Characterization of tumor and adjacent peritumoral stroma in patients with breast cancer using high-resolution diffusion-weighted imaging: Correlation with pathologic biomarkers

Role of inter- and extra-lesion tissue, transfer learning, and fine-tuning in the robust classification of breast lesions

Accurate and unbiased classification of breast lesions is pivotal for early diagnosis and treatment, and a deep learning approach can effectively represent and utilize the digital content of images for more precise medical image analysis. Breast ultrasound imaging is useful for detecting and distinguishing benign masses from malignant masses. Based on the different ways in which benign and malignant tumors affect neighboring tissues, i.e., the pattern of growth and border irregularities, the penetration degree of the adjacent tissue, and tissue-level changes, we investigated the relationship between breast cancer imaging features and the roles of inter- and extra-lesional tissues and their impact on refining the performance of deep learning classification. The novelty of the proposed approach lies in considering the features extracted from the tissue inside the tumor (by performing an erosion operation) and from the lesion and surrounding tissue (by performing a dilation operation) for classification. This study uses these new features and three pre-trained deep neuronal networks to address the challenge of breast lesion classification in ultrasound images. To improve the classification accuracy and interpretability of the model, the proposed model leverages transfer learning to accelerate the training process. Three modern pre-trained CNN architectures (MobileNetV2, VGG16, and EfficientNetB7) are used for transfer learning and fine-tuning for optimization. There are concerns related to the neuronal networks producing erroneous outputs in the presence of noisy images, variations in input data, or adversarial attacks; thus, the proposed system uses the BUS-BRA database (two classes/benign and malignant) for training and testing and the unseen BUSI database (two classes/benign and malignant) for testing. Extensive experiments have recorded accuracy and AUC as performance parameters. The results indicate that the proposed system outperforms the existing breast cancer detection algorithms reported in the literature. AUC values of 1.00 are calculated for VGG16 and EfficientNet-B7 in the dilation cases. The proposed approach will facilitate this challenging and time-consuming classification task.

Association of peritumoral region features assessed on breast MRI and prognosis of breast cancer: a systematic review and meta-analysis

BACKGROUND

Increasing attention has been given to the peritumoral region. However, conflicting findings have been reported regarding the relationship between peritumoral region features on MRI and the prognosis of breast cancer.

PURPOSE

To evaluate the relationship between peritumoral region features on MRI and prognosis of breast cancer.

MATERIALS AND METHODS

A retrospective meta-analysis of observational studies comparing either qualitative or quantitative assessments of peritumoral MRI features on breast cancer with poor prognosis and control subjects was performed for studies published till October 2022. Pooled odds ratios (ORs) or standardized mean differences and 95% confidence intervals (CIs) were estimated by using random-effects models. The heterogeneity across the studies was measured using the statistic I2. Sensitivity analyses were conducted to test this association according to different study characteristics.

RESULTS

Twenty-four studies comprising 1853 breast cancers of poor prognosis and 2590 control participants were included in the analysis. Peritumoral edema was associated with non-luminal breast cancers (OR=3.56; 95%CI: 2.17, 5.83; p=.000), high expression of the Ki-67 index (OR=3.70; 95%CI: 2.41, 5.70; p =.000), high histological grade (OR=5.85; 95%CI: 3.89, 8.80; p=.000), lymph node metastasis (OR=2.83; 95%CI: 1.71, 4.67; p=.000), negative expression of HR (OR=3.15; 95%CI: 2.03, 4.88; p=.000), and lymphovascular invasion (OR=1.72; 95%CI: 1.28, 2.30; p=.000). The adjacent vessel sign was associated with greater odds of breast cancer with poor prognosis (OR=2.02; 95%CI: 1.68, 2.44; p=.000). Additionally, breast cancers with poor prognosis had higher peritumor-tumor ADC ratio (SMD=0.67; 95%CI: 0.54, 0.79; p=.000) and peritumoral ADCmean (SMD=0.29; 95%CI: 0.15, 0.42; p=.000). A peritumoral region of 2-20 mm away from the margin of the tumor is recommended.

CONCLUSION

The presence of peritumoral edema and adjacent vessel signs, higher peritumor-tumor ADC ratio, and peritumoral ADCmean were significantly correlated with poor prognosis of breast cancer.

CLINICAL RELEVANCE STATEMENT

MRI features of the peritumoral region can be used as a non-invasive index for the prognostic evaluation of invasive breast cancer.

KEY POINTS

• Peritumoral edema was positively associated with non-luminal breast cancer, high expression of the Ki-67 index, high histological grade, lymph node metastasis, negative expression of HR, and lymphovascular invasion. • The adjacent vessel sign was associated with greater odds of breast cancers with poor prognosis. • Breast cancers with poor prognosis had higher peritumor-tumor ADC ratio and peritumoral ADCmean.

Preliminary study on DCE-MRI radiomics analysis for differentiation of HER2-low and HER2-zero breast cancer

Purpose

This study aims to evaluate the utility of radiomic features from dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) in distinguishing HER2-low from HER2-zero breast cancer.

Patients and methods

We retrospectively analyzed 118 MRI cases, including 78 HER2-low and 40 HER2-zero patients confirmed by immunohistochemistry or fluorescence in situ hybridization. From each DCE-MRI case, 960 radiomic features were extracted. These features were screened and reduced using intraclass correlation coefficient, Mann-Whitney U test, and least absolute shrinkage to establish rad-scores. Logistic regression (LR) assessed the model's effectiveness in distinguishing HER2-low from HER2-zero. A clinicopathological MRI characteristic model was constructed using univariate and multivariate analysis, and a nomogram was developed combining rad-scores with significant MRI characteristics. Model performance was evaluated using the receiver operating characteristic (ROC) curve, and clinical benefit was assessed with decision curve analysis.

Results

The radiomics model, clinical model, and nomogram successfully distinguished between HER2-low and HER2-zero. The radiomics model showed excellent performance, with area under the curve (AUC) values of 0.875 in the training set and 0.845 in the test set, outperforming the clinical model (AUC = 0.691 and 0.672, respectively). HER2 status correlated with increased rad-score and Time Intensity Curve (TIC). The nomogram outperformed both models, with AUC, sensitivity, and specificity values of 0.892, 79.6%, and 82.8% in the training set, and 0.886, 83.3%, and 90.9% in the test set.

Conclusions

The DCE-MRI-based nomogram shows promising potential in differentiating HER2-low from HER2-zero status in breast cancer patients.

Nomogram for preoperative differentiation of benign and malignant breast tumors using contrast-enhanced cone-beam breast CT (CE CB-BCT) quantitative imaging and assessment features

PURPOSE

Breast cancer's impact necessitates refined diagnostic approaches. This study develops a nomogram using radiology quantitative features from contrast-enhanced cone-beam breast CT for accurate preoperative classification of benign and malignant breast tumors.

MATERIAL AND METHODS

A retrospective study enrolled 234 females with breast tumors, split into training and test sets. Contrast-enhanced cone-beam breast CT-images were acquired using Koning Breast CT-1000. Quantitative assessment features were extracted via 3D-slicer software, identifying independent predictors. The nomogram was constructed to preoperative differentiation benign and malignant breast tumors. Calibration curve was used to assess whether the model showed favorable correspondence with pathological confirmation. Decision curve analysis confirmed the model's superiority.

RESULTS

The study enrolled 234 female patients with a mean age of 50.2 years (SD ± 9.2). The training set had 164 patients (89 benign, 75 malignant), and the test set had 70 patients (29 benign, 41 malignant). The nomogram achieved excellent predictive performance in distinguishing benign and malignant breast lesions with an AUC of 0.940 (95% CI 0.900-0.940) in the training set and 0.970 (95% CI 0.940-0.970) in the test set.

CONCLUSION

This study illustrates the effectiveness of quantitative radiology features derived from contrast-enhanced cone-beam breast CT in distinguishing between benign and malignant breast tumors. Incorporating these features into a nomogram-based diagnostic model allows for breast tumor diagnoses that are objective and possess good accuracy. The application of these insights could substantially increase reliability and efficacy in the management of breast tumors, offering enhanced diagnostic capability.

Diagnosis of benign and malignant nodules with a radiomics model integrating features from nodules and mammary regions on DCE-MRI

Objectives

To establish a radiomics model for distinguishing between the benign and malignant mammary gland nodules via combining the features from nodule and mammary regions on DCE-MRI.

Methods

In this retrospective study, a total of 103 cases with mammary gland nodules (malignant/benign = 80/23) underwent DCE-MRI, and was confirmed by biopsy pathology. Features were extracted from both nodule region and mammary region on DCE-MRI. Three SVM classifiers were built for diagnosis of benign and malignant nodules as follows: the model with the features only from nodule region (N model), with the features only from mammary region (M model) and the model combining the features from nodule region and mammary region (NM model). The performance of models was evaluated with the area under the curve of receiver operating characteristic (AUC).

