Deep learning radiomics can predict axillary lymph node status in early-stage breast cancer

Utilizing multiclassifier radiomics analysis of ultrasound to predict high axillary lymph node tumour burden in node-positive breast cancer patients: a multicentre study

BACKGROUND

The tumor burden within the axillary lymph nodes (ALNs) constitutes a pivotal factor in breast cancer, serving as the primary determinant for treatment decisions and exhibiting a close correlation with prognosis.

OBJECTIVE

This study aimed to investigate the potential of ultrasound-based radiomics and clinical characteristics in non-invasively distinguishing between low tumor burden (1-2 positive nodes) and high tumor burden (more than 2 positive nodes) in patients with node-positive breast cancer.

METHODS

A total of 215 patients with node-positive breast cancer, who underwent preoperative ultrasound examinations, were enrolled in this study. Among these patients, 144 cases were allocated to the training set, 37 cases to the validation set, and 34 cases to the testing set. Postoperative histopathology was used to determine the status of ALN tumor burden. The region of interest for breast cancer was delineated on the ultrasound image. Nine models were developed to predict high ALN tumor burden, employing a combination of three feature screening methods and three machine learning classifiers. Ultimately, the optimal model was selected and tested on both the validation and testing sets. In addition, clinical characteristics were screened to develop a clinical model. Furthermore, Shapley additive explanations (SHAP) values were utilized to provide explanations for the machine learning model.

RESULTS

During the validation and testing sets, the models demonstrated area under the curve (AUC) values ranging from 0.577 to 0.733 and 0.583 to 0.719, and accuracies ranging from 64.9% to 75.7% and 64.7% to 70.6%, respectively. Ultimately, the Boruta_XGB model, comprising five radiomics features, was selected as the final model. The AUC values of this model for distinguishing low from high tumor burden were 0.828, 0.715, and 0.719 in the training, validation, and testing sets, respectively, demonstrating its superiority over the clinical model.

CONCLUSIONS

The developed radiomics models exhibited a significant level of predictive performance. The Boruta_XGB radiomics model outperformed other radiomics models in this study.

Combined deep learning and radiomics in pretreatment radiation esophagitis prediction for patients with esophageal cancer underwent volumetric modulated arc therapy

PURPOSE

To develop a combined radiomics and deep learning (DL) model in predicting radiation esophagitis (RE) of a grade ≥ 2 for patients with esophageal cancer (EC) underwent volumetric modulated arc therapy (VMAT) based on computed tomography (CT) and radiation dose (RD) distribution images.

MATERIALS AND METHODS

A total of 273 EC patients underwent VMAT were retrospectively reviewed and enrolled from two centers and divided into training (n = 152), internal validation (n = 66), and external validation (n = 55) cohorts, respectively. Radiomic and dosiomic features along with DL features using convolutional neural networks were extracted and screened from CT and RD images to predict RE. The performance of these models was evaluated and compared using the area under curve (AUC) of the receiver operating characteristic curves (ROC).

RESULTS

There were 5 and 10 radiomic and dosiomic features were screened, respectively. XGBoost achieved a best AUC of 0.703, 0.694 and 0.801, 0.729 with radiomic and dosiomic features in the internal and external validation cohorts, respectively. ResNet34 achieved a best prediction AUC of 0.642, 0.657 and 0.762, 0.737 for radiomics based DL model (DLR) and RD based DL model (DLD) in the internal and external validation cohorts, respectively. Combined model of DLD + Dosiomics + clinical factors achieved a best AUC of 0.913, 0.821 and 0.805 in the training, internal, and external validation cohorts, respectively.

CONCLUSION

Although the dose was not responsible for the prediction accuracy, the combination of various feature extraction methods was a factor in improving the RE prediction accuracy. Combining DLD with dosiomic features was promising in the pretreatment prediction of RE for EC patients underwent VMAT.

Deep learning applications in breast cancer histopathological imaging: diagnosis, treatment, and prognosis

Breast cancer is the most common malignant tumor among women worldwide and remains one of the leading causes of death among women. Its incidence and mortality rates are continuously rising. In recent years, with the rapid advancement of deep learning (DL) technology, DL has demonstrated significant potential in breast cancer diagnosis, prognosis evaluation, and treatment response prediction. This paper reviews relevant research progress and applies DL models to image enhancement, segmentation, and classification based on large-scale datasets from TCGA and multiple centers. We employed foundational models such as ResNet50, Transformer, and Hover-net to investigate the performance of DL models in breast cancer diagnosis, treatment, and prognosis prediction. The results indicate that DL techniques have significantly improved diagnostic accuracy and efficiency, particularly in predicting breast cancer metastasis and clinical prognosis. Furthermore, the study emphasizes the crucial role of robust databases in developing highly generalizable models. Future research will focus on addressing challenges related to data management, model interpretability, and regulatory compliance, ultimately aiming to provide more precise clinical treatment and prognostic evaluation programs for breast cancer patients.

Radiomics in breast cancer: Current advances and future directions

Breast cancer is a common disease that causes great health concerns to women worldwide. During the diagnosis and treatment of breast cancer, medical imaging plays an essential role, but its interpretation relies on radiologists or clinical doctors. Radiomics can extract high-throughput quantitative imaging features from images of various modalities via traditional machine learning or deep learning methods following a series of standard processes. Hopefully, radiomic models may aid various processes in clinical practice. In this review, we summarize the current utilization of radiomics for predicting clinicopathological indices and clinical outcomes. We also focus on radio-multi-omics studies that bridge the gap between phenotypic and microscopic scale information. Acknowledging the deficiencies that currently hinder the clinical adoption of radiomic models, we discuss the underlying causes of this situation and propose future directions for advancing radiomics in breast cancer research.

Multiview deep learning networks based on automated breast volume scanner images for identifying breast cancer in BI-RADS 4

Objectives

To develop and validate a deep learning (DL) based automatic segmentation and classification system to classify benign and malignant BI-RADS 4 lesions imaged with ABVS.

Methods

From May to December 2020, patients with BI-RADS 4 lesions from Centre 1 and Centre 2 were retrospectively enrolled and divided into a training set (Centre 1) and an independent test set (Centre 2). All included patients underwent an ABVS examination within one week before the biopsy. A two-stage DL framework consisting of an automatic segmentation module and an automatic classification module was developed. The preprocessed ABVS images were input into the segmentation module for BI-RADS 4 lesion segmentation. The classification model was constructed to extract features and output the probability of malignancy. The diagnostic performances among different ABVS views (axial, sagittal, coronal, and multi-view) and DL architectures (Inception-v3, ResNet 50, and MobileNet) were compared.

Results

A total of 251 BI-RADS 4 lesions from 216 patients were included (178 in the training set and 73 in the independent test set). The average Dice coefficient, precision, and recall of the segmentation module in the test set were 0.817 ± 0.142, 0.903 ± 0.183, and 0.886 ± 0.187, respectively. The DL model based on multiview ABVS images and Inception-v3 achieved the best performance, with an AUC, sensitivity, specificity, PPV, and NPV of 0.949 (95% CI: 0.945-0.953), 82.14%, 95.56%, 92.00%, and 89.58%, respectively, in the test set.

Conclusions

The developed multiview DL model enables automatic segmentation and classification of BI-RADS 4 lesions in ABVS images.

Ultrasound Spectral Combined With Clinical Pathological Parameters in Prediction of Axillary Lymph Node Metastatic in Breast Cancer

OBJECTIVES

To explore the clinical value of the nomogram based on ultrasound spectral combined with clinical pathological parameter in predicting axillary lymph node metastasis in breast cancer.

METHODS

We prospectively gathered clinicopathologic and ultrasonic data from 240 patients confirmed breast cancer. The risk factors of axillary lymph node metastasis were analyzed by univariate and multivariate logistic regression, and the prediction model was established. The model calibration, predictive ability, and diagnostic efficiency in the training set and the testing set were analyzed by receiver operating characteristic curve and calibration curve analysis, respectively.

RESULTS

Univariate analysis showed that lymph node metastasis was related with tumor size, Ki-67, axillary ultrasound, ultrasound spectral quantitative parameter, internal echo, and calcification (P < .05). Multivariate logistic regression analysis showed that the Ki-67, axillary ultrasound, quantitative parameter (the mean of the mid-band fit in tumor and posterior tumor) were independent risk factors of axillary lymph node metastasis (P < .05). The models developed using Ki-67, axillary ultrasound, and quantitative parameters for predicting axillary lymph node metastasis demonstrated an area under the receiver operating characteristic curve of 0.83. Additionally, the prediction model exhibited outstanding predictability for axillary lymph node metastasis, as evidenced by a Harrell C-index of 0.83 (95% confidence interval 0.73-0.93).

CONCLUSION

Axillary ultrasound combined with Ki-67 and ultrasound spectral parameters has the potential to predict axillary lymph node metastasis in breast cancer, which is superior to axillary ultrasound alone.

Deep Learning for Distinguishing Mucinous Breast Carcinoma From Fibroadenoma on Ultrasound

PURPOSE

Mucinous breast carcinoma (MBC) tends to be misdiagnosed as fibroadenomas (FA) due to its benign imaging characteristics. We aimed to develop a deep learning (DL) model to differentiate MBC and FA based on ultrasound (US) images. The model could contribute to the diagnosis of MBC for radiologists.

METHODS

In this retrospective study, 884 eligible patients (700 FA patients and 184 MBC patients) with 2257 US images were enrolled. The images were randomly divided into a training set (n = 1805 images) and a test set (n = 452 images) in a ratio of 8:2. First, we used the training set to establish DL model, DL+ age-cutoff model and DL+ age-tree model. Then, we compared the diagnostic performance of three models to get the optimal model. Finally, we evaluated the diagnostic performance of radiologists (4 junior and 4 senior radiologists) with and without the assistance of the optimal model in the test set.

RESULTS

The DL+ age-tree model yielded higher areas under the receiver operating characteristic curve (AUC) than DL model and DL+ age-cutoff model (0.945 vs. 0.835, P < .001; 0.945 vs. 0.931, P < .001, respectively). With the assistance of DL+ age-tree model, both junior and senior radiologists' AUC had significant improvement (0.746-0.818, P = .010, 0.827-0.860, P = .005, respectively).

CONCLUSIONS

The DL+ age-tree model based on US images and age showed excellent performance in the differentiation of MBC and FA. Moreover, it can effectively improve the performance of radiologists with different degrees of experience that may contribute to reducing the misdiagnosis of MBC.

Revolutionizing breast cancer Ki-67 diagnosis: ultrasound radiomics and fully connected neural networks (FCNN) combination method

PURPOSE

This study aims to assess the diagnostic value of ultrasound habitat sub-region radiomics feature parameters using a fully connected neural networks (FCNN) combination method L2,1-norm in relation to breast cancer Ki-67 status.

