively analyzed the relationship between BMI and clinical and ultrasound characteristics in 1670 patients. Among them, breast SFT was measured in 470 patients using mammography and ultrasound, and the correlation between SFT and BMI was assessed. The relationship between ultrasound-measured SFT and pathological as well as ultrasound features was also analyzed. The correlation between breast SFT, BMI, and somatic gene mutations was analyzed in 234 patients. RESULTS:Patients with BMI ≥24 kg/m2 exhibited more malignant ultrasound features. SFT measured by mammography was correlated with SFT measured by ultrasound (r = 0.565, p < .001). Both BMI and SFT measured via mammography (r = 0.578, p < .001) and ultrasound (r = 0.485, p < .001) showed significant correlations. Breast SFT varied significantly among tumors with different shapes (p = .025), boundaries (p < .001), and posterior echo features (p < .001). The area under the curve (AUC) for breast SFT predicting irregular shape, halo, and posterior shadowing was 0.605, 0.666, and 0.632, respectively, with cutoff values of 8.65, 8.35, and 8.35 mm. Patients with breast SFT ≥8.6 mm demonstrated significantly elevated Ki67 levels (p = .004). No differences in somatic mutation frequencies were found at a threshold of 8.60 mm for fat thickness or at a BMI of 24 kg/m2. However, at BMI ≥22 kg/m2, mutation frequencies were higher. CONCLUSIONS:BMI and breast SFT are associated with malignant ultrasound features. While both have diagnostic value, BMI is more reliable than fat thickness for predicting cancer proliferation and invasion, with a BMI threshold of 22 kg/m2 offering higher diagnostic value than 24 kg/m2.

Objectives Obesity is closely associated with the occurrence and progression of breast cancer. While body mass index (BMI) is widely used to diagnose obesity, it has certain limitations. Subcutaneous fat thickness (SFT) also serves as an indicator of body composition. However, studies on breast SFT are scarce. This study aims to investigate the relationship between BMI, breast SFT, and ultrasound features of breast cancer, as well as their associations with tumor proliferation and invasiveness.Methods This study retrospectively analyzed the relationship between BMI and clinical and ultrasound characteristics in 1670 patients. Among them, breast SFT was measured in 470 patients using mammography and ultrasound, and the correlation between SFT and BMI was assessed. The relationship between ultrasound-measured SFT and pathological as well as ultrasound features was also analyzed. The correlation between breast SFT, BMI, and somatic gene mutations was analyzed in 234 patients.Results Patients with BMI >= 24 kg/m2 exhibited more malignant ultrasound features. SFT measured by mammography was correlated with SFT measured by ultrasound (r = 0.565, p < .001). Both BMI and SFT measured via mammography (r = 0.578, p < .001) and ultrasound (r = 0.485, p < .001) showed significant correlations. Breast SFT varied significantly among tumors with different shapes (p = .025), boundaries (p < .001), and posterior echo features (p < .001). The area under the curve (AUC) for breast SFT predicting irregular shape, halo, and posterior shadowing was 0.605, 0.666, and 0.632, respectively, with cutoff values of 8.65, 8.35, and 8.35 mm. Patients with breast SFT >= 8.6 mm demonstrated significantly elevated Ki67 levels (p = .004). No differences in somatic mutation frequencies were found at a threshold of 8.60 mm for fat thickness or at a BMI of 24 kg/m2. However, at BMI >= 22 kg/m2, mutation frequencies were higher.Conclusions BMI and breast SFT are associated with malignant ultrasound features. While both have diagnostic value, BMI is more reliable than fat thickness for predicting cancer proliferation and invasion, with a BMI threshold of 22 kg/m2 offering higher diagnostic value than 24 kg/m2.

Cancer-associated fibroblasts (CAFs) play a pivotal role in inducing photothermal therapy (PTT) resistance of triple-negative breast cancer (TNBC), but with unclear mechanism. Herein, aminoethyl anisamide-modified nano-biomimetic low-density lipoprotein (A-aLDL) is used to target deliver the PTT agent and artesunate (ARS) to both CAFs and cancer cells. Though CAFs are sensitive to PTT and notably transition to heat-resistant phenotype, the formed protective barrier is destroyed by ARS. Subsequently, the outstanding anti-tumor effects are achieved through PTT in multiple models with such kind of combination therapy. Interestingly, the mechanism is discovered that serine metabolism plays a major role in CAF resistance through spatially omics. ARS disrupts serine homeostasis, thereby attenuating the cascade activity of GTPases in MAPK pathway. Meanwhile, MAP2K7 is the most potential target for sensitizing PTT. By integrating ARS with PTT agents, the serine-MAPK axis in CAFs is successfully modulated, thereby overcoming PTT resistance in TNBC therapy.

