ABSTRACT
OBJECTIVES: To develop and validate an ultrasound elastography radiomics nomogram for preoperative evaluation of the axillary lymph node (ALN) burden in early-stage breast cancer. METHODS: Data of 303 patients from hospital #1 (training cohort) and 130 cases from hospital #2 (external validation cohort) between Jun 2016 and May 2019 were enrolled. Radiomics features were extracted from shear-wave elastography (SWE) and corresponding B-mode ultrasound (BMUS) images. The minimum redundancy maximum relevance and least absolute shrinkage and selection operator algorithms were used to select ALN status-related features. Proportional odds ordinal logistic regression was performed using the radiomics signature together with clinical data, and an ordinal nomogram was subsequently developed. We evaluated its performance using C-index and calibration. RESULTS: SWE signature, US-reported LN status, and molecular subtype were independent risk factors associated with ALN status. The nomogram based on these variables showed good discrimination in the training (overall C-index: 0.842; 95%CI, 0.773-0.879) and the validation set (overall C-index: 0.822; 95%CI, 0.765-0.838). For discriminating between disease-free axilla (N0) and any axillary metastasis (N + (≥ 1)), it achieved a C-index of 0.845 (95%CI, 0.777-0.914) for the training cohort and 0.817 (95%CI, 0.769-0.865) for the validation cohort. The tool could also discriminate between low (N + (1-2)) and heavy metastatic ALN burden (N + (≥ 3)), with a C-index of 0.827 (95%CI, 0.742-0.913) in the training cohort and 0.810 (95%CI, 0.755-0.864) in the validation cohort. CONCLUSION: The radiomics model shows favourable predictive ability for ALN staging in patients with early-stage breast cancer, which could provide incremental information for decision-making. KEY POINTS: ⢠Radiomics analysis helps radiologists to evaluate the axillary lymph node status of breast cancer with accuracy. ⢠This multicentre retrospective study showed that radiomics nomogram based on shear-wave elastography provides incremental information for risk stratification. ⢠Treatment can be given with more precision based on the model.
Subject(s)
Breast Neoplasms , Elasticity Imaging Techniques , Axilla/pathology , Breast Neoplasms/pathology , Female , Humans , Lymph Nodes/diagnostic imaging , Lymph Nodes/pathology , Nomograms , Retrospective StudiesABSTRACT
OBJECTIVES: To evaluate the prediction performance of deep convolutional neural network (DCNN) based on ultrasound (US) images for the assessment of breast cancer molecular subtypes. METHODS: A dataset of 4828 US images from 1275 patients with primary breast cancer were used as the training samples. DCNN models were constructed primarily to predict the four St. Gallen molecular subtypes and secondarily to identify luminal disease from non-luminal disease based on the ground truth from immunohistochemical of whole tumor surgical specimen. US images from two other institutions were retained as independent test sets to validate the system. The models' performance was analyzed using per-class accuracy, positive predictive value (PPV), and Matthews correlation coefficient (MCC). RESULTS: The model achieved good performance in identifying the four breast cancer molecular subtypes in the two test sets, with accuracy ranging from 80.07% (95% CI, 76.49-83.23%) to 97.02% (95% CI, 95.22-98.16%) and 87.94% (95% CI, 85.08-90.31%) to 98.83% (95% CI, 97.60-99.43) for the two test cohorts for each sub-category, respectively. In terms of 4-class weighted average MCC, the model achieved 0.59 for test cohort A and 0.79 for test cohort B. Specifically, the DCNN also yielded good diagnostic performance in discriminating luminal disease from non-luminal disease, with a PPV of 93.29% (95% CI, 90.63-95.23%) and 88.21% (95% CI, 85.12-90.73%) for the two test cohorts, respectively. CONCLUSION: Using pretreatment US images of the breast cancer, deep learning model enables the assessment of molecular subtypes with high diagnostic accuracy. TRIAL REGISTRATION: Clinical trial number: ChiCTR1900027676 KEY POINTS: ⢠Deep convolutional neural network (DCNN) helps clinicians assess tumor features with accuracy. ⢠Multicenter retrospective study shows that DCNN derived from pretreatment ultrasound imagine improves the prediction of breast cancer molecular subtypes. ⢠Management of patients becomes more precise based on the DCNN model.