METHODS

Ultrasound images from 528 cases of female breast cancer at the Affiliated Hospital of Xiangnan University and 232 cases of female breast cancer at the Affiliated Rehabilitation Hospital of Xiangnan University were selected for this study. We utilized deep learning methods to automatically outline the gross tumor volume and perform habitat clustering. Subsequently, habitat sub-regions were extracted to identify radiomics features and underwent feature engineering using the L1,2-norm. A prediction model for the Ki-67 status of breast cancer patients was then developed using a FCNN. The model's performance was evaluated using accuracy, area under the curve (AUC), specificity (Spe), positive predictive value (PPV), negative predictive value (NPV), Recall, and F1. In addition, calibration curves and clinical decision curves were plotted for the test set to visually assess the predictive accuracy and clinical benefit of the models.

RESULT

Based on the feature engineering using the L1,2-norm, a total of 9 core features were identified. The predictive model, constructed by the FCNN model based on these 9 features, achieved the following scores: ACC 0.856, AUC 0.915, Spe 0.843, PPV 0.920, NPV 0.747, Recall 0.974, and F1 0.890. Furthermore, calibration curves and clinical decision curves of the validation set demonstrated a high level of confidence in the model's performance and its clinical benefit.

CONCLUSION

Habitat clustering of ultrasound images of breast cancer is effectively supported by the combined implementation of the L1,2-norm and FCNN algorithms, allowing for the accurate classification of the Ki-67 status in breast cancer patients.

Characterizing Sentinel Lymph Node Status in Breast Cancer Patients Using a Deep-Learning Model Compared With Radiologists' Analysis of Grayscale Ultrasound and Lymphosonography

ABSTRACT

The objective of the study was to use a deep learning model to differentiate between benign and malignant sentinel lymph nodes (SLNs) in patients with breast cancer compared to radiologists' assessments.Seventy-nine women with breast cancer were enrolled and underwent lymphosonography and contrast-enhanced ultrasound (CEUS) examination after subcutaneous injection of ultrasound contrast agent around their tumor to identify SLNs. Google AutoML was used to develop image classification model. Grayscale and CEUS images acquired during the ultrasound examination were uploaded with a data distribution of 80% for training/20% for testing. The performance metric used was area under precision/recall curve (AuPRC). In addition, 3 radiologists assessed SLNs as normal or abnormal based on a clinical established classification. Two-hundred seventeen SLNs were divided in 2 for model development; model 1 included all SLNs and model 2 had an equal number of benign and malignant SLNs. Validation results model 1 AuPRC 0.84 (grayscale)/0.91 (CEUS) and model 2 AuPRC 0.91 (grayscale)/0.87 (CEUS). The comparison between artificial intelligence (AI) and readers' showed statistical significant differences between all models and ultrasound modes; model 1 grayscale AI versus readers, P = 0.047, and model 1 CEUS AI versus readers, P < 0.001. Model 2 r grayscale AI versus readers, P = 0.032, and model 2 CEUS AI versus readers, P = 0.041.The interreader agreement overall result showed κ values of 0.20 for grayscale and 0.17 for CEUS.In conclusion, AutoML showed improved diagnostic performance in balance volume datasets. Radiologist performance was not influenced by the dataset's distribution.

Deep learning-based computer-aided diagnosis system for the automatic detection and classification of lateral cervical lymph nodes on original ultrasound images of papillary thyroid carcinoma: a prospective diagnostic study

PURPOSE

This study aims to develop a deep learning-based computer-aided diagnosis (CAD) system for the automatic detection and classification of lateral cervical lymph nodes (LNs) on original ultrasound images of papillary thyroid carcinoma (PTC) patients.

METHODS

A retrospective data set of 1801 cervical LN ultrasound images from 1675 patients with PTC and a prospective test set including 185 images from 160 patients were collected. Four different deep leaning models were trained and validated in the retrospective data set. The best model was selected for CAD system development and compared with three sonographers in the retrospective and prospective test sets.

RESULTS

The Deformable Detection Transformer (DETR) model showed the highest diagnostic efficacy, with a mean average precision score of 86.3% in the retrospective test set, and was therefore used in constructing the CAD system. The detection performance of the CAD system was superior to the junior sonographer and intermediate sonographer with accuracies of 86.3% and 92.4% in the retrospective and prospective test sets, respectively. The classification performance of the CAD system was better than all sonographers with the areas under the curve (AUCs) of 94.4% and 95.2% in the retrospective and prospective test sets, respectively.

CONCLUSIONS

This study developed a Deformable DETR model-based CAD system for automatically detecting and classifying lateral cervical LNs on original ultrasound images, which showed excellent diagnostic efficacy and clinical utility. It can be an important tool for assisting sonographers in the diagnosis process.

A deep learning model for personalized intra-arterial therapy planning in unresectable hepatocellular carcinoma: a multicenter retrospective study

Background

Unresectable Hepatocellular Carcinoma (uHCC) poses a substantial global health challenge, demanding innovative prognostic and therapeutic planning tools for improved patient management. The predominant treatment strategies include Transarterial chemoembolization (TACE) and hepatic arterial infusion chemotherapy (HAIC).

Methods

Between January 2014 and November 2021, a total of 1725 uHCC patients [mean age, 52.8 ± 11.5 years; 1529 males] received preoperative CECT scans and were eligible for TACE or HAIC. Patients were assigned to one of the four cohorts according to their treatment, four transformer models (SELECTION) were trained and validated on each cohort; AUC was used to determine the prognostic performance of the trained models. Patients were stratified into high and low-risk groups based on the survival scores computed by SELECTION. The proposed AI-based treatment decision model (ATOM) utilizes survival scores to further inform final therapeutic recommendation.

Findings

In this study, the training and validation sets included 1448 patients, with an additional 277 patients allocated to the external validation sets. The SELECTION model outperformed both clinical models and the ResNet approach in terms of AUC. Specifically, SELECTION-TACE and SELECTION-HAIC achieved AUCs of 0.761 (95% CI, 0.693-0.820) and 0.805 (95% CI, 0.707-0.881) respectively, in predicting ORR in their external validation cohorts. In predicting OS, SELECTION-TC and SELECTION-HC demonstrated AUCs of 0.736 (95% CI, 0.608-0.841) and 0.748 (95% CI, 0.599-0.865) respectively, in their external validation sets. SELECTION-derived survival scores effectively stratified patients into high and low-risk groups, showing significant differences in survival probabilities (P < 0.05 across all four cohorts). Additionally, the concordance between ATOM and clinician recommendations was associated with significantly higher response/survival rates in cases of agreement, particularly within the TACE, HAIC, and TC cohorts in the external validation sets (P < 0.05).

Interpretation

ATOM was proposed based on SELECTION-derived survival scores, emerges as a promising tool to inform the selection among different intra-arterial interventional therapy techniques.

Funding

This study received funding from the Beijing Municipal Natural Science Foundation, China (Z190024); the Key Program of the National Natural Science Foundation of China, China (81930119); The Science and Technology Planning Program of Beijing Municipal Science & Technology Commission and Administrative Commission of Zhongguancun Science Park, China (Z231100004823012); Tsinghua University Initiative Scientific Research Program of Precision Medicine, China (10001020108); and Institute for Intelligent Healthcare, Tsinghua University, China (041531001).

An unsupervised learning model based on CT radiomics features accurately predicts axillary lymph node metastasis in breast cancer patients: diagnostic study

BACKGROUND

The accuracy of traditional clinical methods for assessing the metastatic status of axillary lymph nodes (ALNs) is unsatisfactory. In this study, the authors propose the use of radiomic technology and three-dimensional (3D) visualization technology to develop an unsupervised learning model for predicting axillary lymph node metastasis in patients with breast cancer (BC), aiming to provide a new method for clinical axillary lymph node assessment in patients with this disease.

METHODS

In this study, we retrospectively analyzed the data of 350 patients with invasive BC who underwent lung-enhanced computed tomography (CT) and axillary lymph node dissection surgery at the Department of Breast Surgery of the Second Xiangya Hospital of Central South University. The authors used 3D visualization technology to create a 3D atlas of ALNs and identified the region of interest for the lymph nodes. Radiomic features were subsequently extracted and selected, and a prediction model for ALNs was constructed using the K-means unsupervised algorithm. To validate the model, the authors prospectively collected data from 128 BC patients who were clinically evaluated as negative at our center.

RESULTS

Using 3D visualization technology, we extracted and selected a total of 36 CT radiomics features. The unsupervised learning model categorized 1737 unlabeled lymph nodes into two groups, and the analysis of the radiomic features between these groups indicated potential differences in lymph node status. Further validation with 1397 labeled lymph nodes demonstrated that the model had good predictive ability for axillary lymph node status, with an area under the curve of 0.847 (0.825-0.869). Additionally, the model's excellent predictive performance was confirmed in the 128 axillary clinical assessment negative cohort (cN0) and the 350 clinical assessment positive (cN+) cohort, for which the correct classification rates were 86.72 and 87.43%, respectively, which were significantly greater than those of clinical assessment methods.

CONCLUSIONS

The authors created an unsupervised learning model that accurately predicts the status of ALNs. This approach offers a novel solution for the precise assessment of ALNs in patients with BC.

The value of quantitative shear wave elastography combined with conventional ultrasound in evaluating and guiding fine needle aspiration biopsy of axillary lymph node for early breast cancer: implication for axillary surgical stage

OBJECTIVES

To investigate the value of conventional ultrasonography (US) combined with quantitative shear wave elastography (SWE) in evaluating and identifying target axillary lymph node (TALN) for fine needle aspiration biopsy (FNAB) of patients with early breast cancer.

MATERIALS AND METHODS

A total of 222 patients with 223 ALNs were prospectively recruited from January 2018 to December 2021. All TALNs were evaluated by US, SWE and subsequently underwent FNAB. The diagnostic performances of US, SWE, UEor (either US or SWE was positive) and UEand (both US and SWE were positive), and FNAB guided by the above four methods for evaluating ALN status were assessed using receiver operator characteristic curve (ROC) analyses. Univariate and multivariate logistic regression analyses used to determine the independent predictors of axillary burden.

RESULTS

The area under the ROC curve (AUC) for diagnosing ALNs using conventional US and SWE were 0.69 and 0.66, respectively, with sensitivities of 78.00% and 65.00% and specificities of 60.98% and 66.67%. The combined method, UEor, demonstrated significantly improved sensitivity of 86.00% (p < 0.001 when compared with US and SWE alone). The AUC of the UEor-guided FNAB [0.85 (95% CI, 0.80-0.90)] was significantly higher than that of US-guided FNAB [0.83 (95% CI, 0.78-0.88), p = 0.042], SWE-guided FNAB [0.79 (95% CI, 0.72-0.84), p = 0.001], and UEand-guided FNAB [0.77 (95% CI, 0.71-0.82), p < 0.001]. Multivariate logistic regression showed that FNAB and number of suspicious ALNs were found independent predictors of axillary burden in patients with early breast cancer.

