Haizhu Xie

Major depressive disorder (MDD) is a clinically heterogeneous disorder. Its mechanism is still unknown. Although the altered intersubject variability in functional connectivity (IVFC) within gray‐matter has been reported in MDD, the alterations to IVFC within white‐matter (WM‐IVFC) remain unknown. Based on the resting‐state functional MRI data of discovery (145 MDD patients and 119 healthy controls [HCs]) and validation cohorts (54 MDD patients, and 78 HCs), we compared the WM‐IVFC between the two groups. We further assessed the meta‐analytic cognitive functions related to the alterations. The discriminant WM‐IVFC values were used to classify MDD patients and predict clinical symptoms in patients. In combination with the Allen Human Brain Atlas, transcriptome‐neuroimaging association analyses were further conducted to investigate gene expression profiles associated with WM‐IVFC alterations in MDD, followed by a set of gene functional characteristic analyses. We found extensive WM‐IVFC alterations in MDD compared to HCs, which were associated with multiple behavioral domains, including sensorimotor processes and higher‐order functions. The discriminant WM‐IVFC could not only effectively distinguish MDD patients from HCs with an area under curve ranging from 0.889 to 0.901 across three classifiers, but significantly predict depression severity (r = 0.575, p = 0.002) and suicide risk (r = 0.384, p = 0.040) in patients. Furthermore, the variability‐related genes were enriched for synapse, neuronal system, and ion channel, and predominantly expressed in excitatory and inhibitory neurons. Our results obtained good reproducibility in the validation cohort. These findings revealed intersubject functional variability changes of brain WM in MDD and its linkage with gene expression profiles, providing potential implications for understanding the high clinical heterogeneity of MDD.

Purpose: The authors aimed to establish an artificial intelligence (AI)-based method for preoperative diagnosis of breast lesions from contrast enhanced mammography (CEM) and to explore its biological mechanism. Materials and methods: This retrospective study includes 1430 eligible patients who underwent CEM examination from June 2017 to July 2022 and were divided into a construction set (n=1101), an internal test set (n=196), and a pooled external test set (n=133). The AI model adopted RefineNet as a backbone network, and an attention sub-network, named convolutional block attention module (CBAM), was built upon the backbone for adaptive feature refinement. An XGBoost classifier was used to integrate the refined deep learning features with clinical characteristics to differentiate benign and malignant breast lesions. The authors further retrained the AI model to distinguish in situ and invasive carcinoma among breast cancer candidates. RNA-sequencing data from 12 patients were used to explore the underlying biological basis of the AI prediction. Results: The AI model achieved an area under the curve of 0.932 in diagnosing benign and malignant breast lesions in the pooled external test set, better than the best-performing deep learning model, radiomics model, and radiologists. Moreover, the AI model has also achieved satisfactory results (an area under the curve from 0.788 to 0.824) for the diagnosis of in situ and invasive carcinoma in the test sets. Further, the biological basis exploration revealed that the high-risk group was associated with the pathways such as extracellular matrix organization. Conclusions: The AI model based on CEM and clinical characteristics had good predictive performance in the diagnosis of breast lesions.