Results

One radiomic features is selected from nodule region and 3 radiomic features is selected from mammary region. Compared with N or M model, NM model exhibited the best performance with an AUC of 0.756.

Conclusions

Compared with the model only using the features from nodule or mammary region, the radiomics-based model combining the features from nodule and mammary region outperformed in the diagnosis of benign and malignant nodules.

Diffusion Tensor Imaging for Characterizing Changes in Triple-Negative Breast Cancer During Neoadjuvant Systemic Therapy

BACKGROUND

Assessment of treatment response in triple-negative breast cancer (TNBC) may guide individualized care for improved patient outcomes. Diffusion tensor imaging (DTI) measures tissue anisotropy and could be useful for characterizing changes in the tumors and adjacent fibroglandular tissue (FGT) of TNBC patients undergoing neoadjuvant systemic treatment (NAST).

PURPOSE

To evaluate the potential of DTI parameters for prediction of treatment response in TNBC patients undergoing NAST.

STUDY TYPE

Prospective.

POPULATION

Eighty-six women (average age: 51 ± 11 years) with biopsy-proven clinical stage I-III TNBC who underwent NAST followed by definitive surgery. 47% of patients (40/86) had pathologic complete response (pCR).

FIELD STRENGTH/SEQUENCE

3.0 T/reduced field of view single-shot echo-planar DTI sequence.

ASSESSMENT

Three MRI scans were acquired longitudinally (pre-treatment, after 2 cycles of NAST, and after 4 cycles of NAST). Eleven histogram features were extracted from DTI parameter maps of tumors, a peritumoral region (PTR), and FGT in the ipsilateral breast. DTI parameters included apparent diffusion coefficients and relative diffusion anisotropies. pCR status was determined at surgery.

STATISTICAL TESTS

Longitudinal changes of DTI features were tested for discrimination of pCR using Mann-Whitney U test and area under the receiver operating characteristic curve (AUC). A P value <0.05 was considered statistically significant.

RESULTS

47% of patients (40/86) had pCR. DTI parameters assessed after 2 and 4 cycles of NAST were significantly different between pCR and non-pCR patients when compared between tumors, PTRs, and FGTs. The median surface/average anisotropy of the PTR, measured after 2 and 4 cycles of NAST, increased in pCR patients and decreased in non-pCR patients (AUC: 0.78; 0.027 ± 0.043 vs. -0.017 ± 0.042 mm2 /s).

DATA CONCLUSION

Quantitative DTI features from breast tumors and the peritumoral tissue may be useful for predicting the response to NAST in TNBC.

EVIDENCE LEVEL

1 TECHNICAL EFFICACY: Stage 4.

Use of a deep learning algorithm for non-mass enhancement on breast MRI: comparison with radiologists' interpretations at various levels

PURPOSE

To evaluate the diagnostic performance of deep learning using the Residual Networks 50 (ResNet50) neural network constructed from different segmentations for distinguishing malignant and benign non-mass enhancement (NME) on breast magnetic resonance imaging (MRI) and conduct a comparison with radiologists with various levels of experience.

MATERIALS AND METHODS

A total of 84 consecutive patients with 86 lesions (51 malignant, 35 benign) presenting NME on breast MRI were analyzed. Three radiologists with different levels of experience evaluated all examinations, based on the Breast Imaging-Reporting and Data System (BI-RADS) lexicon and categorization. For the deep learning method, one expert radiologist performed lesion annotation manually using the early phase of dynamic contrast-enhanced (DCE) MRI. Two segmentation methods were applied: a precise segmentation was carefully set to include only the enhancing area, and a rough segmentation covered the whole enhancing region, including the intervenient non-enhancing area. ResNet50 was implemented using the DCE MRI input. The diagnostic performance of the radiologists' readings and deep learning were then compared using receiver operating curve analysis.

RESULTS

The ResNet50 model from precise segmentation achieved diagnostic accuracy equivalent [area under the curve (AUC) = 0.91, 95% confidence interval (CI) 0.90, 0.93] to that of a highly experienced radiologist (AUC = 0.89, 95% CI 0.81, 0.96; p = 0.45). Even the model from rough segmentation showed diagnostic performance equivalent to a board-certified radiologist (AUC = 0.80, 95% CI 0.78, 0.82 vs. AUC = 0.79, 95% CI 0.70, 0.89, respectively). Both ResNet50 models from the precise and rough segmentation exceeded the diagnostic accuracy of a radiology resident (AUC = 0.64, 95% CI 0.52, 0.76).

CONCLUSION

These findings suggest that the deep learning model from ResNet50 has the potential to ensure accuracy in the diagnosis of NME on breast MRI.

Multimodal ultrasound features of breast cancers: correlation with molecular subtypes

OBJECTIVES

To investigate whether multimodal intratumour and peritumour ultrasound features correlate with specific breast cancer molecular subtypes.

METHODS

From March to November 2021, a total of 85 patients with histologically proven breast cancer underwent B-mode, real-time elastography (RTE), colour Doppler flow imaging (CDFI) and contrast-enhanced ultrasound (CEUS). The time intensity curve (TIC) of CEUS was obtained, and the peak and time to peak (TTP) were analysed. Chi-square and binary logistic regression were used to analyse the connection between multimodal ultrasound imaging features and breast cancer molecular subtype.

RESULTS

Among 85 breast cancers, the subtypes were as follows: luminal A (36 cases, 42.4%), luminal B (20 cases, 23.5%), human epidermal growth factor receptor-2 positive (HER2+) (16 cases, 18.8%), and triple negative breast cancer (TNBC) (13 cases, 15.3%). Binary logistic regression models showed that RTE (P < 0.001) and CDFI (P = 0.036) were associated with the luminal A cancer subtype (C-index: 0.741), RTE (P = 0.016) and the peak ratio between intratumour and corpus mamma (P = 0.036) were related to the luminal B cancer subtype (C-index: 0.788). The peak ratio between peritumour and intratumour (P = 0.039) was related to the HER2 + cancer subtype (C-index: 0.859), and CDFI (P = 0.002) was associated with the TNBC subtype (C-index: 0.847).

CONCLUSIONS

Multimodal ultrasound features could be powerful predictors of specific breast cancer molecular subtypes. The intra- and peritumour CEUS features play assignable roles in separating luminal B and HER2 + breast cancer subtypes.

Quantitative Apparent Diffusion Coefficients From Peritumoral Regions as Early Predictors of Response to Neoadjuvant Systemic Therapy in Triple-Negative Breast Cancer

BACKGROUND

Pathologic complete response (pCR) to neoadjuvant systemic therapy (NAST) in triple-negative breast cancer (TNBC) is a strong predictor of patient survival. Edema in the peritumoral region (PTR) has been reported to be a negative prognostic factor in TNBC.

PURPOSE

To determine whether quantitative apparent diffusion coefficient (ADC) features from PTRs on reduced field-of-view (rFOV) diffusion-weighted imaging (DWI) predict the response to NAST in TNBC.

STUDY TYPE

Prospective.

POPULATION/SUBJECTS

A total of 108 patients with biopsy-proven TNBC who underwent NAST and definitive surgery during 2015-2020.

FIELD STRENGTH/SEQUENCE

A 3.0 T/rFOV single-shot diffusion-weighted echo-planar imaging sequence (DWI).

ASSESSMENT

Three scans were acquired longitudinally (pretreatment, after two cycles of NAST, and after four cycles of NAST). For each scan, 11 ADC histogram features (minimum, maximum, mean, median, standard deviation, kurtosis, skewness and 10th, 25th, 75th, and 90th percentiles) were extracted from tumors and from PTRs of 5 mm, 10 mm, 15 mm, and 20 mm in thickness with inclusion and exclusion of fat-dominant pixels.

STATISTICAL TESTS

ADC features were tested for prediction of pCR, both individually using Mann-Whitney U test and area under the receiver operating characteristic curve (AUC), and in combination in multivariable models with k-fold cross-validation. A P value < 0.05 was considered statistically significant.

RESULTS

Fifty-one patients (47%) had pCR. Maximum ADC from PTR, measured after two and four cycles of NAST, was significantly higher in pCR patients (2.8 ± 0.69 vs 3.5 ± 0.94 mm2 /sec). The top-performing feature for prediction of pCR was the maximum ADC from the 5-mm fat-inclusive PTR after cycle 4 of NAST (AUC: 0.74; 95% confidence interval: 0.64, 0.84). Multivariable models of ADC features performed similarly for fat-inclusive and fat-exclusive PTRs, with AUCs ranging from 0.68 to 0.72 for the cycle 2 and cycle 4 scans.

DATA CONCLUSION

Quantitative ADC features from PTRs may serve as early predictors of the response to NAST in TNBC.

EVIDENCE LEVEL

1 TECHNICAL EFFICACY: Stage 4.

Evaluation of the peritumoral features using radiomics and deep learning technology in non-spiculated and noncalcified masses of the breast on mammography

OBJECTIVE

To assess the significance of peritumoral features based on deep learning in classifying non-spiculated and noncalcified masses (NSNCM) on mammography.