CONCLUSION

The UEor had superior sensitivity compared to US or SWE alone in ALN diagnosis. The UEor-guided FNAB achieved a lower false-negative rate compared to FNAB guided solely by US or SWE, which may be a promising tool for the preoperative diagnosis of ALNs in early breast cancer, and had the potential implication for the selection of axillary surgical modality.

Retrieval of the Clipped Axillary Lymph Node and Its Impact on Treatment Decisions

We examined clinically node-positive (cN+) breast cancer patients undergoing neoadjuvant chemotherapy and clipped lymph node (CLN) localization to determine the rate of CLN = non-sentinel lymph node (SLN), the factors associated with cN+ to pN0 conversion, and the treatment impact. We conducted a single institution review of cN+ patients receiving NAC from 2016 to 2022 with preoperative CLN localization (N = 81). Demographics, hormone receptor (HR) and HER2 status, time to surgery, staging, chemotherapy regimen, localization method, pathology, and adjuvant therapy were analyzed. Pathologic complete response (pCR) of the CLN was observed in 41 patients (50.6%): 18.8% HR+/HER2-, 75% HR+/HER2+, 75% HR-/HER2+, and 62.5% triple-negative breast cancer (p-value = 0.006). CLN = SLN in 68 (84%) patients, while CLN = non-SLN in 13 (16%). In 14 (17.3%) patients, the final treatment was altered based on +CLN status: 11 patients underwent axillary lymph node dissection (ALND), and 3 had systemic treatment changes. pCR rates varied, with the highest conversion rates observed in HER2+ disease and the lowest in HR+/HER2- disease. In 2 (2.5%) patients, adjuvant therapy changes were made based on a non-sentinel CLN, while in 97.5% of patients, a SLN biopsy alone represented the status of the axilla. This demonstrates that a +CLN often alters final plans and that, despite also being a SLN in most cases, a subset of patients will be undertreated by SLN biopsy alone.

Preoperative Ultrasound Radomics to Predict Posthepatectomy Liver Failure in Patients With Hepatocellular Carcinoma

PURPOSE

Posthepatectomy liver failure (PHLF) is a major cause of postoperative mortality in hepatocellular carcinoma (HCC) patients. The study aimed to develop a method based on the two-dimensional shear wave elastography and clinical data to evaluate the risk of PHLF in HCC patients with chronic hepatitis B.

METHODS

This multicenter study proposed a deep learning model (PHLF-Net) incorporating dual-modal ultrasound features and clinical indicators to predict the PHLF risk. The datasets were divided into a training cohort, an internal validation cohort, an internal independent testing cohort, and three external independent testing cohorts. Based on ResNet50 pretrained on ImageNet, PHLF-Net used a progressive training strategy with images of varying granularity and incorporated conventional B-mode and elastography images and clinical indicators related to liver reserve function.

RESULTS

In total, 532 HCC patients who underwent hepatectomy at five hospitals were enrolled. PHLF occurred in 147 patients (27.6%, 147/532). The PHLF-Net combining dual-modal ultrasound and clinical indicators demonstrated high effectiveness for predicting PHLF, with AUCs of 0.957 and 0.923 in the internal validation and testing sets, and AUCs of 0.950, 0.860, and 1.000 in the other three independent external testing sets. The performance of PHLF-Net outperformed models of single- and dual-modal US.

CONCLUSIONS

Preoperative ultrasound imaging combining clinical indicators can effectively predict the PHLF probability in patients with HCC. In the internal and external validation sets, PHLF-Net demonstrated its usefulness in predicting PHLF.

Assessing Axillary Lymph Node Burden and Prognosis in cT1-T2 Stage Breast Cancer Using Machine Learning Methods: A Retrospective Dual-Institutional MRI Study

BACKGROUND

Pathological axillary lymph node (pALN) burden is an important factor for treatment decision-making in clinical T1-T2 (cT1-T2) stage breast cancer. Preoperative assessment of the pALN burden and prognosis aids in the individualized selection of therapeutic approaches.

PURPOSE

To develop and validate a machine learning (ML) model based on clinicopathological and MRI characteristics for assessing pALN burden and survival in patients with cT1-T2 stage breast cancer.

STUDY TYPE

Retrospective.

POPULATION

A total of 506 females (range: 24-83 years) with cT1-T2 stage breast cancer from two institutions, forming the training (N = 340), internal validation (N = 85), and external validation cohorts (N = 81), respectively.

FIELD STRENGTH/SEQUENCE

This study used 1.5-T, axial fat-suppressed T2-weighted turbo spin-echo sequence and axial three-dimensional dynamic contrast-enhanced fat-suppressed T1-weighted gradient echo sequence.

ASSESSMENT

Four ML methods (eXtreme Gradient Boosting [XGBoost], Support Vector Machine, k-Nearest Neighbor, Classification and Regression Tree) were employed to develop models based on clinicopathological and MRI characteristics. The performance of these models was evaluated by their discriminative ability. The best-performing model was further analyzed to establish interpretability and used to calculate the pALN score. The relationships between the pALN score and disease-free survival (DFS) were examined.

STATISTICAL TESTS

Chi-squared test, Fisher's exact test, univariable logistic regression, area under the curve (AUC), Delong test, net reclassification improvement, integrated discrimination improvement, Hosmer-Lemeshow test, log-rank, Cox regression analyses, and intraclass correlation coefficient were performed. A P-value <0.05 was considered statistically significant.

RESULTS

The XGB II model, developed based on the XGBoost algorithm, outperformed the other models with AUCs of 0.805, 0.803, and 0.818 in the three cohorts. The Shapley additive explanation plot indicated that the top variable in the XGB II model was the Node Reporting and Data System score. In multivariable Cox regression analysis, the pALN score was significantly associated with DFS (hazard ratio: 4.013, 95% confidence interval: 1.059-15.207).

DATA CONCLUSION

The XGB II model may allow to evaluate pALN burden and could provide prognostic information in cT1-T2 stage breast cancer patients.

LEVEL OF EVIDENCE

3 TECHNICAL EFFICACY: Stage 2.

Peritumoral edema enhances MRI-based deep learning radiomic model for axillary lymph node metastasis burden prediction in breast cancer

To investigate whether peritumoral edema (PE) could enhance deep learning radiomic (DLR) model in predicting axillary lymph node metastasis (ALNM) burden in breast cancer. Invasive breast cancer patients with preoperative MRI were retrospectively enrolled and categorized into low (< 2 lymph nodes involved (LNs+)) and high (≥ 2 LNs+) burden groups based on surgical pathology. PE was evaluated on T2WI, and intra- and peri-tumoral radiomic features were extracted from MRI-visible tumors in DCE-MRI. Deep learning models were developed for LN burden prediction in the training cohort and validated in an independent cohort. The incremental value of PE was evaluated through receiver operating characteristic (ROC) analysis, confirming the improvement in the area under the curve (AUC) using the Delong test. This was complemented by net reclassification improvement (NRI) and integrated discrimination improvement (IDI) metrics. The deep learning combined model, incorporating PE with selected radiomic features, demonstrated significantly higher AUC values compared to the MRI model and the DLR model in the training cohort (n = 177) (AUC: 0.953 vs. 0.849 and 0.867, p < 0.05) and the validation cohort (n = 111) (AUC: 0.963 vs. 0.883 and 0.882, p < 0.05). The complementary analysis demonstrated that PE significantly enhances the prediction performance of the DLR model (Categorical NRI: 0.551, p < 0.001; IDI = 0.343, p < 0.001). These findings were confirmed in the validation cohort (Categorical NRI: 0.539, p < 0.001; IDI = 0.387, p < 0.001). PE improved preoperative ALNM burden prediction of DLR model, facilitating personalized axillary management in breast cancer patients.

Shear-wave elastography as a supplementary tool for axillary staging in patients undergoing breast cancer diagnosis

BACKGROUND

Preoperative evaluation of axillary lymph node status is crucial for the selection of both systemic and surgical treatment in early breast cancer. This study assessed the particular role of additional shear wave elastography (SWE) in axillary staging in patients undergoing initial breast cancer diagnostics.

METHODS

One hundred patients undergoing axillary lymph node biopsy due to a sonographically suspicious axillary lymph node were prospectively evaluated with SWE using virtual touch tissue imaging quantification (VTIQ). Mean values of tissue stiffness for axillary tissue and lymph node tissue were measured prior to core-cut biopsy of the lymph node. All lymph nodes were clip-marked during the biopsy. Cut-off values to differentiate between malignant and benign lymph nodes were defined using Youden's index.

RESULTS

Lymph nodes with evidence of malignant tumor cells in the final pathological examination showed a significantly higher velocity as measured by SWE, with a mean velocity of 3.48 ± 1.58 m/s compared to 2.33 ± 0.62 m/s of benign lymph nodes (p < 0.0001). The statistically optimal cutoff to differentiate between malignant and benign lymph nodes was 2.66 m/s with a sensitivity of 69.8% and a specificity of 87.5%.

CONCLUSIONS

Lymph node metastases assessed with SWE showed significantly higher elasticity values compared to benign lymph nodes. Thus, SWE provides an additional useful and quantifiable parameter for the sonographic assessment of suspicious axillary lymph nodes in the context of pre-therapeutic axillary staging in order to differentiate between benign and metastatic processes and support the guidance of definitive biopsy work-up.

CRITICAL RELEVANCE STATEMENT

Shear-wave elastography provides an additional useful and quantifiable parameter for the assessment of suspicious axillary lymph nodes in the context of pre-therapeutic axillary staging in order to differentiate between benign and metastatic processes and support guiding the definitive biopsy work-up.

KEY POINTS

SWE is a quantifiable ultrasound parameter in breast cancer diagnosis. SWE shows a significantly higher velocity in malignant lymph nodes. SWE is useful in improving the sensitivity and specificity of axillary staging.

Integrating a deep neural network and Transformer architecture for the automatic segmentation and survival prediction in cervical cancer

Background

Automated tumor segmentation and survival prediction are critical to clinical diagnosis and treatment. This study aimed to develop deep-learning models for automatic tumor segmentation and survival prediction in magnetic resonance imaging (MRI) of cervical cancer (CC) by combining deep neural networks and Transformer architecture.