Spatial transcriptomics has revolutionized tissue biology by enabling spatially resolved gene expression profiling. Nonetheless, current spot-level spatial transcriptomic technologies consolidate signals from multiple cells, complicating cellular-level analysis. Moreover, matched single-cell references required by reference-based deconvolution methods are frequently unavailable. To overcome these limitations, we present SURF, a reference-free deconvolution tool that integrates high-dimensional gene data analysis with self-supervised deep learning to effectively model nonlinear gene interactions and leverage spot relationships. Benchmarking on both synthetic and real datasets shows that SURF consistently outperforms existing reference-free methods and exceeds reference-based approaches when appropriate references are absent. Applications across datasets with varying resolutions, species, spatial patterns, and tissue states demonstrate SURF's robust capacity to precisely represent tissue microenvironments. Importantly, SURF successfully identifies clinically significant epithelial-to-mesenchymal transition states within tumor regions in a dataset of human colorectal liver metastasis, highlighting its utility in uncovering critical biological mechanisms relevant to disease progression.

Breast cancer Mammography (MAM) screening was proven to improve survival worldwide. However, younger patients with higher breast density made MAM less effective in China. It is necessary to establish Chinese-specific effective screening strategies. This study aims to explore the efficacy of artificial intelligence (AI)-assisted ultrasound breast cancer screening in China. Eligible participants were those aged 35–69 years and were attending the Chinese "Two Cancer (breast and cervical cancer) Screening" program. Two districts were selected as cluster to receive either AI-assisted ultrasound screening or routine ultrasound screening. We obtained data on cancer diagnosis through active follow-up and linkage with municipal cancer registry. The primary outcome was improved screening sensitivity enabling the detection of more true-positive cases. This study is registered at ClinicalTrials.gov under the number NCT06521788 (Initial Release Date: 07/22/2024). A total of 21,790 individuals in two districts were included in this study, with 8,736 participants in Hongkou district receiving AI-assisted ultrasound screening and 13,054 in Pudong district undergoing routine ultrasound screening. Of the 21,790 screened participants, 232 (10.7‰) tested positive, with AI detecting similar positivity rates compared to routine screening (12.2‰ vs. 9.6‰, P = 0.07). After one year of follow-up, 49 participants were diagnosed with breast cancer: 30 were screen-detected cancers, and 19 were interval cancers. The AI group demonstrated a significantly higher screening sensitivity (75

Triple-negative breast cancer (TNBC) is highly malignant, with rapid tumor growth and metastasis. Due to ER-, PR- and HER2-of TNBC, FGFR pathway play a pivotal role in the progression of TNBC. Its ligand FGFs is mostly released from the extracellular matrix by fibroblast growth factor binding protein 1 (FGFBP1). However, little is known about the role of FGFBP1 in TNBC. In this study, we found that overexpression of FGFBP1 significantly promoted the proliferation, migration and invasion of TNBC cells in vitro and in vivo and vice versa. Mechanistically, overexpression of FGFBP1 upregulated the expression of KLK10, thereby activating AKT, which led to proliferation, migration and invasion of TNBC cells. After knocking down FGFBP1, the expression of KLK10 was reduced and the AKT pathway was inhibited. In addition, knocking down KLK10 or inhibiting AKT pathway impaired the promotion effect of overexpression of FGFBP1 on the proliferation and invasion of TNBC cells. These results suggest that FGFBP1 may promote the proliferation, migration and invasion of TNBC cells through the KLK10-AKT axis. Targeting FGFBP1 may serve as a new therapeutic strategy for TNBC.