METHODS

We retrospectively screened the digital mammography data of 2254 patients who underwent surgery for breast lesions in Harbin Medical University Cancer Hospital from January to December 2018. Deep learning and radiomics models were constructed. The classification efficacy in ROI and patient levels of AUC, accuracy, sensitivity, and specificity were compared. Stratified analysis was conducted to analyze the influence of primary factors on the AUC of the deep learning model. The image filter and CAM were used to visualize the radiomics and depth features.

RESULTS

For 1298 included patients, 771 (59.4%) were benign, and 527 (40.6%) were malignant. The best model was the deep learning combined model (2 mm), in which the AUC was 0.884 (P < 0.05); especially the AUC of breast composition B reached 0.941. All the deep learning models were superior to the radiomics models (P < 0.05), and the class activation map (CAM) showed a high expression of signals around the tumor of the deep learning model. The deep learning model achieved higher AUC for large size, age >60 years, and breast composition type B (P < 0.05).

CONCLUSION

Combining the tumoral and peritumoral features resulted in better identification of malignant NSNCM on mammography, and the performance of the deep learning model exceeded the radiomics model. Age, tumor size, and the breast composition type are essential for diagnosis.

Evaluation of the efficacy of transcatheter arterial embolization combined with apatinib on rabbit VX2 liver tumors by intravoxel incoherent motion diffusion-weighted MR imaging

Background and purpose

It is crucial to evaluate the efficacy, recurrence, and metastasis of liver tumors after clinical treatment. This study aimed to investigate the value of Introvoxel Incoherent Motion (IVIM) imaging in the evaluation of rabbit VX2 liver tumors treated with Transcatheter Arterial Embolization (TAE) combined with apatinib.

Methods

Twenty rabbit VX2 liver tumor models were established and randomly divided into either the experimental group (n=15) or the control group (n=5). The experimental group was treated with TAE combined with oral apatinib after successful tumor inoculation, while no treatment was administered following inoculation in the control group. IVIM sequence scan was performed in the experimental group before treatment, at 7 and 14 days after treatment. All rabbits were sacrificed after the last scan of the experimental group. Marginal tissues from the tumors of both groups were excised for immunohistochemical analysis to observe and compare the expression of microvessel density (MVD). The alterations of IVIM-related parameters of tumor tissues in the experimental group, including Apparent Diffusion Coefficient (ADC), True Diffusion Coefficient (D), Pseudodiffusion Coefficient (D*), and Perfusion Fraction (f) were compared at different periods, and the correlation between these parameters and MVD was analyzed.

Results

After treatment, ADC and D values significantly increased, whereas D* and f values both decreased, with statistically significant differences.(P<0.05). The average tumor MVD of the experimental group after TAE combined with apatinib ((33.750 ± 6.743) bars/high power field (HPF)) was significantly lower than that in the control group ((64.200 ± 10.164) bars/HPF)). Moreover, D and f were positively correlated with tumor MVD in the experimental group (r=0.741 for D and r=0.668 for f, P<0.05). However, there was no significant correlation between ADC and D* values of the experimental group and tumor MVD (r=0.252 for ADC and r=0.198 for D*, P>0.05).

Conclusion

IVIM imaging can be employed to evaluate the efficacy of TAE combined with apatinib in rabbit VX2 liver tumors. Alterations in D and f values were closely related to the MVD of liver tumor tissues.

The potential of predictive and prognostic breast MRI (P2-bMRI)

Magnetic resonance imaging (MRI) is an important part of breast cancer diagnosis and multimodal workup. It provides unsurpassed soft tissue contrast to analyse the underlying pathophysiology, and it is adopted for a variety of clinical indications. Predictive and prognostic breast MRI (P2-bMRI) is an emerging application next to these indications. The general objective of P2-bMRI is to provide predictive and/or prognostic biomarkers in order to support personalisation of breast cancer treatment. We believe P2-bMRI has a great clinical potential, thanks to the in vivo examination of the whole tumour and of the surrounding tissue, establishing a link between pathophysiology and response to therapy (prediction) as well as patient outcome (prognostication). The tools used for P2-bMRI cover a wide spectrum: standard and advanced multiparametric pulse sequences; structured reporting criteria (for instance BI-RADS descriptors); artificial intelligence methods, including machine learning (with emphasis on radiomics data analysis); and deep learning that have shown compelling potential for this purpose. P2-bMRI reuses the imaging data of examinations performed in the current practice. Accordingly, P2-bMRI could optimise clinical workflow, enabling cost savings and ultimately improving personalisation of treatment. This review introduces the concept of P2-bMRI, focusing on the clinical application of P2-bMRI by using semantic criteria.

Swin Transformer Improves the IDH Mutation Status Prediction of Gliomas Free of MRI-Based Tumor Segmentation

Background: Deep learning (DL) could predict isocitrate dehydrogenase (IDH) mutation status from MRIs. Yet, previous work focused on CNNs with refined tumor segmentation. To bridge the gap, this study aimed to evaluate the feasibility of developing a Transformer-based network to predict the IDH mutation status free of refined tumor segmentation. Methods: A total of 493 glioma patients were recruited from two independent institutions for model development (TCIA; N = 259) and external test (AHXZ; N = 234). IDH mutation status was predicted directly from T2 images with a Swin Transformer and conventional ResNet. Furthermore, to investigate the necessity of refined tumor segmentation, seven strategies for the model input image were explored: (i) whole tumor slice; (ii-iii) tumor mask and/or not edema; (iv-vii) tumor bounding box of 0.8, 1.0, 1.2, 1.5 times. Performance comparison was made among the networks of different architectures along with different image input strategies, using area under the curve (AUC) and accuracy (ACC). Finally, to further boost the performance, a hybrid model was built by incorporating the images with clinical features. Results: With the seven proposed input strategies, seven Swin Transformer models and seven ResNet models were built, respectively. Based on the seven Swin Transformer models, an averaged AUC of 0.965 (internal test) and 0.842 (external test) were achieved, outperforming 0.922 and 0.805 resulting from the seven ResNet models, respectively. When a bounding box of 1.0 times was used, Swin Transformer (AUC = 0.868, ACC = 80.7%), achieved the best results against the one that used tumor segmentation (Tumor + Edema, AUC = 0.862, ACC = 78.5%). The hybrid model that integrated age and location features into images yielded improved performance (AUC = 0.878, Accuracy = 82.0%) over the model that used images only. Conclusions: Swin Transformer outperforms the CNN-based ResNet in IDH prediction. Using bounding box input images benefits the DL networks in IDH prediction and makes the IDH prediction free of refined glioma segmentation feasible.

The value of noncontrast MRI in evaluating breast imaging reporting and data system category 0 lesions on digital mammograms

Background

Benign and malignant diagnosis of nonpalpable breast imaging reporting and data system (BI-RADS) category 0 lesions on digital mammograms (DMs) is very important. We compared the diagnostic performance of non-contrast-enhanced magnetic resonance imaging (MRI) and dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) for them. We sought to evaluate BI-RADS category 0 lesions using 3 MRI sequences: short tau inversion recovery (STIR), STIR combined with high b value diffusion-weighted imaging (STIR-DWI), and DCE-MRI.

Methods

We retrospectively reviewed 114 breast DMs rated as nonpalpable BI-RADS category 0 lesions in 112 patients from January 2014 to June 2019. STIR, high b value DWI, and DCE-MRI were performed for all patients. Two breast radiologists read individual sequences (STIR, DWI, DCE-MRI) and pairs of sequences (STIR-DWI) to detect BI-RADS category 0 lesions in DMs. Receiver operating characteristic (ROC) curve analysis was used to assess diagnostic performance according to a best valuable comparator that combined MRI imaging, clinical, and pathological data.

Results

Among of 114 lesions (the median age of patients was 47 years; the median size of the lesion was 19 mm), 32 (48.5%) malignant lesions were missed by STIR, 9 (13.6%) malignant lesions were missed by STIR-DWI, and 3 (4.5%) malignant lesions were missed by DCE-MRI. The principal finding of our study was that STIR-DWI and DCE-MRI showed higher diagnostic accuracy than did STIR (P<0.01). STIR-DWI showed higher accuracy [area under the curve (AUC) =0.858; sensitivity =87.8%] for BI-RADS category 0 lesions in DMs than did STIR (AUC =0.754; sensitivity =51.5%), while the performance was comparable to that of DCE-MRI (AUC =0.884; sensitivity =95.5%).

Conclusions

Using pairs of sequences (STIR-DWI) is a non-contrast-enhanced MRI technique and had an equal diagnostic performance in distinguishing benign from malignant lesions among nonpalpable BI-RADS category 0 lesions to that of DCE-MRI. As a result, STIR-DWI as having the potential to improve the safety and efficacy in of breast cancer screening, especially in nonpalpable BI-RADS category 0 lesions at in DMs.