Methods

This study included 406 patients with CC, each with comprehensive clinical information and MRI scans. We randomly divided patients into training, validation, and independent test cohorts in a 6:2:2 ratio. During the model training, we employed two architecture types: one being a hybrid model combining convolutional neural network (CNN) and ransformer (CoTr) and one of pure CNNs. For survival prediction, the hybrid model combined tumor image features extracted by segmentation models with clinical information. The performance of the segmentation models was evaluated using the Dice similarity coefficient (DSC) and 95% Hausdorff distance (HD95). The performance of the survival models was assessed using the concordance index.

Results

The CoTr model performed well in both contrast-enhanced T1-weighted (ceT1W) and T2-weighted (T2W) imaging segmentation tasks, with average DSCs of 0.827 and 0.820, respectively, which outperformed other the CNN models such as U-Net (DSC: 0.807 and 0.808), attention U-Net (DSC: 0.814 and 0.811), and V-Net (DSC: 0.805 and 0.807). For survival prediction, the proposed deep-learning model significantly outperformed traditional methods, yielding a concordance index of 0.732. Moreover, it effectively divided patients into low-risk and high-risk groups for disease progression (P<0.001).

Conclusions

Combining Transformer architecture with a CNN can improve MRI tumor segmentation, and this deep-learning model excelled in the survival prediction of patients with CC as compared to traditional methods.

Improving ultrasound diagnostic Precision for breast cancer and adenosis with modality-specific enhancement (MSE) - Breast Net

Adenosis is a benign breast condition whose lesions can mimic breast carcinoma and is evaluated for malignancy with the Breast Imaging-Reporting and Data System (BI-RADS). We construct and validate the performance of modality-specific enhancement (MSE)-Breast Net based on multimodal ultrasound images and compare it to the BI-RADS in differentiating adenosis from breast cancer. A total of 179 patients with breast carcinoma and 229 patients with adenosis were included in this retrospective, two-institution study, then divided into a training cohort (institution I, n = 292) and a validation cohort (institution II, n = 116). In the training cohort, the final model had a significantly greater AUC (0.82; P < 0.05) than B-mode-based model (0.69, 95% CI [0.49-0.90]). In the validation cohort, the AUC of the final model was 0.81, greater than that of the BI-RADS (0.75, P < 0.05). The multimodal model outperformed the individual and bimodal models, reaching a significantly greater AUC of 0.87 (95% CI = 0.69-1.0) (P < 0.05). MSE-Breast Net, based on multimodal ultrasound images, exhibited better diagnostic performance than the BI-RADS in differentiating adenosis from breast cancer and may contribute to clinical diagnosis and treatment.

Teacher-student guided knowledge distillation for unsupervised convolutional neural network-based speckle tracking in ultrasound strain elastography

Accurate and efficient motion estimation is a crucial component of real-time ultrasound elastography (USE). However, obtaining radiofrequency ultrasound (RF) data in clinical practice can be challenging. In contrast, although B-mode (BM) data is readily available, elastographic data derived from BM data results in sub-optimal elastographic images. Furthermore, existing conventional ultrasound devices (e.g., portable devices) cannot provide elastography modes, which has become a significant obstacle to the widespread use of traditional ultrasound devices. To address the challenges above, we developed a teacher-student guided knowledge distillation for an unsupervised convolutional neural network (TSGUPWC-Net) to improve the accuracy of BM motion estimation by employing a well-established convolutional neural network (CNN) named modified pyramid warping and cost volume network (MPWC-Net). A pre-trained teacher model based on RF is utilized to guide the training of a student model using BM data. Innovations outlined below include employing spatial attention transfer at intermediate layers to enhance the guidance effect of the model. The loss function consists of smoothness of the displacement field, knowledge distillation loss, and intermediate layer loss. We evaluated our method on simulated data, phantoms, and in vivo ultrasound data. The results indicate that our method has higher signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) values in axial strain estimation than the model trained on BM. The model is unsupervised and requires no ground truth labels during training, making it highly promising for motion estimation applications.

Repeatability and reproducibility of deep learning features for lung adenocarcinoma subtypes with nodules less than 10 mm in size: a multicenter thin-slice computed tomography phantom and clinical validation study

Background

Deep learning features (DLFs) derived from radiomics features (RFs) fused with deep learning have shown potential in enhancing diagnostic capability. However, the limited repeatability and reproducibility of DLFs across multiple centers represents a challenge in the clinically validation of these features. This study thus aimed to evaluate the repeatability and reproducibility of DLFs and their potential efficiency in differentiating subtypes of lung adenocarcinoma less than 10 mm in size and manifesting as ground-glass nodules (GGNs).

Methods

A chest phantom with nodules was scanned repeatedly using different thin-slice computed tomography (TSCT) scanners with varying acquisition and reconstruction parameters. The robustness of the DLFs was measured using the concordance correlation coefficient (CCC) and intraclass correlation coefficient (ICC). A deep learning approach was used for visualizing the DLFs. To assess the clinical effectiveness and generalizability of the stable and informative DLFs, three hospitals were used to source 275 patients, in whom 405 nodules were pathologically differentially diagnosed as GGN lung adenocarcinoma less than 10 mm in size and were retrospectively reviewed for clinical validation.

Results

A total of 64 DLFs were analyzed, which revealed that the variables of slice thickness and slice interval (ICC, 0.79±0.18) and reconstruction kernel (ICC, 0.82±0.07) were significantly associated with the robustness of DLFs. Feature visualization showed that the DLFs were mainly focused around the nodule areas. In the external validation, a subset of 28 robust DLFs identified as stable under all sources of variability achieved the highest area under curve [AUC =0.65, 95% confidence interval (CI): 0.53-0.76] compared to other DLF models and the radiomics model.

Conclusions

Although different manufacturers and scanning schemes affect the reproducibility of DLFs, certain DLFs demonstrated excellent stability and effectively improved diagnostic the efficacy for identifying subtypes of lung adenocarcinoma. Therefore, as the first step, screening stable DLFs in multicenter DLFs research may improve diagnostic efficacy and promote the application of these features.

Integrating intratumoral and peritumoral radiomics with deep transfer learning for DCE-MRI breast lesion differentiation: A multicenter study comparing performance with radiologists

PURPOSE

To conduct the fusion of radiomics and deep transfer learning features from the intratumoral and peritumoral areas in breast DCE-MRI images to differentiate between benign and malignant breast tumors, and to compare the diagnostic accuracy of this fusion model against the assessments made by experienced radiologists.

MATERIALS AND METHODS

This multi-center study conducted a retrospective analysis of DCE-MRI images from 330 women diagnosed with breast cancer, with 138 cases categorized as benign and 192 as malignant. The training and internal testing sets comprised 270 patients from center 1, while the external testing cohort consisted of 60 patients from center 2. A fusion feature set consisting of radiomics features and deep transfer learning features was constructed from both intratumoral (ITR) and peritumoral (PTR) areas. The Least absolute shrinkage and selection operator (LASSO) based support vector machine was chosen as the classifier by comparing its performance with five other machine learning models. The diagnostic performance and clinical usefulness of fusion model were verified and assessed through the area under the receiver operating characteristics (ROC) and decision curve analysis. Additionally, the performance of the fusion model was compared with the diagnostic assessments of two experienced radiologists to evaluate its relative accuracy. The study strictly adhered to CLEAR and METRICS guidelines for standardization to ensure rigorous and reproducible methods.

RESULTS

The findings show that the fusion model, utilizing radiomics and deep transfer learning features from the ITR and PTR, exhibited exceptional performance in classifying breast tumors, achieving AUCs of 0.950 in the internal testing set and 0.921 in the external testing set. This performance significantly surpasses that of models relying on singular regional radiomics or deep transfer learning features alone. Moreover, the fusion model demonstrated superior diagnostic accuracy compared to the evaluations conducted by two experienced radiologists, thereby highlighting its potential to support and enhance clinical decision-making in the differentiation of benign and malignant breast tumors.

CONCLUSION

The fusion model, combining multi-regional radiomics with deep transfer learning features, not only accurately differentiates between benign and malignant breast tumors but also outperforms the diagnostic assessments made by experienced radiologists. This underscores the model's potential as a valuable tool for improving the accuracy and reliability of breast tumor diagnosis.

Prediction of breast cancer and axillary positive-node response to neoadjuvant chemotherapy based on multi-parametric magnetic resonance imaging radiomics models

PURPOSE

Accurate identification of primary breast cancer and axillary positive-node response to neoadjuvant chemotherapy (NAC) is important for determining appropriate surgery strategies. We aimed to develop combining models based on breast multi-parametric magnetic resonance imaging and clinicopathologic characteristics for predicting therapeutic response of primary tumor and axillary positive-node prior to treatment.

MATERIALS AND METHODS

A total of 268 breast cancer patients who completed NAC and underwent surgery were enrolled. Radiomics features and clinicopathologic characteristics were analyzed through the analysis of variance and the least absolute shrinkage and selection operator algorithm. Finally, 24 and 28 optimal features were selected to construct machine learning models based on 6 algorithms for predicting each clinical outcome, respectively. The diagnostic performances of models were evaluated in the testing set by the area under the curve (AUC), sensitivity, specificity, and accuracy.

RESULTS

Of the 268 patients, 94 (35.1 %) achieved breast cancer pathological complete response (bpCR) and of the 240 patients with clinical positive-node, 120 (50.0 %) achieved axillary lymph node pathological complete response (apCR). The multi-layer perception (MLP) algorithm yielded the best diagnostic performances in predicting apCR with an AUC of 0.825 (95 % CI, 0.764-0.886) and an accuracy of 77.1 %. And MLP also outperformed other models in predicting bpCR with an AUC of 0.852 (95 % CI, 0.798-0.906) and an accuracy of 81.3 %.

CONCLUSIONS

Our study established non-invasive combining models to predict the therapeutic response of primary breast cancer and axillary positive-node prior to NAC, which may help to modify preoperative treatment and determine post-NAC surgery strategy.

Bimodal machine learning model for unstable hips in infants: integration of radiographic images with automatically-generated clinical measurements

Bimodal convolutional neural networks (CNNs) are frequently combined with patient information or several medical images to enhance the diagnostic performance. However, the technologies that integrate automatically generated clinical measurements within the images are scarce. Hence, we developed a bimodal model that produced automatic algorithm for clinical measurement (aaCM) from radiographic images and integrated the model with CNNs. In this multicenter research project, the diagnostic performance of the model was investigated with 813 radiographic hip images of infants at risk of developmental dysplasia of the hips (232 and 581 images of unstable and stable hips, respectively), with the ground truth defined by provocative examinations. The results indicated that the accuracy of aaCM was equal or higher than that of specialists, and the bimodal model showed better diagnostic performance than LightGBM, XGBoost, SVM, and single CNN models. aaCM can provide expert's knowledge in a high level, and our proposed bimodal model has better performance than the state-of-art models.