Background: Accurate diagnosis of breast lesions and discrimination of axillary lymph node (ALN) metastases largely depend on radiologist experience.Purpose: To develop a deep learning‐based whole‐process system (DLWPS) for segmentation and diagnosis of breast lesions and discrimination of ALN metastasis.Study Type: Retrospective.Population: 1760 breast patients, who were divided into training and validation sets (1110 patients), internal (476 patients), and external (174 patients) test sets.Field Strength/Sequence: 3.0T/dynamic contrast‐enhanced (DCE)‐MRI sequence.Assessment: DLWPS was developed using segmentation and classification models. The DLWPS‐based segmentation model was developed by the U‐Net framework, which combined the attention module and the edge feature extraction module. The average score of the output scores of three networks was used as the result of the DLWPS‐based classification model. Moreover, the radiologists' diagnosis without and with the DLWPS‐assistance was explored. To reveal the underlying biological basis of DLWPS, genetic analysis was performed based on RNA‐sequencing data.Statistical Tests: Dice similarity coefficient (DI), area under receiver operating characteristic curve (AUC), accuracy, sensitivity, specificity, and kappa value.Results: The segmentation model reached a DI of 0.828 and 0.813 in the internal and external test sets, respectively. Within the breast lesions diagnosis, the DLWPS achieved AUCs of 0.973 in internal test set and 0.936 in external test set. For ALN metastasis discrimination, the DLWPS achieved AUCs of 0.927 in internal test set and 0.917 in external test set. The agreement of radiologists improved with the DLWPS‐assistance from 0.547 to 0.794, and from 0.848 to 0.892 in breast lesions diagnosis and ALN metastasis discrimination, respectively. Additionally, 10 breast cancers with ALN metastasis were associated with pathways of aerobic electron transport chain and cytoplasmic translation.Data Conclusion: The performance of DLWPS indicates that it can promote radiologists in the judgment of breast lesions and ALN metastasis and nonmetastasis.Level of Evidence: 4Technical Efficacy Stage: 3

Objective: To develop and test a contrast-enhanced mammography (CEM)-based radiomics model using intratumoral and peritumoral regions to predict non-sentinel lymph node (NSLN) metastasis in breast cancer before surgery. Methods: This multicenter study included 365 breast cancer patients with sentinel lymph node metastasis. Intratumoral regions of interest (ROIs) were manually delineated, and peritumoral ROIs (5 and 10 mm) were automatically obtained. Five models, including intratumoral model, peritumoral (5 and 10 mm) models, and intratumoral+peritumoral (5 and 10 mm) models, were constructed by support vector machine classifier on the basis of optimal features selected by variance threshold, SelectKbest, and least absolute shrinkage and selection operator algorithms. The predictive performance of radiomics models was evaluated by receiver operating characteristic curves. An external testing set was used to test the model. The Memorial Sloan Kettering Cancer Center (MSKCC) model was used to compare the predictive performance with radiomics model. Results: The intratumoral ROI and intratumoral+peritumoral 10-mm ROI-based radiomics model achieved the best performance with an area under the curve (AUC) of 0.8000 (95% confidence interval [CI]: 0.6871–0.8266) in the internal testing set. In the external testing set, the AUC of radiomics model was 0.7567 (95% CI: 0.6717–0.8678), higher than that of MSKCC model (AUC = 0.6681, 95% CI: 0.5148–0.8213) (p = 0.361). Conclusions: The intratumoral and peritumoral radiomics based on CEM had an acceptable predictive performance in predicting NSLN metastasis in breast cancer, which could be seen as a supplementary predicting tool to help clinicians make appropriate surgical plans. Advances in knowledge: The intratumoral and peritumoral CEM-based radiomics model could noninvasively predict NSLN metastasis in breast cancer patients before surgery.

Background: Characterization of the dynamics of functional brain network has gained increased attention in the study of depression. However, most studies have focused on single temporal dimension, while ignoring spatial dimensional information, hampering the discovery of validated biomarkers for depression.Purpose: To integrate temporal and spatial functional MRI variability features of dynamic brain network in machine‐learning techniques to distinguish patients with major depressive disorder (MDD) from healthy controls (HCs).Study Type: Prospective.Population: A discovery cohort including 119 patients and 106 HCs and an external validation cohort including 126 patients and 124 HCs from Rest‐meta‐MDD consortium.Field Strength/Sequence: A 3.0 T/resting‐state functional MRI using the gradient echo sequence.Assessment: A random forest (RF) model integrating temporal and spatial variability features of dynamic brain networks with separate feature selection method (M_SFS) was implemented for MDD classification. Its performance was compared with three RF models that used: temporal variability features (M_TVF), spatial variability features (M_SVF), and integrated temporal and spatial variability features with hybrid feature selection method (M_HFS). A linear regression model based on M_SFS was further established to assess MDD symptom severity, with prediction performance evaluated by the correlations between true and predicted scores.Statistical Tests: Receiver operating characteristic analyses with the area under the curve (AUC) were used to evaluate models' performance. Pearson's correlation was used to assess relationship of predicted scores and true scores. P < 0.05 was considered statistically significant.Results: The model with M_SFS achieved the best performance, with AUCs of 0.946 and 0.834 in the discovery and validation cohort, respectively. Additionally, altered temporal and spatial variability could significantly predict the severity of depression (r = 0.640) and anxiety (r = 0.616) in MDD.Data Conclusion: Integration of temporal and spatial variability features provides potential assistance for clinical diagnosis and symptom prediction of MDD.Evidence Level: 2.Technical Efficacy: Stage 2.