Apparent Diffusion Coefficient-Based Convolutional Neural Network Model Can Be Better Than Sole Diffusion-Weighted Magnetic Resonance Imaging to Improve the Differentiation of Invasive Breast Cancer From Breast Ductal Carcinoma In Situ

BACKGROUND AND PURPOSE

Breast ductal carcinoma in situ (DCIS) has no metastatic potential, and has better clinical outcomes compared with invasive breast cancer (IBC). Convolutional neural networks (CNNs) can adaptively extract features and may achieve higher efficiency in apparent diffusion coefficient (ADC)-based tumor invasion assessment. This study aimed to determine the feasibility of constructing an ADC-based CNN model to discriminate DCIS from IBC.

METHODS

The study retrospectively enrolled 700 patients with primary breast cancer between March 2006 and June 2019 from our hospital, and randomly selected 560 patients as the training and validation sets (ratio of 3 to 1), and 140 patients as the internal test set. An independent external test set of 102 patients during July 2019 and May 2021 from a different scanner of our hospital was selected as the primary cohort using the same criteria. In each set, the status of tumor invasion was confirmed by pathologic examination. The CNN model was constructed to discriminate DCIS from IBC using the training and validation sets. The CNN model was evaluated using the internal and external tests, and compared with the discriminating performance using the mean ADC. The area under the curve (AUC), sensitivity, specificity, and accuracy were calculated to evaluate the performance of the previous model.

RESULTS

The AUCs of the ADC-based CNN model using the internal and external test sets were larger than those of the mean ADC (AUC: 0.977 vs. 0.866, P = 0.001; and 0.926 vs. 0.845, P = 0.096, respectively). Regarding the internal test set and external test set, the ADC-based CNN model yielded sensitivities of 0.893 and 0.873, specificities of 0.929 and 0.894, and accuracies of 0.907 and 0.902, respectively. Regarding the two test sets, the mean ADC showed sensitivities of 0.845 and 0.818, specificities of 0.821 and 0.829, and accuracies of 0.836 and 0.824, respectively. Using the ADC-based CNN model, the prediction only takes approximately one second for a single lesion.

CONCLUSION

The ADC-based CNN model can improve the differentiation of IBC from DCIS with higher accuracy and less time.

Classification of MR-Detected Additional Lesions in Patients With Breast Cancer Using a Combination of Radiomics Analysis and Machine Learning

Objective

This study was conducted in order to investigate the feasibility of using radiomics analysis (RA) with machine learning algorithms based on breast magnetic resonance (MR) images for discriminating malignant from benign MR-detected additional lesions in patients with primary breast cancer.

Materials and Methods

One hundred seventy-four MR-detected additional lesions (benign, n = 86; malignancy, n = 88) from 158 patients with ipsilateral primary breast cancer from a tertiary medical center were included in this retrospective study. The entire data were randomly split to training (80%) and independent test sets (20%). In addition, 25 patients (benign, n = 21; malignancy, n = 15) from another tertiary medical center were included for the external test. Radiomics features that were extracted from three regions-of-interest (ROIs; intratumor, peritumor, combined) using fat-saturated T1-weighted images obtained by subtracting pre- from postcontrast images (SUB) and T2-weighted image (T2) were utilized to train the support vector machine for the binary classification. A decision tree method was utilized to build a classifier model using clinical imaging interpretation (CII) features assessed by radiologists. Area under the receiver operating characteristic curve (AUROC), accuracy, sensitivity, and specificity were used to compare the diagnostic performance.

Results

The RA models trained using radiomics features from the intratumor-ROI showed comparable performance to the CII model (accuracy, AUROC: 73.3%, 69.6% for the SUB RA model; 70.0%, 75.1% for the T2 RA model; 73.3%, 72.0% for the CII model). The diagnostic performance increased when the radiomics and CII features were combined to build a fusion model. The fusion model that combines the CII features and radiomics features from multiparametric MRI data demonstrated the highest performance with an accuracy of 86.7% and an AUROC of 91.1%. The external test showed a similar pattern where the fusion models demonstrated higher levels of performance compared with the RA- or CII-only models. The accuracy and AUROC of the SUB+T2 RA+CII model in the external test were 80.6% and 91.4%, respectively.

Conclusion

Our study demonstrated the feasibility of using RA with machine learning approach based on multiparametric MRI for quantitatively characterizing MR-detected additional lesions. The fusion model demonstrated an improved diagnostic performance over the models trained with either RA or CII alone.

Quantitative Three-Dimensional Assessment of the Pharmacokinetic Parameters of Intra- and Peri-tumoural Tissues on Breast Dynamic Contrast-Enhanced Magnetic Resonance Imaging

We aimed to assess the feasibility of three-dimensional (3D) segmentation and to investigate whether semi-quantitative dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) parameters are associated with traditional prognostic factors for breast cancer. In addition, we evaluated whether both intra-tumoural and peri-tumoural DCE parameters can differentiate the breast cancers that are more aggressive from those that are less aggressive. Consecutive patients with newly diagnosed invasive breast cancer and structural breast MRI (3.0 T) were included after informed consent. Fifty-six patients (mean age, 57 years) with mass lesions of > 7 mm in diameter were included. A semi-automatic image post-processing algorithm was developed to measure 3D pharmacokinetic information from the DCE-MRI images. The kinetic parameters were extracted from time-signal curves, and the absolute tissue contrast agent concentrations were calculated with a reference tissue model. Markedly, higher intra-tumoural and peri-tumoural tissue concentrations of contrast agent were found in high-grade tumours (n = 44) compared to low-grade tumours (n = 12) at every time point (P = 0.006-0.040), providing positive predictive values of 90.6-92.6% in the classification of high-grade tumours. The intra-tumoural and peri-tumoural signal enhancement ratios correlated with tumour grade, size, and Ki67 activity. The intra-observer reproducibility was excellent. We developed a model to measure the 3D intensity data of breast cancers. Low- and high-grade tumours differed in their intra-tumoural and peri-tumoural enhancement characteristics. We anticipate that pharmacokinetic parameters will be increasingly used as imaging biomarkers to model and predict tumour behavior, prognoses, and responses to treatment.

Diagnostic performance of perilesional radiomics analysis of contrast-enhanced mammography for the differentiation of benign and malignant breast lesions

OBJECTIVE

To conduct perilesional region radiomics analysis of contrast-enhanced mammography (CEM) images to differentiate benign and malignant breast lesions.

METHODS AND MATERIALS

This retrospective study included patients who underwent CEM from November 2017 to February 2020. Lesion contours were manually delineated. Perilesional regions were automatically obtained. Seven regions of interest (ROIs) were obtained for each lesion, including the lesion ROI, annular perilesional ROIs (1 mm, 3 mm, 5 mm), and lesion + perilesional ROIs (1 mm, 3 mm, 5 mm). Overall, 4,098 radiomics features were extracted from each ROI. Datasets were divided into training and testing sets (1:1). Seven classification models using features from the seven ROIs were constructed using LASSO regression. Model performance was assessed by the AUC with 95% CI.

RESULTS

Overall, 190 women with 223 breast lesions (101 benign; 122 malignant) were enrolled. In the testing set, the annular perilesional ROI of 3-mm model showed the highest AUC of 0.930 (95% CI: 0.882-0.977), followed by the annular perilesional ROI of 1 mm model (AUC = 0.929; 95% CI: 0.881-0.978) and the lesion ROI model (AUC = 0.909; 95% CI: 0.857-0.961). A new model was generated by combining the predicted probabilities of the lesion ROI and annular perilesional ROI of 3-mm models, which achieved a higher AUC in the testing set (AUC = 0.940).

CONCLUSIONS

Annular perilesional radiomics analysis of CEM images is useful for diagnosing breast cancers. Adding annular perilesional information to the radiomics model built on the lesion information may improve the diagnostic performance.

KEY POINTS

• Radiomics analysis of the annular perilesional region of 3 mm in CEM images may provide valuable information for the differential diagnosis of benign and malignant breast lesions. • The radiomics information from the lesion region and the annular perilesional region may be complementary. Combining the predicted probabilities of the models constructed by the features from the two regions may improve the diagnostic performance of radiomics models.

Machine learning classification of texture features of MRI breast tumor and peri-tumor of combined pre- and early treatment predicts pathologic complete response

Purpose

This study used machine learning classification of texture features from MRI of breast tumor and peri-tumor at multiple treatment time points in conjunction with molecular subtypes to predict eventual pathological complete response (PCR) to neoadjuvant chemotherapy.

Materials and method

This study employed a subset of patients (N = 166) with PCR data from the I-SPY-1 TRIAL (2002–2006). This cohort consisted of patients with stage 2 or 3 breast cancer that underwent anthracycline–cyclophosphamide and taxane treatment. Magnetic resonance imaging (MRI) was acquired pre-neoadjuvant chemotherapy, early, and mid-treatment. Texture features were extracted from post-contrast-enhanced MRI, pre- and post-contrast subtraction images, and with morphological dilation to include peri-tumoral tissue. Molecular subtypes and Ki67 were also included in the prediction model. Performance of classification models used the receiver operating characteristics curve analysis including area under the curve (AUC). Statistical analysis was done using unpaired two-tailed t-tests.