Anti-HER2 therapy response assessment for guiding treatment (de-)escalation in early HER2-positive breast cancer using a novel deep learning radiomics model

OBJECTIVES

Anti-HER2 targeted therapy significantly reduces risk of relapse in HER2 + breast cancer. New measures are needed for a precise risk stratification to guide (de-)escalation of anti-HER2 strategy.

METHODS

A total of 726 HER2 + cases who received no/single/dual anti-HER2 targeted therapies were split into three respective cohorts. A deep learning model (DeepTEPP) based on preoperative breast magnetic resonance (MR) was developed. Patients were scored and categorized into low-, moderate-, and high-risk groups. Recurrence-free survival (RFS) was compared in patients with different risk groups according to the anti-HER2 treatment they received, to validate the value of DeepTEPP in predicting treatment efficacy and guiding anti-HER2 strategy.

RESULTS

DeepTEPP was capable of risk stratification and guiding anti-HER2 treatment strategy: DeepTEPP-Low patients (60.5%) did not derive significant RFS benefit from trastuzumab (p = 0.144), proposing an anti-HER2 de-escalation. DeepTEPP-Moderate patients (19.8%) significantly benefited from trastuzumab (p = 0.048), but did not obtain additional improvements from pertuzumab (p = 0.125). DeepTEPP-High patients (19.7%) significantly benefited from dual HER2 blockade (p = 0.045), suggesting an anti-HER2 escalation.

CONCLUSIONS

DeepTEPP represents a pioneering MR-based deep learning model that enables the non-invasive prediction of adjuvant anti-HER2 effectiveness, thereby providing valuable guidance for anti-HER2 (de-)escalation strategies. DeepTEPP provides an important reference for choosing the appropriate individualized treatment in HER2 + breast cancer patients, warranting prospective validation.

CLINICAL RELEVANCE STATEMENT

We built an MR-based deep learning model DeepTEPP, which enables the non-invasive prediction of adjuvant anti-HER2 effectiveness, thus guiding anti-HER2 (de-)escalation strategies in early HER2-positive breast cancer patients.

KEY POINTS

• DeepTEPP is able to predict anti-HER2 effectiveness and to guide treatment (de-)escalation. • DeepTEPP demonstrated an impressive prognostic efficacy for recurrence-free survival and overall survival. • To our knowledge, this is one of the very few, also the largest study to test the efficacy of a deep learning model extracted from breast MR images on HER2-positive breast cancer survival and anti-HER2 therapy effectiveness prediction.

Endobronchial Ultrasound-Based Support Vector Machine Model for Differentiating between Benign and Malignant Mediastinal and Hilar Lymph Nodes

INTRODUCTION

The aim of the study was to establish an ultrasonographic radiomics machine learning model based on endobronchial ultrasound (EBUS) to assist in diagnosing benign and malignant mediastinal and hilar lymph nodes (LNs).

METHODS

The clinical and ultrasonographic image data of 197 patients were retrospectively analyzed. The radiomics features extracted by EBUS-based radiomics were analyzed by the least absolute shrinkage and selection operator. Then, we used a support vector machine (SVM) algorithm to establish an EBUS-based radiomics model. A total of 205 lesions were randomly divided into training (n = 143) and validation (n = 62) groups. The diagnostic efficiency was evaluated by receiver operating characteristic (ROC) curve analysis.

RESULTS

A total of 13 stable radiomics features with non-zero coefficients were selected. The SVM model exhibited promising performance in both groups. In the training group, the SVM model achieved an ROC area under the curve (AUC) of 0.892 (95% CI: 0.885-0.899), with an accuracy of 85.3%, sensitivity of 93.2%, and specificity of 79.8%. In the validation group, the SVM model had an ROC AUC of 0.906 (95% CI: 0.890-0.923), an accuracy of 74.2%, a sensitivity of 70.3%, and a specificity of 74.1%.

CONCLUSION

The EBUS-based radiomics model can be used to differentiate mediastinal and hilar benign and malignant LNs. The SVM model demonstrated excellent potential as a diagnostic tool in clinical practice.

A cutting-edge deep learning-and-radiomics-based ultrasound nomogram for precise prediction of axillary lymph node metastasis in breast cancer patients ≥ 75 years

Objective

The 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.

Methods

The 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.

Results

Histological 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.

Conclusion

The 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.

Predicting axillary lymph node metastasis in breast cancer patients: A radiomics-based multicenter approach with interpretability analysis

PURPOSE

To develop a MRI-based radiomics model, integrating the intratumoral and peritumoral imaging information to predict axillary lymph node metastasis (ALNM) in patients with breast cancer and to elucidate the model's decision-making process via interpretable algorithms.

METHODS

This study included 376 patients from three institutions who underwent contrast-enhanced breast MRI between 2021 and 2023. We used multiple machine learning algorithms to combine peritumoral, intratumoral, and radiological characteristics with the building of radiological, radiomics, and combined models. The model's performance was compared based on the area under the curve (AUC) obtained from the receiver operating characteristic analysis and interpretable machine learning techniques to analyze the operating mechanism of the model.

RESULTS

The radiomics model, incorporating features from both intratumoral tissue and the 3 mm peritumoral region and utilizing the backpropagation neural network (BPNN) algorithm, demonstrated superior diagnostic efficacy, achieving an AUC of 0.820. The AUC of the combination of the RAD score, clinical T stage, and spiculated margin was as high as 0.855. Furthermore, we conducted SHapley Additive exPlanations (SHAP) analysis to evaluate the contributions of RAD score, clinical T stage, and spiculated margin in ALNM status prediction.

CONCLUSIONS

The interpretable radiomics model we propose can better predict the ALNM status of breast cancer and help inform clinical treatment decisions.

Ultrasound-based radiomics for the differential diagnosis of breast masses: A systematic review and meta-analysis

OBJECTIVES

Ultrasound-based radiomics has demonstrated excellent diagnostic performance in differentiating benign and malignant breast masses. Given a few clinical studies on their diagnostic role, we conducted a meta-analysis of the potential effects of ultrasound-based radiomics for the differential diagnosis of breast masses, aiming to provide evidence-based medical basis for clinical research.

MATERIALS AND METHODS

We searched Embase, Web of Science, Cochrane Library, and PubMed databases from inception through to February 2023. The methodological quality assessment of the included studies was performed according to Quality Assessment of Diagnostic Accuracy Studies checklist. A diagnostic test accuracy systematic review and meta-analysis was performed in accordance with PRISMA guidelines. Sensitivity, specificity, and area under curve delineating benign and malignant lesions were recorded. We also used sensitivity analysis and subgroup analysis to explore potential sources of heterogeneity. Deeks' funnel plots was used to examine the publication bias.

RESULTS

A total of 11 studies were included in this meta-analysis. For the diagnosis of malignant breast masses worldwide, the overall mean rates of sensitivity and specificity of ultrasound-based radiomics were 0.90 (95% confidence interval [CI], 0.83-0.95) and 0.89 (95% CI, 0.82-0.94), respectively. The summary diagnostic odds ratio was 76 (95% CI, 26-219), and the area under the curve for the summary receiver operating characteristic curve was 0.95 (95% CI, 0.93-0.97).

CONCLUSION

Ultrasound-based radiomics has the potential to improve diagnostic accuracy to discriminate between benign and malignant breast masses, and could reduce unnecessary biopsies.

Multiparametric magnetic resonance imaging (MRI)-based radiomics model explained by the Shapley Additive exPlanations (SHAP) method for predicting complete response to neoadjuvant chemoradiotherapy in locally advanced rectal cancer: a multicenter retrospective study

Background

Predicting the response to neoadjuvant chemoradiotherapy (nCRT) before initiating treatment is essential for tailoring therapeutic strategies and monitoring prognosis in locally advanced rectal cancer (LARC). In this study, we aimed to develop and validate radiomic-based models to predict clinical and pathological complete responses (cCR and pCR, respectively) by incorporating the Shapley Additive exPlanations (SHAP) method for model interpretation.

Methods

A total of 285 patients with complete pretreatment clinical characteristics and T1-weighted (T1W) and T2-weighted (T2W) magnetic resonance imaging (MRI) at 3 centers were retrospectively recruited. The features of tumor lesions were extracted by PyRadiomics and selected using least absolute shrinkage and selection operator (LASSO) algorithm. The selected features were used to build multilayer perceptron (MLP) models alone or combined with clinical features. Area under the receiver operating characteristic curve (AUC), decision curve, and calibration curve were applied to evaluate performance of models. The SHAP method was adopted to explain the prediction models.

Results

The radiomic-based models all showed better performances than clinical models. The clinical-radiomic models showed the best differentiation on cCR and pCR with mean AUCs of 0.718 and 0.810 in the validation set, respectively. The decision curves of the clinical-radiomic models showed its values in clinical application. The SHAP method powerfully interpreted the prediction models both at a holistic and individual levels.

Conclusions

Our study highlights that the radiomic-based prediction models have more excellent abilities than clinical models and can effectively predict treatment response and optimize therapeutic strategies for patients with LARCs.

Brain metastasis magnetic resonance imaging-based deep learning for predicting epidermal growth factor receptor (EGFR) mutation and subtypes in metastatic non-small cell lung cancer

Background

The preoperative identification of epidermal growth factor receptor (EGFR) mutations and subtypes based on magnetic resonance imaging (MRI) of brain metastases (BM) is necessary to facilitate individualized therapy. This study aimed to develop a deep learning model to preoperatively detect EGFR mutations and identify the location of EGFR mutations in patients with non-small cell lung cancer (NSCLC) and BM.

Methods

We included 160 and 72 patients who underwent contrast-enhanced T1-weighted (T1w-CE) and T2-weighted (T2W) MRI at Liaoning Cancer Hospital and Institute (center 1) and Shengjing Hospital of China Medical University (center 2) to form a training cohort and an external validation cohort, respectively. A multiscale feature fusion network (MSF-Net) was developed by adaptively integrating features based on different stages of residual network (ResNet) 50 and by introducing channel and spatial attention modules. The external validation set from center 2 was used to assess the performance of MSF-Net and to compare it with that of handcrafted radiomics features. Receiver operating characteristic (ROC) curves, accuracy, precision, recall, and F1-score were used to evaluate the effectiveness of the models. Gradient-weighted class activation mapping (Grad-CAM) was used to demonstrate the attention of the MSF-Net model.

Results

The developed MSF-Net generated a better diagnostic performance than did the handcrafted radiomics in terms of the microaveraged area under the curve (AUC) (MSF-Net: 0.91; radiomics: 0.80) and macroaveraged AUC (MSF-Net: 0.90; radiomics: 0.81) for predicting EGFR mutations and subtypes.