Subject(s)
Breast Neoplasms , Deep Learning , Breast Neoplasms/diagnostic imaging , Humans , Neural Networks, Computer , Retrospective Studies , UltrasonographyABSTRACT
RATIONALE AND OBJECTIVES: Nottingham histological grade (NHG) 2 breast cancer has an intermediate risk of recurrence, which is not informative for therapeutic decision-making. We sought to develop and independently validate an MRI-based radiomics signature (Rad-Grade) to improve prognostic re-stratification of NHG 2 tumors. MATERIALS AND METHODS: Nine hundred-eight subjects with invasive breast cancer and preoperative MRI scans were retrospectively obtained. The NHG 1 and 3 tumors were randomly split into training and independent test cohort, with the NHG 2 as the prognostic validation set. From MRI image features, a radiomics-based signature predictive of the histological grade was built by use of the LASSO logistic regression algorithm. The model was developed for identifying NHG 1 and 3 radiological patterns, followed with re-stratification of NHG 2 tumors into Rad-Grade (RG)2-low (NHG 1-like) and RG2-high (NHG 3-like) subtypes using the learned patterns, and the prognostic value was assessed in terms of recurrence-free survival (RFS). RESULTS: The Rad-Grade showed independent prognostic value for re-stratification of NHG 2 tumors, where RG2-high had an increased risk for recurrence (HR 2.20, 1.10-4.40, p = 0.026) compared with RG2-low after adjusting for established risk factors. RG2-low shared similar phenotypic characteristics and RFS outcomes with NHG 1, and RG2-high with NHG 3, revealing that the model captures radiomic features in NHG 2 that are associated with different aggressiveness. The prognostic value of Rad-Grade was further validated in the NHG2 ER+ (HR 2.53, 1.13-5.56, p = 0.023) and NHG 2 ER+LN- (HR 5.72, 1.24-26.44, p = 0.025) subgroups, and in specific treatment contexts. CONCLUSION: The radiomics-based re-stratification of NHG 2 tumors offers a cost-effective promising alternative to gene expression profiling for tumor grading and thus may improve clinical decisions.
Subject(s)
Breast Neoplasms , Humans , Female , Breast Neoplasms/diagnostic imaging , Breast Neoplasms/pathology , Retrospective Studies , Magnetic Resonance Imaging/methods , Prognosis , Neoplasm GradingABSTRACT
PURPOSE: The aim of the study was to develop and validate a deep learning radiomic nomogram (DLRN) for preoperatively assessing breast cancer pathological complete response (pCR) after neoadjuvant chemotherapy (NAC) based on the pre- and post-treatment ultrasound. METHODS: Patients with locally advanced breast cancer (LABC) proved by biopsy who proceeded to undergo preoperative NAC were enrolled from hospital #1 (training cohort, 356 cases) and hospital #2 (independent external validation cohort, 236 cases). Deep learning and handcrafted radiomic features reflecting the phenotypes of the pre-treatment (radiomic signature [RS] 1) and post-treatment tumour (RS2) were extracted. The minimum redundancy maximum relevance algorithm and the least absolute shrinkage and selection operator regression were used for feature selection and RS construction. A DLRN was then developed based on the RSs and independent clinicopathological risk factors. The performance of the model was assessed with regard to calibration, discrimination and clinical usefulness. RESULTS: The DLRN predicted the pCR status with accuracy, yielded an area under the receiver operator characteristic curve of 0.94 (95% confidence interval, 0.91-0.97) in the validation cohort, with good calibration. The DLRN outperformed the clinical model and single RS within both cohorts (P < 0.05, as per the DeLong test) and performed better than two experts' prediction of pCR (both P < 0.01 for comparison of total accuracy). Besides, prediction within the hormone receptor-positive/human epidermal growth factor receptor 2 (HER2)-negative, HER2+ and triple-negative subgroups also achieved good discrimination performance, with an AUC of 0.90, 0.95 and 0.93, respectively, in the external validation cohort. Decision curve analysis confirmed that the model was clinically useful. CONCLUSION: A deep learning-based radiomic nomogram had good predictive value for pCR in LABC, which could provide valuable information for individual treatment.
Subject(s)
Breast Neoplasms/drug therapy , Deep Learning , Image Interpretation, Computer-Assisted , Neoadjuvant Therapy , Nomograms , Ultrasonography, Mammary , Adult , Breast Neoplasms/diagnostic imaging , Breast Neoplasms/pathology , Chemotherapy, Adjuvant , Clinical Decision-Making , Female , Humans , Mastectomy , Middle Aged , Neoadjuvant Therapy/adverse effects , Neoplasm Invasiveness , Observer Variation , Predictive Value of Tests , Reproducibility of Results , Retrospective Studies , Treatment OutcomeABSTRACT
PURPOSE: To develop a nomogram incorporating B-mode ultrasound (BMUS) and shear-wave elastography (SWE) radiomics to predict malignant status of breast lesions seen on US non-invasively. METHODS: Data on 278 consecutive patients from Hospital #1 (training cohort) and 123 cases from Hospital #2 (external validation cohort) referred for breast US with subsequent histopathologic analysis between May 2017 and October 2019 were retrospectively collected. Using their BMUS and SWE images, we built a radiomics nomogram to improve radiology workflow for management of breast lesions. The performance of the algorithm was compared with a consensus of three ACR BI-RADS committee experts and four individual radiologists, all of whom interpreted breast US images in clinical practice. RESULTS: Twelve features from BMUS and three from SWE were selected finally to construct the respective radiomic signature. The nomogram based on the dual-modal US radiomics achieved good diagnostic performance in the training (AUC 0.96; 95% confidence intervals [CI], 0.94-0.98) and the validation set (AUC 0.92; 95% CI, 0.87-0.97). For the 123 test lesions, the algorithm achieved 105 of 123 (85%) accuracy, comparable to the expert consensus (104 of 123 [85%], Pâ¯=⯠0.86) and four individual radiologists (93, 99, 95 and 97 of 123, with P value of 0.05, 0.31, 0.10 and 0.18 respectively). Furthermore, the model also performed well in the BI-RADS 4 and 5 categories. CONCLUSIONS: Performance of a dual-model US radiomics nomogram based on SWE for breast lesion classification may comparable to that of expert radiologists who used ACR BI-RADS guideline.