Results

Molecular subtypes alone yielded moderate prediction performance of PCR (AUC = 0.82, p = 0.07). Pre-, early, and mid-treatment data alone yielded moderate performance (AUC = 0.88, 0.72, and 0.78, p = 0.03, 0.13, 0.44, respectively). The combined pre- and early treatment data markedly improved performance (AUC = 0.96, p = 0.0003). Addition of molecular subtypes improved performance slightly for individual time points but substantially for the combined pre- and early treatment (AUC = 0.98, p = 0.0003). The optimal morphological dilation was 3–5 pixels. Subtraction of post- and pre-contrast MRI further improved performance (AUC = 0.98, p = 0.00003). Finally, among the machine-learning algorithms evaluated, the RUSBoosted Tree machine-learning method yielded the highest performance.

Conclusion

AI-classification of texture features from MRI of breast tumor at multiple treatment time points accurately predicts eventual PCR. Longitudinal changes in texture features and peri-tumoral features further improve PCR prediction performance. Accurate assessment of treatment efficacy early on could minimize unnecessary toxic chemotherapy and enable mid-treatment modification for patients to achieve better clinical outcomes.

A Novel Nodule Edge Sharpness Radiomic Biomarker Improves Performance of Lung-RADS for Distinguishing Adenocarcinomas from Granulomas on Non-Contrast CT Scans

Simple Summary

The great majority of pulmonary nodules on screening CT scans are benign (95%). Due to inaccurate diagnoses of granulomas from adenocarcinomas on CT scans, many patients with benign nodules are subjected to unnecessary surgical procedures. The aim of this retrospective study is to evaluate the discriminability of a new radiomic feature, nodule edge/interface sharpness (NIS), for distinguishing lung adenocarcinomas from benign granulomas on non-contrast CT scans. Moreover, we aim to evaluate whether NIS can improve the performance of Lung-RADS, by reclassifying benign nodules that were initially assessed as suspicious. In a cohort of 352 patients with diagnostic non-contrast CT scans, NIS radiomics was able to classify nodules with an area under the receiver operating characteristic curve (ROC AUC) of 0.77, and when combined with intra-tumoral textural and shape features, classification performance increased to AUC of 0.84. Additionally, the NIS classifier correctly reclassified 46% of those lesions that were actually benign but deemed suspicious by Lung-RADS. Combining NIS with Lung-RADS has the potential to alter patient management by significantly decreasing unnecessary biopsies/follow up imaging.

Abstract

The aim of this study is to evaluate whether NIS radiomics can distinguish lung adenocarcinomas from granulomas on non-contrast CT scans, and also to improve the performance of Lung-RADS by reclassifying benign nodules that were initially assessed as suspicious. The screening or standard diagnostic non-contrast CT scans of 362 patients was divided into training (S_t, N = 145), validation (S_v, N = 145), and independent validation (S_iv, N = 62) sets from different institutions. Nodules were identified and manually segmented on CT images by a radiologist. A series of 264 features relating to the edge sharpness transition from the inside to the outside of the nodule were extracted. The top 10 features were used to train a linear discriminant analysis (LDA) machine learning classifier on St. In conjunction with the LDA classifier, NIS radiomics classified nodules with an AUC of 0.82 ± 0.04, 0.77, and 0.71 respectively on S_t, S_v, and S_iv. We evaluated the ability of the NIS classifier to determine the proportion of the patients in S_v that were identified initially as suspicious by Lung-RADS but were reclassified as benign by applying the NIS scores. The NIS classifier was able to correctly reclassify 46% of those lesions that were actually benign but deemed suspicious by Lung-RADS alone on S_v.

The rate of breast fibroglandular enhancement during dynamic contrast-enhanced MRI reflects response to neoadjuvant therapy

PURPOSE

This study assesses the rate of enhancement of breast fibroglandular tissue after administration of a magnetic resonance imaging (MRI) gadolinium-based contrast agent and determines its relationship with response to neoadjuvant therapy (NAT) in women with breast cancer.

METHOD

Women with locally advanced breast cancer (N = 19) were imaged four times over the course of NAT. Dynamic contrast-enhanced (DCE) MRI was acquired after administration of a gadolinium-based contrast agent with a temporal resolution of 7.27 s. The tumor, fibroglandular tissue, and adipose tissue were semi-automatically segmented using a manually drawn region of interest encompassing the tumor followed by fuzzy c-means clustering. The rate and relative intensity of signal enhancement were calculated for each voxel within the tumor and fibroglandular tissue.

RESULTS

The rate of fibroglandular tissue enhancement after contrast agent injection declined by an average of 29 % over the course of NAT. This decline was present in 16 of the 19 patients in the study. The rate of enhancement is significantly higher in women who achieve pathological complete response (pCR) after both 1 cycle (68 % higher, p < 0.05) and after 3-5 cycles of NAT (58 % higher; p < 0.05). The relative intensity of fibroglandular enhancement correlates with the rate of enhancement (R2 = 0.64, p < 0.001) and is higher in women who achieve pCR after both 1 cycle and after 3-5 cycles of NAT (p < 0.05, both timepoints).

CONCLUSION

The rate of fibroglandular tissue enhancement declines over the course of therapy, provides novel information not reflected by tumoral measures, and may predict pathological response early in the course of therapy, with smaller declines in enhancement in women who achieve favorable response.

Pattern Classification Approaches for Breast Cancer Identification via MRI: State-Of-The-Art and Vision for the Future

Mining algorithms for Dynamic Contrast Enhanced Magnetic Resonance Imaging (DCE-MRI) of breast tissue are discussed. The algorithms are based on recent advances in multi-dimensional signal processing and aim to advance current state-of-the-art computer-aided detection and analysis of breast tumours when these are observed at various states of development. The topics discussed include image feature extraction, information fusion using radiomics, multi-parametric computer-aided classification and diagnosis using information fusion of tensorial datasets as well as Clifford algebra based classification approaches and convolutional neural network deep learning methodologies. The discussion also extends to semi-supervised deep learning and self-supervised strategies as well as generative adversarial networks and algorithms using generated confrontational learning approaches. In order to address the problem of weakly labelled tumour images, generative adversarial deep learning strategies are considered for the classification of different tumour types. The proposed data fusion approaches provide a novel Artificial Intelligence (AI) based framework for more robust image registration that can potentially advance the early identification of heterogeneous tumour types, even when the associated imaged organs are registered as separate entities embedded in more complex geometric spaces. Finally, the general structure of a high-dimensional medical imaging analysis platform that is based on multi-task detection and learning is proposed as a way forward. The proposed algorithm makes use of novel loss functions that form the building blocks for a generated confrontation learning methodology that can be used for tensorial DCE-MRI. Since some of the approaches discussed are also based on time-lapse imaging, conclusions on the rate of proliferation of the disease can be made possible. The proposed framework can potentially reduce the costs associated with the interpretation of medical images by providing automated, faster and more consistent diagnosis.

Correlation of apparent diffusion coefficient values and peritumoral edema with pathologic biomarkers in patients with breast cancer

PURPOSE

To investigate the relationship between breast cancer imaging features on magnetic resonance imaging (MRI) and histopathological characteristics.

METHODS AND MATERIALS

We prospectively enrolled 46 patients who underwent 1.5-T MRI with 68 breast malignant lesions from 2017 until 2019. Peritumoral edema was determined based on visual assessment on T2 weighted imaging. Lesions were categorized into two groups: A: with edema (48 lesions) and B: without edema (20 lesions).

RESULTS

The tumor size was not different among two groups but multifocal-multicentric lesions were more common in the group A (70% vs. 35%). The axillary lymph nodes are most involved in group A. ER and PR positive lesions were more common in group B (90% vs. 56.3%) but in the group A, HER2 positive lesions were found to be more common (31.3% vs. 15%). The mean ADC value in tumors and peritumoral regions were lower (0.97 × 10^-3 mm²/s, P = 0.023) and higher (1.85 × 10^-3 mm²/s, P < 0.0001) in group A, respectively. Peritumoral ADC value was significantly higher in HER2-positive group.

CONCLUSION

Breast carcinomas with peritumoral edema were found to be more multifocal-multicentric, with higher prevalence of axillary lymph node involvement, more HER 2-positive, with lower prevalence of ER/PR-positive, lower tumoral ADC and higher peritumoral ADC values.