Conclusions

This study provides an end-to-end and noninvasive imaging tool for the preoperative prediction of EGFR mutation status and subtypes based on BM, which may be helpful for facilitating individualized clinical treatment plans.

Dual-source dual-energy CT and deep learning for equivocal lymph nodes on CT images for thyroid cancer

OBJECTIVES

This study investigated the diagnostic performance of dual-energy computed tomography (CT) and deep learning for the preoperative classification of equivocal lymph nodes (LNs) on CT images in thyroid cancer patients.

METHODS

In this prospective study, from October 2020 to March 2021, 375 patients with thyroid disease underwent thin-section dual-energy thyroid CT at a small field of view (FOV) and thyroid surgery. The data of 183 patients with 281 LNs were analyzed. The targeted LNs were negative or equivocal on small FOV CT images. Six deep-learning models were used to classify the LNs on conventional CT images. The performance of all models was compared with pathology reports.

RESULTS

Of the 281 LNs, 65.5% had a short diameter of less than 4 mm. Multiple quantitative dual-energy CT parameters significantly differed between benign and malignant LNs. Multivariable logistic regression analyses showed that the best combination of parameters had an area under the curve (AUC) of 0.857, with excellent consistency and discrimination, and its diagnostic accuracy and sensitivity were 74.4% and 84.2%, respectively (p < 0.001). The visual geometry group 16 (VGG16) based model achieved the best accuracy (86%) and sensitivity (88%) in differentiating between benign and malignant LNs, with an AUC of 0.89.

CONCLUSIONS

The VGG16 model based on small FOV CT images showed better diagnostic accuracy and sensitivity than the spectral parameter model. Our study presents a noninvasive and convenient imaging biomarker to predict malignant LNs without suspicious CT features in thyroid cancer patients.

CLINICAL RELEVANCE STATEMENT

Our study presents a deep-learning-based model to predict malignant lymph nodes in thyroid cancer without suspicious features on conventional CT images, which shows better diagnostic accuracy and sensitivity than the regression model based on spectral parameters.

KEY POINTS

Many cervical lymph nodes (LNs) do not express suspicious features on conventional computed tomography (CT). Dual-energy CT parameters can distinguish between benign and malignant LNs. Visual geometry group 16 model shows superior diagnostic accuracy and sensitivity for malignant LNs.

Ultrasound-based radiomics nomogram for predicting axillary lymph node metastasis in invasive breast cancer

OBJECTIVE

To develop a nomogram for predicting axillary lymph node metastasis (ALNM) in patients with invasive breast cancer.

METHODS

We included 307 patients with clinicopathologically confirmed invasive breast cancer. The cohort was divided into a training group (n=215) and a validation group (n=92). Ultrasound images were used to extract radiomics features. The least absolute shrinkage and selection operator (LASSO) algorithm helped select pertinent features, from which Radiomics Scores (Radscores) were calculated using the LASSO regression equation. We developed three logistic regression models based on Radscores and 2D image features, and assessed the models' performance in the validation group. A nomogram was created from the best-performing model.

RESULTS

In the training set, the area under the curve (AUC) for the Radscore model, 2D feature model, and combined model were 0.76, 0.85, and 0.88, respectively. In the validation set, the AUCs were 0.71, 0.78, and 0.83, respectively. The combined model demonstrated good calibration and promising clinical utility.

CONCLUSION

Our ultrasound-based radiomics nomogram can accurately and non-invasively predict ALNM in breast cancer, suggesting potential clinical applications to optimize surgical and medical strategies.

Parallel CNN-Deep Learning Clinical-Imaging Signature for Assessing Pathologic Grade and Prognosis of Soft Tissue Sarcoma Patients

BACKGROUND

Traditional biopsies pose risks and may not accurately reflect soft tissue sarcoma (STS) heterogeneity. MRI provides a noninvasive, comprehensive alternative.

PURPOSE

To assess the diagnostic accuracy of histological grading and prognosis in STS patients when integrating clinical-imaging parameters with deep learning (DL) features from preoperative MR images.

STUDY TYPE

Retrospective/prospective.

POPULATION

354 pathologically confirmed STS patients (226 low-grade, 128 high-grade) from three hospitals and the Cancer Imaging Archive (TCIA), divided into training (n = 185), external test (n = 125), and TCIA cohorts (n = 44). 12 patients (6 low-grade, 6 high-grade) were enrolled into prospective validation cohort.

FIELD STRENGTH/SEQUENCE

1.5 T and 3.0 T/Unenhanced T1-weighted and fat-suppressed-T2-weighted.

ASSESSMENT

DL features were extracted from MR images using a parallel ResNet-18 model to construct DL signature. Clinical-imaging characteristics included age, gender, tumor-node-metastasis stage and MRI semantic features (depth, number, heterogeneity at T1WI/FS-T2WI, necrosis, and peritumoral edema). Logistic regression analysis identified significant risk factors for the clinical model. A DL clinical-imaging signature (DLCS) was constructed by incorporating DL signature with risk factors, evaluated for risk stratification, and assessed for progression-free survival (PFS) in retrospective cohorts, with an average follow-up of 23 ± 22 months.

STATISTICAL TESTS

Logistic regression, Cox regression, Kaplan-Meier curves, log-rank test, area under the receiver operating characteristic curve (AUC)，and decision curve analysis. A P-value <0.05 was considered significant.

RESULTS

The AUC values for DLCS in the external test, TCIA, and prospective test cohorts (0.834, 0.838, 0.819) were superior to clinical model (0.662, 0.685, 0.694). Decision curve analysis showed that the DLCS model provided greater clinical net benefit over the DL and clinical models. Also, the DLCS model was able to risk-stratify patients and assess PFS.

DATA CONCLUSION

The DLCS exhibited strong capabilities in histological grading and prognosis assessment for STS patients, and may have potential to aid in the formulation of personalized treatment plans.

LEVEL OF EVIDENCE: 4

TECHNICAL EFFICACY

Stage 2.

Prediction value study of breast cancer tumor infiltrating lymphocyte levels based on ultrasound imaging radiomics

Objective

Construct models based on grayscale ultrasound and radiomics and compare the efficacy of different models in preoperatively predicting the level of tumor-infiltrating lymphocytes in breast cancer.

Materials and methods

This study retrospectively collected clinical data and preoperative ultrasound images from 185 breast cancer patients confirmed by surgical pathology. Patients were randomly divided into a training set (n=111) and a testing set (n=74) using a 6:4 ratio. Based on a 10% threshold for tumor-infiltrating lymphocytes (TIL) levels, patients were classified into low-level and high-level groups. Radiomic features were extracted and selected using the training set. The evaluation included assessing the relationship between TIL levels and both radiomic features and grayscale ultrasound features. Subsequently, grayscale ultrasound models, radiomic models, and nomograms combining radiomics score (Rad-score) and grayscale ultrasound features were established. The predictive performance of different models was evaluated through receiver operating characteristic (ROC) analysis. Calibration curves assessed the fit of the nomograms, and decision curve analysis (DCA) evaluated the clinical effectiveness of the models.

Results

Univariate analyses and multivariate logistic regression analyses revealed that indistinct margin (P<0.001, Odds Ratio [OR]=0.214, 95% Confidence Interval [CI]: 0.103-1.026), posterior acoustic enhancement (P=0.027, OR=2.585, 95% CI: 1.116-5.987), and ipsilateral axillary lymph node enlargement (P=0.001, OR=4.214, 95% CI: 1.798-9.875) were independent predictive factors for high levels of TIL in breast cancer. In comparison to grayscale ultrasound model (Training set: Area under curve [AUC] 0.795; Testing set: AUC 0.720) and radiomics model (Training set: AUC 0.803; Testing set: AUC 0.759), the nomogram demonstrated superior discriminative ability on both the training (AUC 0.884) and testing (AUC 0.820) datasets. Calibration curves indicated high consistency between the nomogram model's predicted probability of breast cancer TIL levels and the actual occurrence probability. DCA revealed that the radiomics model and the nomogram model achieved higher clinical net benefits compared to the grayscale ultrasound model.

Conclusion

The nomogram based on preoperative ultrasound radiomics features exhibits robust predictive capacity for the non-invasive evaluation of breast cancer TIL levels, potentially providing a significant basis for individualized treatment decisions in breast cancer.

Integration of ultrasound-based radiomics with clinical features for predicting cervical lymph node metastasis in postoperative patients with differentiated thyroid carcinoma

OBJECTIVE

The primary objective was to establish a radiomics model utilizing longitudinal +cross-sectional ultrasound (US) images of lymph nodes (LNs) to predict cervical lymph node metastasis (CLNM) following differentiated thyroid carcinoma (DTC) surgery.

METHODS

A retrospective collection of 211 LNs from 211 postoperative DTC patients who underwent neck US with suspicious LN fine needle aspiration cytopathology findings at our institution was conducted between June 2021 and April 2023. Conventional US and clinicopathological information of patients were gathered. Based on the pathological results, patients were categorized into CLNM and non-CLNM groups. The database was randomly divided into a training cohort (n = 147) and a test cohort (n = 64) at a 7:3 ratio. The least absolute shrinkage and selection operator algorithm was applied to screen the most relevant radiomic features from the longitudinal + cross-sectional US images, and a radiomics model was constructed. Univariate and multivariate analyses were used to assess US and clinicopathological significance features. Subsequently, a combined model for predicting CLNM was constructed by integrating radiomics, conventional US, and clinicopathological features and presented as a nomogram.

RESULTS

The area under the curves (AUCs) of the longitudinal + cross-sectional radiomics models were 0.846 and 0.801 in the training and test sets, respectively, outperforming the single longitudinal and cross-sectional models (p < 0.05). In the testing cohort, the AUC of the combined model in predicting CLNM was 0.901, surpassing that of the single US model (AUC, 0.731) and radiomics model (AUC, 0.801).

CONCLUSIONS

The US-based radiomics model exhibits the potential to accurately predict CLNM following DTC surgery, thereby enhancing diagnostic accuracy.

Prediction of Parkinson's disease by transcranial sonography-based deep learning

OBJECTIVES

Transcranial sonography has been used as a valid neuroimaging tool to diagnose Parkinson's disease (PD). This study aimed to develop a modified transcranial sonography (TCS) technique based on a deep convolutional neural network (DCNN) model to predict Parkinson's disease.