Background: Previous studies have explored the potential on radiomics features of primary breast cancer tumor to identify axillary lymph node (ALN) metastasis. However, the value of deep learning (DL) to identify ALN metastasis remains unclear.Purpose: To investigate the potential of the proposed attention‐based DL model for the preoperative differentiation of ALN metastasis in breast cancer on dynamic contrast‐enhanced MRI (DCE‐MRI).Study Type: Retrospective.Population: A total of 941 breast cancer patients who underwent DCE‐MRI before surgery were included in the training (742 patients), internal test (83 patients), and external test (116 patients) cohorts.Field Strength/Sequence: A 3.0 TMRscanner,DCE‐MRIsequence.Assessment: A DL model containing a 3D deep residual network (ResNet) architecture and a convolutional block attention module, named RCNet, was proposed for ALN metastasis identification. Three RCNet models were established based on the tumor, ALN, and combined tumor‐ALN regions on the images. The performance of these models was compared with ResNet models, radiomics models, the Memorial Sloan‐Kettering Cancer Center (MSKCC) model, and three radiologists (W.L., H.S., and F. L.).Statistical Tests: Dice similarity coefficient for breast tumor and ALN segmentation. Accuracy, sensitivity, specificity, intercorrelation and intracorrelation coefficients, area under the curve (AUC), and Delong test for ALN classification.Results: The optimal RCNet model, that is, RCNet_−tumor+ALN, achieved an AUC of 0.907, an accuracy of 0.831, a sensitivity of 0.824, and a specificity of 0.837 in the internal test cohort, as well as an AUC of 0.852, an accuracy of 0.828, a sensitivity of 0.792, and a specificity of 0.853 in the external test cohort. Additionally, with the assistance of RCNet_−tumor+ALN, the radiologists' performance was improved (external test cohort,P < 0.05).Data Conclusion: DCE‐MRI‐based RCNet model could provide a noninvasive auxiliary tool to identify ALN metastasis preoperatively in breast cancer, which may assist radiologists in conducting more accurate evaluation of ALN status.Evidence Level: 3Technical Efficacy: Stage 2

Background Postpartum depression (PPD) is a common mood disorder with increasing incidence year by year. However, the dynamic changes in local neural activity remain unclear. In this study, we utilized the dynamic amplitude of low-frequency (dALFF) to investigate the abnormal temporal variability of local neural activity. Methods Twenty-four patients with PPD and nineteen healthy postpartum women controls (HCs) matched for age, education level and body mass index were examined by resting-state functional magnetic resonance imaging (rs-fMRI). A sliding-window method was used to assess the dALFF, and a k-means clustering method was used to identify dALFF states. Two-sample t-test was used to compare the differences of dALFF variability and state metrics between PPD and HCs. Pearson correlation analysis was used to analyze the relationship between dALFF variability, states metrics and clinical severity. Results (1) Patients with PPD had lower variance of dALFF than HCs in the cognitive control network, cerebellar network, and sensorimotor network. (2) Four dALFF states were identified, and the number of transitions between the four dALFF states increased in the patients compared with that in HCs. (3) Multiple dALFF states were found to be correlated with the severity of depression. The variance of dALFF in the right middle frontal gyrus was negatively correlated with the Edinburgh postnatal depression scale score. Conclusion This study provides new insights into the brain dysfunction of PPD from the perspective of dynamic local brain activity, and highlights its important role in understanding the neurophysiological mechanisms of PPD.