Subvariants of testicular germ cell tumor (TGCT) significantly affect therapeutic strategies and patient prognosis. However, preoperatively distinguishing seminoma (SE) from non-seminoma (n-SE) remains a challenge. This study aimed to evaluate the performance of a deep learning-based super-resolution (SR) US radiomics model for SE/n-SE differentiation. This international multicenter retrospective study recruited patients with confirmed TGCT between 2015 and 2023. A pre-trained SR reconstruction algorithm was applied to enhance native resolution (NR) images. NR and SR radiomics models were constructed, and the superior model was then integrated with clinical features to construct clinical-radiomics models. Diagnostic performance was evaluated by ROC analysis (AUC) and compared with radiologists' assessments using the DeLong test. A total of 486 male patients were enrolled for training (n = 338), domestic (n = 92), and international (n = 59) validation sets. The SR radiomics model achieved AUCs of 0.90, 0.82, and 0.91, respectively, in the training, domestic, and international validation sets, significantly surpassing the NR model (p < 0.001, p = 0.031, and p = 0.001, respectively). The clinical-radiomics model exhibited a significantly higher across both domestic and international validation sets compared to the SR radiomics model alone (0.95 vs 0.82, p = 0.004; 0.97 vs 0.91, p = 0.031). Moreover, the clinical-radiomics model surpassed the performance of experienced radiologists in both domestic (AUC, 0.95 vs 0.85, p = 0.012) and international (AUC, 0.97 vs 0.77, p < 0.001) validation cohorts. The SR-based clinical-radiomics model can effectively differentiate between SE and n-SE. This international multicenter study demonstrated that a radiomics model of deep learning-based SR reconstructed US images enabled effective differentiation between SE and n-SE. Clinical parameters and radiologists' assessments exhibit limited diagnostic accuracy for SE/n-SE differentiation in TGCT. Based on scrotal US images of TGCT, the SR radiomics models performed better than the NR radiomics models. The SR-based clinical-radiomics model outperforms both the radiomics model and radiologists' assessment, enabling accurate, non-invasive preoperative differentiation between SE and n-SE.

Inadequate generality across different organs and tasks constrains the application of ultrasound (US) image analysis methods in smart healthcare. Building a universal US foundation model holds the potential to address these issues. Nevertheless, the development of such foundation models encounters intrinsic challenges in US analysis, i.e., insufficient databases, low quality, and ineffective features. In this paper, we present a universal US foundation model, named USFM, generalized to diverse tasks and organs towards label efficient US image analysis. First, a large-scale Multi-organ, Multi-center, and Multi-device US database was built, comprehensively containing over two million US images. Organ-balanced sampling was employed for unbiased learning. Then, USFM is self-supervised pre-trained on the sufficient US database. To extract the effective features from low-quality US images, we proposed a spatial-frequency dual masked image modeling method. A productive spatial noise addition-recovery approach was designed to learn meaningful US information robustly, while a novel frequency band-stop masking learning approach was also employed to extract complex, implicit grayscale distribution and textural variations. Extensive experiments were conducted on the various tasks of segmentation, classification, and image enhancement from diverse organs and diseases. Comparisons with representative US image analysis models illustrate the universality and effectiveness of USFM. The label efficiency experiments suggest the USFM obtains robust performance with only 20% annotation, laying the groundwork for the rapid development of US models in clinical practices.

ObjectiveThe objective of this study was to develop a deep learning-and-radiomics-based ultrasound nomogram for the evaluation of axillary lymph node (ALN) metastasis risk in breast cancer patients ≥ 75 years.MethodsThe study enrolled breast cancer patients ≥ 75 years who underwent either sentinel lymph node biopsy or ALN dissection at Fudan University Shanghai Cancer Center. DenseNet-201 was employed as the base model, and it was trained using the Adam optimizer and cross-entropy loss function to extract deep learning (DL) features from ultrasound images. Additionally, radiomics features were extracted from ultrasound images utilizing the Pyradiomics tool, and a Rad-Score (RS) was calculated employing the Lasso regression algorithm. A stepwise multivariable logistic regression analysis was conducted in the training set to establish a prediction model for lymph node metastasis, which was subsequently validated in the validation set. Evaluation metrics included area under the curve (AUC), accuracy, sensitivity, specificity, positive predictive value, negative predictive value, and F1-score. The calibration of the model’s performance and its clinical prediction accuracy were assessed using calibration curves and decision curves respectively. Furthermore, integrated discrimination improvement and net reclassification improvement were utilized to quantify enhancements in RS.ResultsHistological grade, axillary ultrasound, and RS were identified as independent risk factors for predicting lymph node metastasis. The integration of the RS into the clinical prediction model significantly improved its predictive performance, with an AUC of 0.937 in the training set, surpassing both the clinical model and the RS model alone. In the validation set, the integrated model also outperformed other models with AUCs of 0.906, 0.744, and 0.890 for the integrated model, clinical model, and RS model respectively. Experimental results demonstrated that this study’s integrated prediction model could enhance both accuracy and generalizability.ConclusionThe DL and radiomics-based model exhibited remarkable accuracy and reliability in predicting ALN status among breast cancer patients ≥ 75 years, thereby contributing to the enhancement of personalized treatment strategies’ efficacy and improvement of patients’ quality of life.