METHODS

This retrospective diagnostic study was conducted using 1529 transcranial sonography images collected from 854 patients with PD and 775 normal controls admitted to the Second Affiliated Hospital of Soochow University (Suzhou, Jiangsu, China) between September 2019 and May 2022. The data set was divided into training cohorts (570 PD patients and 541 normal controls), and the validation set (184 PD patients and 234 normal controls). Using these datasets, we developed four different DCNN models (ResNet18, ResNet50, ResNet152, and DenseNet121). We then assessed their diagnostic performance, including the area under the receiver operating characteristic (AUROC) curve, specificity, sensitivity, positive predictive value (PPV), negative predictive value (NPV), and F1 score and compared with traditional diagnostic criteria.

RESULTS

Among the 1529 TCS images, 570 PD patients and 541 normal controls from 4 of 6 sonographers of the TCS team were selected as the training cohort, and 184 PD patients and 234 normal controls from the other 2 sonographers were chosen as the validation cohort. There were no sex and age differences between PD patients and normal control subjects in the training and validation cohorts (P values > 0.05). All DCNN models achieved good performance in distinguishing PD patients from normal control subjects on the validation datasets, with diagnostic AUROCs and accuracy of 0.949 (95% CI 0.925, 0.965) and 86.60 for the RestNet18 model, 0.949 (95% CI 0.929, 0.971) and 87.56 for ResNet50, 0.945 (95% CI 0.931, 0.969) and 88.04 for ResNet152, 0.953 (95% CI 0.935, 0.971) and 87.80 for DenseNet121, respectively. On the other hand, the diagnostic accuracy of the traditional diagnostic method was 82.30. The accuracy of all DCNN models was higher than that of traditional diagnostic method. Moreover, the 5k-fold cross-validation results in train datasets showed that these DCNN models are robust.

CONCLUSION

The developed transcranial sonography-based DCNN models performed better than traditional diagnostic criteria, thus improving the sonographer's accuracy in diagnosing PD.

Machine learning approaches in the prediction of positive axillary lymph nodes post neoadjuvant chemotherapy using MRI, CT, or ultrasound: A systematic review

Background and objective

Neoadjuvant chemotherapy is a standard treatment approach for locally advanced breast cancer. Conventional imaging modalities, such as magnetic resonance imaging (MRI), computed tomography (CT), and ultrasound, have been used for axillary lymph node evaluation which is crucial for treatment planning and prognostication. This systematic review aims to comprehensively examine the current research on applying machine learning algorithms for predicting positive axillary lymph nodes following neoadjuvant chemotherapy utilizing imaging modalities, including MRI, CT, and ultrasound.

Methods

A systematic search was conducted across databases, including PubMed, Scopus, and Web of Science, to identify relevant studies published up to December 2023. Articles employing machine learning algorithms to predict positive axillary lymph nodes using MRI, CT, or ultrasound data after neoadjuvant chemotherapy were included. The review follows the preferred reporting items for systematic reviews and meta-analyses (PRISMA) guidelines, encompassing data extraction and quality assessment.

Results

Seven studies were included, comprising 1502 patients. Four studies used MRI, two used CT, and one applied ultrasound. Two studies developed deep-learning models, while five used classic machine-learning models mainly based on multiple regression. Across the studies, the models showed high predictive accuracy, with the best-performing models combining radiomics and clinical data.

Conclusion

This systematic review demonstrated the potential of utilizing advanced data analysis techniques, such as deep learning radiomics, in improving the prediction of positive axillary lymph nodes in breast cancer patients following neoadjuvant chemotherapy.

Prediction of histopathologic grades of bladder cancer with radiomics based on MRI: Comparison with traditional MRI

PURPOSE

To compare biparametric magnetic resonance imaging (bp-MRI) radiomics signatures and traditional MRI model for the preoperative prediction of bladder cancer (BCa) grade.

MATERIALS AND METHODS

This retrospective study included 255 consecutive patients with pathologically confirmed 113 low-grade and 142 high-grade BCa. The traditional MRI nomogram model was developed using univariate and multivariate logistic regression by the mean apparent diffusion coefficient (ADC), vesical imaging reporting and data system, tumor size, and the number of tumors. Volumes of interest were manually drawn on T2-weighted imaging (T2WI) and ADC maps by 2 radiologists. Using one-way analysis of variance, correlation, and least absolute shrinkage and selection operator methods to select features. Then, a logistic regression classifier was used to develop the radiomics signatures. Receiver operating characteristic (ROC) analysis was used to compare the diagnostic abilities of the radiomics and traditional MRI models by the DeLong test. Finally, decision curve analysis was performed by estimating the clinical usefulness of the 2 models.

RESULTS

The area under the ROC curves (AUCs) of the traditional MRI model were 0.841 in the training cohort and 0.806 in the validation cohort. The AUCs of the 3 groups of radiomics model [ADC, T2WI, bp-MRI (ADC and T2WI)] were 0.888, 0.875, and 0.899 in the training cohort and 0.863, 0.805, and 0.867 in the validation cohort, respectively. The combined radiomics model achieved higher AUCs than the traditional MRI model. decision curve analysis indicated that the radiomics model had higher net benefits than the traditional MRI model.

CONCLUSION

The bp-MRI radiomics model may help distinguish high-grade and low-grade BCa and outperforming the traditional MRI model. Multicenter validation is needed to acquire high-level evidence for its clinical application.

Ultrasound-based radiomics-clinical nomogram for noninvasive prediction of residual cancer burden grading in breast cancer

PURPOSE

To assess the predictive value of an ultrasound-based radiomics-clinical nomogram for grading residual cancer burden (RCB) in breast cancer patients.

METHODS

This retrospective study of breast cancer patients who underwent neoadjuvant therapy (NAC) and ultrasound scanning between November 2020 and July 2023. First, a radiomics model was established based on ultrasound images. Subsequently, multivariate LR (logistic regression) analysis incorporating both radiomic scores and clinical factors was performed to construct a nomogram. Finally, Receiver operating characteristics (ROC) curve analysis and decision curve analysis (DCA) were employed to evaluate and validate the diagnostic accuracy and effectiveness of the nomogram.

RESULTS

A total of 1122 patients were included in this study. Among them, 427 patients exhibited a favorable response to NAC chemotherapy, while 695 patients demonstrated a poor response to NAC therapy. The radiomics model achieved an AUC value of 0.84 in the training cohort and 0.83 in the validation cohort. The ultrasound-based radiomics-clinical nomogram achieved an AUC value of 0.90 in the training cohort and 0.91 in the validation cohort.

CONCLUSIONS

Ultrasound-based radiomics-clinical nomogram can accurately predict the effectiveness of NAC therapy by predicting RCB grading in breast cancer patients.

A Review of Machine Learning Approaches for Brain Positron Emission Tomography Data Analysis

Positron emission tomography (PET) imaging has moved forward the development of medical diagnostics and research across various domains, including cardiology, neurology, infection detection, and oncology. The integration of machine learning (ML) algorithms into PET data analysis has further enhanced their capabilities of including disease diagnosis and classification, image segmentation, and quantitative analysis. ML algorithms empower researchers and clinicians to extract valuable insights from complex big PET datasets, which enabling automated pattern recognition, predictive health outcome modeling, and more efficient data analysis. This review explains the basic knowledge of PET imaging, statistical methods for PET image analysis, and challenges of PET data analysis. We also discussed the improvement of analysis capabilities by combining PET data with machine learning algorithms and the application of this combination in various aspects of PET image research. This review also highlights current trends and future directions in PET imaging, emphasizing the driving and critical role of machine learning and big PET image data analytics in improving diagnostic accuracy and personalized medical approaches. Integration between PET imaging will shape the future of medical diagnosis and research.

Deep learning for risk stratification of thymoma pathological subtypes based on preoperative CT images

OBJECTIVES

This study aims to develop an innovative, deep model for thymoma risk stratification using preoperative CT images. Current algorithms predominantly focus on radiomic features or 2D deep features and require manual tumor segmentation by radiologists, limiting their practical applicability.

METHODS

The deep model was trained and tested on a dataset comprising CT images from 147 patients (82 female; mean age, 54 years ± 10) who underwent surgical resection and received subsequent pathological confirmation. The eligible participants were divided into a training cohort (117 patients) and a testing cohort (30 patients) based on the CT scan time. The model consists of two stages: 3D tumor segmentation and risk stratification. The radiomic model and deep model (2D) were constructed for comparative analysis. Model performance was evaluated through dice coefficient, area under the curve (AUC), and accuracy.

RESULTS

In both the training and testing cohorts, the deep model demonstrated better performance in differentiating thymoma risk, boasting AUCs of 0.998 and 0.893 respectively. This was compared to the radiomic model (AUCs of 0.773 and 0.769) and deep model (2D) (AUCs of 0.981 and 0.760). Notably, the deep model was capable of simultaneously identifying lesions, segmenting the region of interest (ROI), and differentiating the risk of thymoma on arterial phase CT images. Its diagnostic prowess outperformed that of the baseline model.

CONCLUSIONS

The deep model has the potential to serve as an innovative decision-making tool, assisting on clinical prognosis evaluation and the discernment of suitable treatments for different thymoma pathological subtypes.

KEY POINTS

• This study incorporated both tumor segmentation and risk stratification. • The deep model, using clinical and 3D deep features, effectively predicted thymoma risk. • The deep model improved AUCs by 16.1pt and 17.5pt compared to radiomic model and deep model (2D) respectively.

A combinatorial MRI sequence-based radiomics model for preoperative prediction of microsatellite instability status in rectal cancer

This study aimed to develop an optimal radiomics model for preoperatively predicting microsatellite instability (MSI) in patients with rectal cancer (RC) based on multiparametric magnetic resonance imaging. The retrospective study included 308 RC patients who did not receive preoperative antitumor therapy, among whom 51 had MSI. Radiomics features were extracted and dimensionally reduced from T2-weighted imaging (T2WI), T1-weighted imaging (T1WI), diffusion-weighted imaging (DWI), and T1-weighted contrast enhanced (T1CE) images for each patient, and the features of each sequence were combined. Multifactor logistic regression was used to screen the optimal feature set for each combination. Different machine learning methods were applied to construct predictive MSI status models. Relative standard deviation values were determined to evaluate model performance and select the optimal model. Receiver operating characteristic (ROC) curve, calibration curve, and decision curve analyses were performed to evaluate model performance. The model constructed using the k-nearest neighbor (KNN) method combined with T2WI and T1CE images performed best. The area under the curve values for prediction of MSI with this model were 0.849 (0.804-0.887), with a sensitivity of 0.784 and specificity of 0.805. The Delong test showed no significant difference in diagnostic efficacy between the KNN-derived model and the traditional logistic regression model constructed using T1WI + DWI + T1CE and T2WI + T1WI + DWI + T1CE data (P > 0.05) and the diagnostic efficiency of the KNN-derived model was slightly better than that of the traditional model. From ROC curve analysis, the KNN-derived model significantly distinguished patients at low- and high-risk of MSI with the optimal threshold of 0.2, supporting the clinical applicability of the model. The model constructed using the KNN method can be applied to noninvasively predict MSI status in RC patients before surgery based on radiomics features from T2WI and T1CE images. Thus, this method may provide a convenient and practical tool for formulating treatment strategies and optimizing individual clinical decision-making for patients with RC.