Correlation of Peri-Tumoral Edema Determined in T2 Weighted Imaging with Apparent Diffusion Coefficient of Peritumoral Area in Patients with Breast Carcinoma

Background: Breast cancer may result in remodeling of adjacent normal appearing breast tissues. Magnetic resonance imaging (MRI) is increasingly used in the diagnosis and follow-up of breast cancer by means of diffusion weighted imaging, which is based on thermal motion of water molecules in the extracellular fluid. Objectives: We investigated the correlation of visual assessment of peri-tumoral edema with peri-tumoral and tumoral apparent diffusion coefficient (ADC) values. Patients and Methods: In this cross-sectional study, from 2016 to 2018, 78 patients with 89 malignant breast lesions (mean age, 47 years) were examined by 1.5-T breast MRI. The lesions were categorized based on the visual assessment of peri-tumoral edema on T2 weighted imaging (T2WI) into two groups: (A) with edema (36 lesions) and (B) without edema (53 lesions). Measuring ADC values in the contralateral normal breast tissue, peri-tumoral tissue and peri-tumoral-normal tissue ADC ratio were compared between the two groups for all lesions. Results: The number of in situ lesions was higher in group B (7.5% vs 2.7%) with the p value of 0.01. The mean of ADC values in the normal breast tissue was 1.76 × 10-3mm2/s. Tumor ADCs were significantly lower in group A compared to group B (0.95 × 10-3mm2/s vs. 1.11 × 10-3mm2/s) with the P value of 0.003. However, peri-tumoral ADCs were significantly higher in group A (1.82 × 10-3mm2/s vs. 1.53 × 10-3mm2/s) with the p value of 0.005. The peri-tumoral-normal tissue ADC ratio was 0.87 in group B and about 1 in group A. However, the difference between normal tissue ADCs and peri-tumoral ADCs was only significant (P value of 0.005) in group B. The cut-off point value for differentiating normal tissue ADCs and peri-tumoral ADCs was 1.61 × 10-3mm2/s with the sensitivity of 65% and specificity of 70%. Conclusion: Breast cancer with peri-tumoral edema has lower tumoral ADC values, higher peri-tumoral ADC values and lower prevalence of in situ lesions. Visual assessment of peri-tumoral edema on T2WI could predict the tumoral characteristic on diffusion-weighted imaging.

Generative adversarial network-based super-resolution of diffusion-weighted imaging: Application to tumour radiomics in breast cancer

Diffusion-weighted imaging (DWI) is increasingly used to guide the clinical management of patients with breast tumours. However, accurate tumour characterization with DWI and the corresponding apparent diffusion coefficient (ADC) maps are challenging due to their limited resolution. This study aimed to produce super-resolution (SR) ADC images and to assess the clinical utility of these SR images by performing a radiomic analysis for predicting the histologic grade and Ki-67 expression status of breast cancer. To this end, 322 samples of dynamic enhanced magnetic resonance imaging (DCE-MRI) and the corresponding DWI data were collected. A SR generative adversarial (SRGAN) and an enhanced deep SR (EDSR) network along with the bicubic interpolation were utilized to generate SR-ADC images from which radiomic features were extracted. The dataset was randomly separated into a development dataset (n = 222) to establish a deep SR model using DCE-MRI and a validation dataset (n = 100) to improve the resolution of ADC images. This random separation of datasets was performed 10 times, and the results were averaged. The EDSR method was significantly better than the SRGAN and bicubic methods in terms of objective quality criteria. Univariate and multivariate predictive models of radiomic features were established to determine the area under the receiver operating characteristic curve (AUC). Individual features from the tumour SR-ADC images showed a higher performance with the EDSR and SRGAN methods than with the bicubic method and the original images. Multivariate analysis of the collective radiomics showed that the EDSR- and SRGAN-based SR-ADC images performed better than the bicubic method and original images in predicting either Ki-67 expression levels (AUCs of 0.818 and 0.801, respectively) or the tumour grade (AUCs of 0.826 and 0.828, respectively). This work demonstrates that in addition to improving the resolution of ADC images, deep SR networks can also improve tumour image-based diagnosis in breast cancer.

Peritumor to tumor apparent diffusion coefficient ratio is associated with biologically more aggressive breast cancer features and correlates with the prognostication tools

PURPOSE

The apparent diffusion coefficient (ADC) is increasingly used to characterize breast cancer. The peritumor/tumor ADC ratio is suggested to be a reliable and generally applicable index. However, its overall prognostication value remains unclear. We aimed to evaluate the associations between the peritumor/tumor ADC ratio and histopathological biomarkers and published prognostic tools in patients with invasive breast cancer.

MATERIALS AND METHODS

This prospective study included 88 lesions (five bilateral) in 83 patients with primary invasive breast cancer who underwent preoperative 3.0-T magnetic resonance imaging. The lowest intratumoral mean ADC value on the slice with the largest tumor cross-sectional area was designated the tumor ADC, and the highest mean ADC value on the peritumoral breast parenchymal tissue adjacent to the tumor border was designated the peritumor ADC. The peritumor/tumor ADC ratio was then calculated. The tumor and peritumor ADC values and peritumor/tumor ADC ratios were compared with histopathological parameters using an unpaired t test, and their correlations with published prognostic tools were evaluated with Pearson's correlation coefficient.

RESULTS

The peritumor/tumor ADC ratio was significantly associated with tumor size (p<0.001), histological grade (p = 0.005), Ki-67 index (p = 0.006), axillary-lymph-node metastasis (p = 0.001), and lymphovascular invasion (p = 0.006), but was not associated with estrogen receptor status (p = 0.931), progesterone receptor status (p = 0.160), or human epidermal growth factor receptor 2 status (p = 0.259). The peritumor/tumor ADC ratio showed moderate positive correlations with the Nottingham Prognostic Index (r = 0.498, p<0.001) and mortality predicted using PREDICT (r = 0.436, p<0.001).

CONCLUSION

The peritumor/tumor ADC ratio was correlated with histopathological biomarkers in patients with invasive breast cancer, showed significant correlations with published prognostic indexes, and may provide an easily applicable imaging index for the preoperative prognostic evaluation of breast cancer.

Can conventional DWI accurately assess the size of endometrial cancer?

PURPOSE

To compare T2-weighted image (T2WI) and conventional Diffusion-weighted image (cDWI) of magnetic resonance imaging (MRI) for sensitivity of qualitative diagnosis and accuracy of tumor size (TS) measurement in endometrial cancer (EC). Meanwhile, the effect of the lesion size itself and tumor grade on the ability of T2WI and cDWI of TS assessment was explored. Ultimately, the reason of deviation on size evaluation was studied.

MATERIALS AND METHODS

34 patients with EC were enrolled. They were all treated with radical hysterectomy and performed MR examinations before operation. Firstly, the sensitivity of T2WI alone and T2WI-DWI in qualitative diagnosis of EC were compared according to pathology. Secondly, TS on T2WI and cDWI described with longitudinal (LD) and horizontal diameter (HD) were compared to macroscopic surgical specimen (MSS) quantitatively in the entire lesions and the subgroup lesions which grouped by postoperative tumor size itself and tumor grade. Thirdly, the discrepancy of mean ADC values (ADC mean) and range ADC values (ADC range) between different zones of EC were explored.

RESULTS

For qualitative diagnosis, the sensitivity of T2WI-DWI (97%) was higher than T2WI alone (85%) (p = 0.046).For TS estimation, no significant difference (PLD = 0.579; PHD = 0.261) was observed between T2WI (LDT2WI = 3.90 cm; HDT2WI = 2.88 cm) and MSS (LD = 4.00 cm; HD = 3.06 cm), whereas TS of cDWI (LDDWI = 3.01 cm; HDDWI = 2.54 cm) were smaller than MSS (PLD = 0.002; PHD = 0.002) in all lesions. In subgroup of tumor with G1 (grade 1) and small lesion (defined as maximum diameter < 3 cm), both T2WI and cDWI were not significantly different from MSS; In subgroup of tumor with G2 + 3 (grade 2 and grade 3) and big lesion (maximum diameter ≥ 3 cm), T2WI matched well with MSS still, but DWI lost accuracy significantly. The result of ADC values between different zones of tumor showed ADC mean of EC rose from central zone to peripheral zone of tumor gradually and ADC range widened gradually.

CONCLUSION

cDWI can detect EC very sensitively. The TS on cDWI was smaller than the fact for the ECs with G2/3 and big size. The TS of T2WI was in accordance with the actual size for all ECs. The heterogeneity may be responsible for the inaccuracy of cDWI.

Diagnosis of Benign and Malignant Breast Lesions on DCE-MRI by Using Radiomics and Deep Learning With Consideration of Peritumor Tissue

BACKGROUND

Computer-aided methods have been widely applied to diagnose lesions detected on breast MRI, but fully-automatic diagnosis using deep learning is rarely reported.

PURPOSE

To evaluate the diagnostic accuracy of mass lesions using region of interest (ROI)-based, radiomics and deep-learning methods, by taking peritumor tissues into consideration.

STUDY TYPE

Retrospective.

POPULATION

In all, 133 patients with histologically confirmed 91 malignant and 62 benign mass lesions for training (74 patients with 48 malignant and 26 benign lesions for testing).