Domain shift problem is commonplace for ultrasound image analysis due to difference imaging setting and diverse medical centers, which lead to poor generalizability of deep learning-based methods. Multi-Source Domain Transformation (MSDT) provides a promising way to tackle the performance degeneration caused by the domain shift, which is more practical and challenging compared to conventional single-source transformation tasks. An effective unsupervised domain combination strategy is highly required to handle multiple domains without annotations. Fidelity and quality of generated images are also important to ensure the accuracy of computer-aided diagnosis. However, existing MSDT approaches underperform in above two areas. In this paper, an efficient domain transformation model named M2O-DiffGAN is introduced to achieve a unified mapping from multiple unlabeled source domains to the target domain. A cycle-consistent “many-to-one” adversarial learning architecture is introduced to model various unlabeled domains jointly. A condition adversarial diffusion process is employed to generate images with high-fidelity, combining an adversarial projector to capture reverse transition probabilities over large step sizes for accelerating sampling. Considering the limited perceptual information of ultrasound images, an ultrasound-specific content loss helps to capture more perceptual features for synthesizing high-quality ultrasound images. Massive comparisons on six clinical datasets covering thyroid, carotid and breast demonstrate the superiority of the M2O-DiffGAN in the performance of bridging the domain gaps and enlarging the generalization of downstream analysis methods compared to state-of-the-art algorithms. It improves the mean MI, Bhattacharyya Coefficient, dice and IoU assessments by 0.390, 0.120, 0.245 and 0.250, presenting promising clinical applications.

Deep learning (DL) has proven highly effective for ultrasound-based computer-aided diagnosis (CAD) of breast cancers. In an automatic CAD system, lesion detection is critical for the following diagnosis. However, existing DL-based methods generally require voluminous manually-annotated region of interest (ROI) labels and class labels to train both the lesion detection and diagnosis models. In clinical practice, the ROI labels, i.e. ground truths, may not always be optimal for the classification task due to individual experience of sonologists, resulting in the issue of coarse annotation to limit the diagnosis performance of a CAD model. To address this issue, a novel Two-Stage Detection and Diagnosis Network (TSDDNet) is proposed based on weakly supervised learning to improve diagnostic accuracy of the ultrasound-based CAD for breast cancers. In particular, all the initial ROI-level labels are considered as coarse annotations before model training. In the first training stage, a candidate selection mechanism is then designed to refine manual ROIs in the fully annotated images and generate accurate pseudo-ROIs for the partially annotated images under the guidance of class labels. The training set is updated with more accurate ROI labels for the second training stage. A fusion network is developed to integrate detection network and classification network into a unified end-to-end framework as the final CAD model in the second training stage. A self-distillation strategy is designed on this model for joint optimization to further improves its diagnosis performance. The proposed TSDDNet is evaluated on three B-mode ultrasound datasets, and the experimental results indicate that it achieves the best performance on both lesion detection and diagnosis tasks, suggesting promising application potential.

PurposeThe detection of human epidermal growth factor receptor 2 (HER2) expression status is essential to determining the chemotherapy regimen for breast cancer patients and to improving their prognosis. We developed a deep learning radiomics (DLR) model combining time-frequency domain features of ultrasound (US) video of breast lesions with clinical parameters for predicting HER2 expression status.Patients and MethodsData for this research was obtained from 807 breast cancer patients who visited from February 2019 to July 2020. Ultimately, 445 patients were included in the study. Pre-operative breast ultrasound examination videos were collected and split into a training set and a test set. Building a training set of DLR models combining time-frequency domain features and clinical features of ultrasound video of breast lesions based on the training set data to predict HER2 expression status. Test the performance of the model using test set data. The final models integrated with different classifiers are compared, and the best performing model is finally selected.ResultsThe best diagnostic performance in predicting HER2 expression status is provided by an Extreme Gradient Boosting (XGBoost)-based time-frequency domain feature classifier combined with a logistic regression (LR)-based clinical parameter classifier of clinical parameters combined DLR, particularly with a high specificity of 0.917. The area under the receiver operating characteristic curve (AUC) for the test cohort was 0.810.ConclusionOur study provides a non-invasive imaging biomarker to predict HER2 expression status in breast cancer patients.