Deep learning driven diagnosis of malignant soft tissue tumors based on dual-modal ultrasound images and clinical indexes

Background

Soft tissue tumors (STTs) are benign or malignant superficial neoplasms arising from soft tissues throughout the body with versatile pathological types. Although Ultrasonography (US) is one of the most common imaging tools to diagnose malignant STTs, it still has several drawbacks in STT diagnosis that need improving.

Objectives

The study aims to establish this deep learning (DL) driven Artificial intelligence (AI) system for predicting malignant STTs based on US images and clinical indexes of the patients.

Methods

We retrospectively enrolled 271 malignant and 462 benign masses to build the AI system using 5-fold validation. A prospective dataset of 44 malignant masses and 101 benign masses was used to validate the accuracy of system. A multi-data fusion convolutional neural network, named ultrasound clinical soft tissue tumor net (UC-STTNet), was developed to combine gray scale and color Doppler US images and clinic features for malignant STTs diagnosis. Six radiologists (R1-R6) with three experience levels were invited for reader study.

Results

The AI system achieved an area under receiver operating curve (AUC) value of 0.89 in the retrospective dataset. The diagnostic performance of the AI system was higher than that of one of the senior radiologists (AUC of AI vs R2: 0.89 vs. 0.84, p=0.022) and all of the intermediate and junior radiologists (AUC of AI vs R3, R4, R5, R6: 0.89 vs 0.75, 0.81, 0.80, 0.63; p <0.01). The AI system also achieved an AUC of 0.85 in the prospective dataset. With the assistance of the system, the diagnostic performances and inter-observer agreement of the radiologists was improved (AUC of R3, R5, R6: 0.75 to 0.83, 0.80 to 0.85, 0.63 to 0.69; p<0.01).

Conclusion

The AI system could be a useful tool in diagnosing malignant STTs, and could also help radiologists improve diagnostic performance.

Machine Learning Prediction of Lymph Node Metastasis in Breast Cancer: Performance of a Multi-institutional MRI-based 4D Convolutional Neural Network

Purpose To develop a custom deep convolutional neural network (CNN) for noninvasive prediction of breast cancer nodal metastasis. Materials and Methods This retrospective study included patients with newly diagnosed primary invasive breast cancer with known pathologic (pN) and clinical nodal (cN) status who underwent dynamic contrast-enhanced (DCE) breast MRI at the authors' institution between July 2013 and July 2016. Clinicopathologic data (age, estrogen receptor and human epidermal growth factor 2 status, Ki-67 index, and tumor grade) and cN and pN status were collected. A four-dimensional (4D) CNN model integrating temporal information from dynamic image sets was developed. The convolutional layers learned prognostic image features, which were combined with clinicopathologic measures to predict cN0 versus cN+ and pN0 versus pN+ disease. Performance was assessed with the area under the receiver operating characteristic curve (AUC), with fivefold nested cross-validation. Results Data from 350 female patients (mean age, 51.7 years ± 11.9 [SD]) were analyzed. AUC, sensitivity, and specificity values of the 4D hybrid model were 0.87 (95% CI: 0.83, 0.91), 89% (95% CI: 79%, 93%), and 76% (95% CI: 68%, 88%) for differentiating pN0 versus pN+ and 0.79 (95% CI: 0.76, 0.82), 80% (95% CI: 77%, 84%), and 62% (95% CI: 58%, 67%), respectively, for differentiating cN0 versus cN+. Conclusion The proposed deep learning model using tumor DCE MR images demonstrated high sensitivity in identifying breast cancer lymph node metastasis and shows promise for potential use as a clinical decision support tool. Keywords: MR Imaging, Breast, Breast Cancer, Breast MRI, Machine Learning, Metastasis, Prognostic Prediction Supplemental material is available for this article. Published under a CC BY 4.0 license.

Multitask Deep Learning-Based Whole-Process System for Automatic Diagnosis of Breast Lesions and Axillary Lymph Node Metastasis Discrimination from Dynamic Contrast-Enhanced-MRI: A Multicenter Study

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

4 TECHNICAL EFFICACY STAGE: 3.

Prediction of early hematoma expansion of spontaneous intracerebral hemorrhage based on deep learning radiomics features of noncontrast computed tomography

OBJECTIVES

Aimed to develop a nomogram model based on deep learning features and radiomics features for the prediction of early hematoma expansion.

METHODS

A total of 561 cases of spontaneous intracerebral hemorrhage (sICH) with baseline Noncontrast Computed Tomography (NCCT) were included. The metrics of hematoma detection were evaluated by Intersection over Union (IoU), Dice coefficient (Dice), and accuracy (ACC). The semantic features of sICH were judged by EfficientNet-B0 classification model. Radiomics analysis was performed based on the region of interest which was automatically segmented by deep learning. A combined model was constructed in order to predict the early expansion of hematoma using multivariate binary logistic regression, and a nomogram and calibration curve were drawn to verify its predictive efficacy by ROC analysis.

RESULTS

The accuracy of hematoma detection by segmentation model was 98.2% for IoU greater than 0.6 and 76.5% for IoU greater than 0.8 in the training cohort. In the validation cohort, the accuracy was 86.6% for IoU greater than 0.6 and 70.0% for IoU greater than 0.8. The AUCs of the deep learning model to judge semantic features were 0.95 to 0.99 in the training cohort, while in the validation cohort, the values were 0.71 to 0.83. The deep learning radiomics model showed a better performance with higher AUC in training cohort (0.87), internal validation cohort (0.83), and external validation cohort (0.82) than either semantic features or Radscore.

CONCLUSION

The combined model based on deep learning features and radiomics features has certain efficiency for judging the risk grade of hematoma.

CLINICAL RELEVANCE STATEMENT

Our study revealed that the deep learning model can significantly improve the work efficiency of segmentation and semantic feature classification of spontaneous intracerebral hemorrhage. The combined model has a good prediction efficiency for early hematoma expansion.

KEY POINTS

• We employ a deep learning algorithm to perform segmentation and semantic feature classification of spontaneous intracerebral hemorrhage and construct a prediction model for early hematoma expansion. • The deep learning radiomics model shows a favorable performance for the prediction of early hematoma expansion. • The combined model holds the potential to be used as a tool in judging the risk grade of hematoma.

Preoperative Discrimination of CDKN2A/B Homozygous Deletion Status in Isocitrate Dehydrogenase-Mutant Astrocytoma: A Deep Learning-Based Radiomics Model Using MRI

BACKGROUND

Cyclin-dependent kinase inhibitor 2A/B (CDKN2A/B) homozygous deletion has been verified as an independent and critical biomarker of negative prognosis and short survival in isocitrate dehydrogenase (IDH)-mutant astrocytoma. Therefore, noninvasive and accurate discrimination of CDKN2A/B homozygous deletion status is essential for the clinical management of IDH-mutant astrocytoma patients.

PURPOSE

To develop a noninvasive, robust preoperative model based on MR image features for discriminating CDKN2A/B homozygous deletion status of IDH-mutant astrocytoma.

STUDY TYPE

Retrospective.

POPULATION

Two hundred fifty-one patients: 107 patients with CDKN2A/B homozygous deletion and 144 patients without CDKN2A/B homozygous deletion.

FIELD STRENGTH/SEQUENCE

3.0 T/1.5 T: Contrast-enhanced T1-weighted spin-echo inversion recovery sequence (CE-T1WI) and T2-weighted fluid-attenuation spin-echo inversion recovery sequence (T2FLAIR).

ASSESSMENT

A total of 1106 radiomics and 1000 deep learning features extracted from CE-T1WI and T2FLAIR were used to develop models to discriminate the CDKN2A/B homozygous deletion status. Radiomics models, deep learning-based radiomics (DLR) models and the final integrated model combining radiomics features with deep learning features were developed and compared their preoperative discrimination performance.

STATISTICAL TESTING

Pearson chi-square test and Mann Whitney U test were used for assessing the statistical differences in patients' clinical characteristics. The Delong test compared the statistical differences of receiver operating characteristic (ROC) curves and area under the curve (AUC) of different models. The significance threshold is P < 0.05.

RESULTS

The final combined model (training AUC = 0.966; validation AUC = 0.935; test group: AUC = 0.943) outperformed the optimal models based on only radiomics or DLR features (training: AUC = 0.916 and 0.952; validation: AUC = 0.886 and 0.912; test group: AUC = 0.862 and 0.902).

DATA CONCLUSION

Whether based on a single sequence or a combination of two sequences, radiomics and DLR models have achieved promising performance in assessing CDKN2A/B homozygous deletion status. However, the final model combining both deep learning and radiomics features from CE-T1WI and T2FLAIR outperformed the optimal radiomics or DLR model.

EVIDENCE LEVEL

4 TECHNICAL EFFICACY: Stage 2.

The use of longitudinal CT-based radiomics and clinicopathological features predicts the pathological complete response of metastasized axillary lymph nodes in breast cancer

BACKGROUND

Accurate assessment of axillary status after neoadjuvant therapy for breast cancer patients with axillary lymph node metastasis is important for the selection of appropriate subsequent axillary treatment decisions. Our objectives were to accurately predict whether the breast cancer patients with axillary lymph node metastases could achieve axillary pathological complete response (pCR).

METHODS

We collected imaging data to extract longitudinal CT image features before and after neoadjuvant chemotherapy (NAC), analyzed the correlation between radiomics and clinicopathological features, and developed models to predict whether patients with axillary lymph node metastasis can achieve axillary pCR after NAC. The clinical utility of the models was determined via decision curve analysis (DCA). Subgroup analyses were also performed. Then, a nomogram was developed based on the model with the best predictive efficiency and clinical utility and was validated using the calibration plots.

RESULTS

A total of 549 breast cancer patients with metastasized axillary lymph nodes were enrolled in this study. 42 independent radiomics features were selected from LASSO regression to construct a logistic regression model with clinicopathological features (LR radiomics-clinical combined model). The AUC of the LR radiomics-clinical combined model prediction performance was 0.861 in the training set and 0.891 in the testing set. For the HR + /HER2 - , HER2 + , and Triple negative subtype, the LR radiomics-clinical combined model yields the best prediction AUCs of 0.756, 0.812, and 0.928 in training sets, and AUCs of 0.757, 0.777 and 0.838 in testing sets, respectively.

CONCLUSIONS

The combination of radiomics features and clinicopathological characteristics can effectively predict axillary pCR status in NAC breast cancer patients.