FIELD STRENGTH/SEQUENCE

3T, using the volume imaging for breast assessment (VIBRANT) dynamic contrast-enhanced (DCE) sequence.

ASSESSMENT

3D tumor segmentation was done automatically by using fuzzy-C-means algorithm with connected-component labeling. A total of 99 texture and histogram parameters were calculated for each case, and 15 were selected using random forest to build a radiomics model. Deep learning was implemented using ResNet50, evaluated with 10-fold crossvalidation. The tumor alone, smallest bounding box, and 1.2, 1.5, 2.0 times enlarged boxes were used as inputs.

STATISTICAL TESTS

The malignancy probability was calculated using each model, and the threshold of 0.5 was used to make a diagnosis.

RESULTS

In the training dataset, the diagnostic accuracy was 76% using three ROI-based parameters, 84% using the radiomics model, and 86% using ROI + radiomics model. In deep learning using the per-slice basis, the area under the receiver operating characteristic (ROC) was comparable for tumor alone, smallest and 1.2 times box (AUC = 0.97-0.99), which were significantly higher than 1.5 and 2.0 times box (AUC = 0.86 and 0.71, respectively). For per-lesion diagnosis, the highest accuracy of 91% was achieved when using the smallest bounding box, and that decreased to 84% for tumor alone and 1.2 times box, and further to 73% for 1.5 times box and 69% for 2.0 times box. In the independent testing dataset, the per-lesion diagnostic accuracy was also the highest when using the smallest bounding box, 89%.

DATA CONCLUSION

Deep learning using ResNet50 achieved a high diagnostic accuracy. Using the smallest bounding box containing proximal peritumor tissue as input had higher accuracy compared to using tumor alone or larger boxes.

LEVEL OF EVIDENCE

3 Technical Efficacy: Stage 2.

Peritumoral ADC values in breast cancer: region of interest selection, associations with hyaluronan intensity, and prognostic significance

Objectives

We aimed to evaluate the differences in peritumoral apparent diffusion coefficient (ADC) values by four different ROI selection methods and to validate the optimal method. Furthermore, we aimed to evaluate if the peritumor-tumor ADC ratios are correlated with axillary lymph node positivity and hyaluronan accumulation.

Methods

Altogether, 22 breast cancer patients underwent 3.0-T breast MRI, histopathological evaluation, and hyaluronan assay. Paired t and Friedman tests were used to compare minimum, mean, and maximum values of tumoral and peritumoral ADC by four methods: (M1) band ROI, (M2) whole tumor surrounding ROI, (M3) clockwise multiple ROI, and (M4) visual assessment of ROI selection. Subsequently, peritumor/tumor ADC ratios were compared with hyaluronan levels and axillary lymph node status by the Mann-Whitney U test.

Results

No statistically significant differences were found among the four ROI selection methods regarding minimum, mean, or maximum values of tumoral and peritumoral ADC. Visual assessment ROI measurements represented the less time-consuming evaluation method for the peritumoral area, and with sufficient accuracy. Peritumor/tumor ADC ratios obtained by all methods except the clockwise ROI (M3) showed a positive correlation with hyaluronan content (M1, p = 0.004; M2, p = 0.012; M3, p = 0.20; M4, p = 0.025) and lymph node metastasis (M1, p = 0.001; M2, p = 0.007; M3, p = 0.22; M4, p = 0.015), which are established factors for unfavorable prognosis.

Conclusions

Our results suggest that the peritumor/tumor ADC ratio could be a readily applicable imaging index associated with axillary lymph node metastasis and extensive hyaluronan accumulation. It could be related to the biological aggressiveness of breast cancer and therefore might serve as an additional prognostic factor.

Key Points

• Out of four different ROI selection methods for peritumoral ADC evaluation, measurements based on visual assessment provided sufficient accuracy and were the less time-consuming method.

• The peritumor/tumor ADC ratio can provide an easily applicable supplementary imaging index for breast cancer assessment.

• A higher peritumor/tumor ADC ratio was associated with axillary lymph node metastasis and extensive hyaluronan accumulation and might serve as an additional prognostic factor.

Non-contrast MRI for breast screening: preliminary study on detectability of benign and malignant lesions in women with dense breasts

PURPOSE

The importance of breast cancer screening has long been known. Unfortunately, there is no imaging modality for screening women with dense breasts that is both sensitive and without concerns regarding potential side effects. The purpose of this study is to explore the possibility of combined diffusion-weighted imaging and turbo inversion recovery magnitude MRI (DWI + TIRM) to overcome the difficulty of detection sensitivity and safety.

METHODS

One hundred and seventy-six breast lesions from 166 women with dense breasts were retrospectively evaluated. The lesion visibility, area under the curve (AUC), sensitivity and specificity of cancer detection by MG, DWI + TIRM, and clinical MRI were evaluated and compared. MG plus clinical MRI served as the gold standard for lesion detection and pathology served as the gold standard for cancer detection.

RESULTS

Lesion visibility of DWI + TIRM (96.6%) was significantly superior to MG (67.6%) in women with dense breasts (p < 0.001). There was no significant difference compared with clinical MRI. DWI + TIRM showed higher accuracy (AUC = 0.935) and sensitivity (93.68%) for breast cancer detection than MG (AUC = 0.783, sensitivity = 46.32%), but was comparable to clinical MRI (AUC = 0.944, sensitivity = 93.68%). The specificity of DWI + TIRM (83.95%) was lower than MG (98.77%), but higher than clinical MRI (77.78%).

CONCLUSIONS

DWI combined with TIRM could be a safe, sensitive, and practical alternative for screening women with dense breasts.

Apparent diffusion coefficient values in differential diagnosis and prognostic prediction of solitary of fibrous tumor/hemangiopericytoma (WHOII) and atypical meningioma

BACKGROUND

It is difficult to distinguish solitary of fibrous tumor/hemangiopericytoma (SFT/HPC) from atypical meningioma (AM) by conventional imaging.As far as we know,diffusion weighting imaging may identify them effectively.

OBJECTIVE

The purpose of this study was to determine the role of apparent diffusion coefficient (ADC) values to distinguish and predict prognosis of solitary of fibrous tumor/hemangiopericytoma (SFT/HPC) (WHOII) and atypical meningioma (AM).

METHODS

Preoperative diffusion-weighted imaging (DWI) of 30 cases with histopathologic and immunhistochemical testified SFT/HPC WHOII (n= 11) and AM (n= 19) were performed retrospectively. The ADC values of lesion, peritumoral edema, normal white matter and lesion NADC ratio (lesion ADC values/ADC values of normal white matter (NWN ADC)) were compared. The immunhistochemical markers (Ki-67, CD34, Vim, EMA, GFAP, S-100, PR, CD56) were compared. The correlation between the ADC values and Ki-67 index was evaluated.

RESULTS

The mean lesion ADC values of SFT/HPC (1.15 ± 0.04 × 10-3 mm2/s) was significantly higher than that of AM (0.80 ± 0.04 × 10-3 mm2/s) (t= 23.824, p< 0.05). The mean NADC ratio was lower for AM (1.03 ± 0.06) compared with SFT/HPC (1.51 ± 0.05) (t= 23.105, p< 0.05). The mean edema ADC for SFT/HPC (1.47 ± 0.06 × 10-3 mm2/s) was lower compared with AM (1.68 ± 0.05 × 10-3 mm2/s) (t=-9.926, p< 0.05 ). There was no statistical difference between the two groups of NWM ADC (t=-1.475, p> 0.05) . The mean Ki-67 of SFT/HPC (7.18 ± 2.60%) was lower than the mean Ki-67 of AM (13.58 ± 4.50%) (t=-4.934, p< 0.05). The CD34 showed statistically differences between two groups (X2= 13.659, p< 0.05). The EMA also showed statistically differences between two groups (X2= 4.474, p< 0.05). Vim,GFAP, S-100, PR, CD56 showed no statistical difference in the two group (p> 0.05). The pearson analysis indicated that there was a negative correlation between lesion ADC and Ki-67 in SFT/HPC group (r=-0.770, p< 0.05) and AM group (r=-0.727, p< 0.05). There was also a negative correlation between lesion NADC ratio and Ki-67 in SFT/HPC group (r=-0.673, p< 0.05) and AM group (r=-0.707, p< 0.05). There was a positive correlation between edema ADC and Ki-67 in SFT/HPC group (r= 0.819, p< 0.05) and AM group (r= 0.942, p< 0.05). Furthermore,there was no correlation between NWM A DC and Ki-67 in SFT/HPC group (r=-0.403, p> 0.05) and AM group (r= 0.202, p> 0.05).

CONCLUSIONS

The lesion ADC, lesion NADC ratio and edema ADC can distinguish the SFT/HPC WHO II from AM and be helpful to predict prognosis of the two tumors before operation. Further more, histopathologic and immunhistochemical can make a definite diagnosis of the two tumors.

Quantitative analysis of peri-tumor fat in different molecular subtypes of breast cancer

BACKGROUND AND PURPOSES

The aim of this study was to develop morphological analytic methods to analyze the tumor-fat interface and in different peritumoral shells away from the tumor, and to compare the results among three molecular subtypes of breast cancer.

MATERIALS AND METHODS

A total of 102 women (mean age 48.5 y/o) with solitary well-defined breast cancers were analyzed, including 46 human epidermal growth factor receptor 2 (HER2) (+), 46 HER2(-) hormonal receptor (HR) (+), and 10 triple negative (TN) breast cancers. The tumor lesion, the breast, the fibroglandular and fatty tissue were segmented using well-established methods. The whole breast fat percentage and the peri-tumor interface fat percentage were measured. Three shells (SH1, SH2, SH3) surrounding the convex hall of the three dimensional (3D) tumor were defined and in each shell the volumetric percentage of fat was calculated. The peri-tumor interface fat percentage and the volumetric percentage of fat in the three peri-tumoral shells were compared among different subtypes.

RESULTS

In the TN group, the fat percentage on the tumor boundary was 43 ± 20% and 78 ± 12% for two dimensional (2D) and 3D measurement, respectively, which were the highest among the three subtypes but not significantly different. The fat percentage in SH2 and SH3 in the TN group was 82 ± 7% and 85 ± 7%, which was significantly higher compared to the two other two subtypes. The results remained after controlling for the whole breast fat percentage.

CONCLUSIONS

This study provided a feasible method for quantitative analysis of peri-tumoral tissue characteristics. Because of small patient number, the finding that TN tumors had the highest peri-tumor fat content among the three subtypes needs to be further verified with a large cohort study.

Comparison of peritumoral stromal tissue stiffness obtained by shear wave elastography between benign and malignant breast lesions

Background Aggressive breast cancers produce abnormal peritumoral stiff areas, which can differ between benign and malignant lesions and between different subtypes of breast cancer. Purpose To compare the tissue stiffness of the inner tumor, tumor border, and peritumoral stroma (PS) between benign and malignant breast masses by shear wave elastography (SWE). Material and Methods We enrolled 133 consecutive patients who underwent preoperative SWE. Using OsiriX commercial software, we generated multiple 2-mm regions of interest (ROIs) in a linear arrangement on the inner tumor, tumor border, and PS. We obtained the mean elasticity value (E_mean) of each ROI, and compared the E_mean between benign and malignant tumors. Odds ratios (ORs) for prediction of malignancy were calculated. Subgroup analyses were performed among tumor subtypes. Results There were 85 malignant and 48 benign masses. The E_mean of the tumor border and PS were significantly different between benign and malignant masses ( P < 0.05 for all). ORs for malignancy were 1.06, 1.08, 1.05, and 1.04 for stiffness of the tumor border, proximal PS, middle PS, and distal PS, respectively ( P < 0.05 for all). Malignant masses with a stiff rim were significantly larger than malignant masses without a stiff rim, and were more commonly associated with the luminal B and triple negative subtypes. Conclusion Stiffness of the tumor border and PS obtained by SWE were significantly different between benign and malignant masses. Malignant masses with a stiff rim were larger in size and associated with more aggressive pathologic subtypes.

Diffusion-weighted imaging features of breast tumours and the surrounding stroma reflect intrinsic heterogeneous characteristics of molecular subtypes in breast cancer

Breast cancer heterogeneity is the main obstacle preventing the identification of patients with breast cancer with poor prognoses and treatment responses; however, such heterogeneity has not been well characterized. The purpose of this retrospective study was to reveal heterogeneous patterns in the apparent diffusion coefficient (ADC) signals in tumours and the surrounding stroma to predict molecular subtypes of breast cancer. A dataset of 126 patients with breast cancer, who underwent preoperative diffusion-weighted imaging (DWI) on a 3.0-T image system, was collected. Breast images were segmented into regions comprising the tumour and surrounding stromal shells in which features that reflect heterogeneous ADC signal distribution were extracted. For each region, imaging features were computed, including the mean, minimum, variance, interquartile range (IQR), range, skewness, kurtosis and entropy of ADC values. Univariate and stepwise multivariate logistic regression modelling was performed to identify the magnetic resonance imaging features that optimally discriminate luminal A, luminal B, human epidermal growth factor 2 (HER2)-enriched and basal-like molecular subtypes. The performance of the predictive models was evaluated using the area under the receiver operating characteristic curve (AUC). Univariate logistic regression analysis showed that the skewness in the tumour boundary achieved an AUC of 0.718 for discrimination between luminal A and non-luminal A tumours, whereas the IQR of the ADC value in the tumour boundary had an AUC of 0.703 for classification of the HER2-enriched subtype. Imaging features in the tumour boundary and the proximal peritumoral stroma corresponded to a higher overall prediction performance than those in other regions. A multivariate logistic regression model combining features in all the regions achieved an overall AUC of 0.800 for the classification of the four tumour subtypes. These findings suggest that features in the tumour boundary and stroma around the tumour may be further assessed as potential predictors of molecular subtypes of breast cancer.

Comparison of microvascular perfusion evaluation among IVIM-DWI, CT perfusion imaging and histological microvessel density in rabbit liver VX2 tumors

OBJECT

To explore microcirculation features with intravoxel incoherent motion (IVIM) and to compare IVIM with CT perfusion imaging (CTPI) and microvessel density (MVD).

MATERIALS AND METHODS

Hepatic CTPI and IVIM were performed in 16 rabbit liver VX2 tumor models. Hepatic arterial perfusion (HAP), hepatic arterial perfusion index (HPI), Blood flow (BF), and blood volume (BV) from CTPI were measured. Apparent diffusion coefficient (ADC), true diffusion coefficient (D), perfusion fraction (f), and pseudo-diffusion coefficient (D*) from IVIM were measured. MVD was counted with CD34 stain. The microcirculation features with IVIM were compared with CTPI parameters and MVD.

RESULTS

Strong linear correlations were found between D value (0.89±0.21×10^-3mm²/s) and HAP (15.83±6.97ml/min/100mg) (r=0.755, P=0.001) and between f value (12.64±6.66%) and BV (9.74±5.04ml/100mg) (r=0.693, P=0.004). Moderate linear correlations were observed between ADC (1.07±0.32×10^-3mm²/s) and HAP (r=0.538, P=0.039), respectively; and between D value and MVD (9.31±2.57 vessels at 400×magnification) (r=0.509, P=0.044). No correlations were found between D* (119.90±37.67×10^-3mm²/s) and HAP, HPI (68.34±12.91%), BF (4.95±2.16ml/min/100mg), BV.

CONCLUSION

IVIM parameters can characterize microcirculation to certain extent and separate it from pure water molecular diffusion. There is fair correlation between D or ADC value and CTPI parameters or MVD, but no correlation between D* or f value and CTPI parameters or MVD except f value and BV, which is still unclear and need further clinical studies to validate.

Heterogeneity of Diffusion-Weighted Imaging in Tumours and the Surrounding Stroma for Prediction of Ki-67 Proliferation Status in Breast Cancer

Breast tissue heterogeneity is related to risk factors that lead to more aggressive tumour growth and worse prognosis, yet such heterogeneity has not been well characterized. The aim of this study is to reveal the heterogeneous signal patterns of the apparent diffusion coefficient (ADC) of a tumour and its surrounding stromal tissue and to predict the Ki-67 proliferation status in oestrogen receptor (ER)-positive breast cancer patients. A dataset of 82 patients who underwent diffusion-weighted imaging (DWI) examination was collected. The ADC map was segmented into regions comprising the tumour and the surrounding stromal shells. To reflect correlations between each region in terms of its mean ADC value, a functional graph was constructed consisting of nodes as regions and edges as interactions between two nodes. Analysis of the graph revealed a higher average degree in samples over-expressing Ki-67 than in samples with low Ki-67 expression. In the low-Ki-67 group, most of the identified edges represented correlations between adjacent regions, whereas additional edges representing correlations between non-adjacent regions were found in the high-Ki-67 group. The ADC signal in various breast stromal regions surrounding the tumour showed a discriminative pattern and would be valuable for estimating the Ki-67 proliferation status by DWI.

Radiogenomic Analysis of Oncological Data: A Technical Survey

In the last few years, biomedical research has been boosted by the technological development of analytical instrumentation generating a large volume of data. Such information has increased in complexity from basic (i.e., blood samples) to extensive sets encompassing many aspects of a subject phenotype, and now rapidly extending into genetic and, more recently, radiomic information. Radiogenomics integrates both aspects, investigating the relationship between imaging features and gene expression. From a methodological point of view, radiogenomics takes advantage of non-conventional data analysis techniques that reveal meaningful information for decision-support in cancer diagnosis and treatment. This survey is aimed to review the state-of-the-art techniques employed in radiomics and genomics with special focus on analysis methods based on molecular and multimodal probes. The impact of single and combined techniques will be discussed in light of their suitability in correlation and predictive studies of specific oncologic diseases.