Differentiation between idiopathic granulomatous mastitis and invasive breast carcinoma, both presenting with non-mass enhancement without rim-enhanced masses: The value of whole-lesion histogram and texture analysis using apparent diffusion coefficient

Differentiation of Malignancy and Idiopathic Granulomatous Mastitis Presenting as Non-mass Lesions on MRI: Radiological, Clinical, Radiomics, and Clinical-Radiomics Models

RATIONALE AND OBJECTIVES

To investigate the effectiveness of machine learning-based clinical, radiomics, and combined models in differentiating idiopathic granulomatous mastitis (IGM) from malignancy, both presenting as non-mass enhancement (NME) lesions on magnetic resonance imaging (MRI), and to compare these models with radiological evaluation.

MATERIAL AND METHODS

A total of 178 patients (69 IGM and 109 breast cancer patients) with NME on breast MRI evaluated between March 2018 and April 2022, were included in this two-center study. Age, skin changes, presence of fistula, and abscess were recorded from hospital records. Two experienced radiologists evaluated MRI images according to the breast imaging reporting and data system 2013 lexicon. Lesions were segmented independently on T2-weighted, apparent diffusion coefficient, and post-contrast-T1-weighted sequences. Data were split into training and external testing sets. Machine learning models were built using Light GBM (light gradient-boosting machine). Radiological, clinical, radiomics, and clinical-radiomics models were created and compared. Decision curve analysis was performed. Quality of reporting and that of methodology were evaluated using CLEAR and METRICS tools.

RESULTS

IGM group was younger (p = 0.014). Abscesses (p < 0.001), fistulas (p < 0.001), and skin changes (p < 0.001) were significantly more common in the IGM group. No significant difference was detected in terms of lesion size (p = 0.213). In the evaluation of NME, the lowest performance belonged to the radiologists' evaluation (AUC for training, 0.740; for testing, 0.737), while the highest AUC was achieved by the model developed by combined clinical and radiomics features (AUC for training, 0.979; for testing, 0.942).

CONCLUSION

Our study has shown that the machine learning-based clinical-radiomics model might have the potential to accurately discriminate IGM and malignant lesions in evaluating NME areas.

Granulomatous lobular mastitis co-existing with ductal carcinoma in situ: Report of three cases and review of the literature

Granulomatous lobular mastitis (GLM) is a benign and infrequent chronic breast ailment. Although this lesion can be clinically and radiographically mistaken for early-onset breast cancer, it is a rare occurrence for the two to coexist. This report describes three such cases. In all three patients, the primary signs and symptoms were related to the formation of diffuse breast masses or abscesses. Breast ultrasound and MRI revealed glandular edema and dilated breast ducts. The biopsies of all lesions exhibited both granulomatous inflammation confined to the lobules of the breast, abundant interstitial inflammatory cell infiltrates, and apparently cancerous cells located in dilated ducts with intact basement membranes. The surgically excised specimens confirmed the diagnosis of GLM and ductal carcinoma in situ (DCIS) in all three patients who underwent breast mass resection. By clinical imaging and clinical manifestations, GLM may obscure a concurrent DCIS, as highlighted by the cases reported herein.

A sonogram radiomics model for differentiating granulomatous lobular mastitis from invasive breast cancer: a multicenter study

PURPOSE

To construct a nomogram based on sonogram features and radiomics features to differentiate granulomatous lobular mastitis (GLM) from invasive breast cancer (IBC).

MATERIALS AND METHODS

A retrospective collection of 213 GLMs and 472 IBCs from three centers was divided into a training set, an internal validation set, and an external validation set. A radiomics model was built based on radiomics features, and the RAD score of the lesion was calculated. The sonogram radiomics model was constructed using ultrasound features and RAD scores. Finally, the diagnostic efficacy of the three sonographers with different levels of experience before and after combining the RAD score was assessed in the external validation set.

RESULTS

The RAD score, lesion diameter, orientation, echogenicity, and tubular extension showed significant differences in GLM and IBC (p < 0.05). The sonogram radiomics model based on these factors achieved optimal performance, and its area under the curve (AUC) was 0.907, 0.872, and 0.888 in the training, internal, and external validation sets, respectively. The AUCs before and after combining the RAD scores were 0.714, 0.750, and 0.830 and 0.834, 0.853, and 0.878, respectively, for sonographers with different levels of experience. The diagnostic efficacy was comparable for all sonographers when combined with the RAD score (p > 0.05).

CONCLUSION

Radiomics features effectively enhance the ability of sonographers to discriminate between GLM and IBC and reduce interobserver variation. The nomogram combining ultrasound features and radiomics features show promising diagnostic efficacy and can be used to identify GLM and IBC in a noninvasive approach.

Idiopathic Granulomatous Mastitis or Breast Cancer? A Comparative MRI Study in Patients Presenting with Non-Mass Enhancement

OBJECTIVE

To compare and determine discriminative magnetic resonance imaging (MRI) findings of idiopathic granulomatous mastitis (IGM) and breast cancer (BC) that present as non-mass enhancement.

MATERIALS AND METHODS

This retrospective study includes 68 IGM and 75 BC cases that presented with non-mass enhancement on breast MRI. All patients with a previous history of breast surgery, radiotherapy, or chemotherapy due to BC or a previous history of mastitis were excluded. On MRI images, presence of architectural distortion skin thickening, edema, hyperintense ducts containing protein, dilated fat-containing ducts and axillary adenopathies were noted. Cysts with enhancing walls, lesion size, lesion location, fistulas, distribution, internal enhancement pattern and kinetic features of non-mass enhancement were recorded. Apparent diffusion coefficient (ADC) values were calculated. Pearson chi-square test, Fisher's exact test, independent t test and Mann-Whitney U test were used as needed for statistical analysis and comparison. Multivariate logistic regression model was used to determine the independent predictors.

RESULTS

IGM patients were significantly younger than BC patients (p < 0.001). Cysts with thin (p < 0.05) or thick walls (p = 0.001), multiple cystic lesions, (p < 0.001), cystic lesions draining to the skin (p < 0.001), and skin fistulas (p < 0.05) were detected more often in IGM. Central (p < 0.05) and periareolar (p < 0.001) location and focal skin thickening (p < 0.05) were significantly more common in IGM. Architectural distortion (p = 0.001) and diffuse skin thickening (p < 0.05) were associated with BC. Multiple regional distribution was more common in IGM, whereas diffuse distribution and clumped enhancement were more common in BC (p < 0.05). In kinetic analysis, persistent enhancement was more common in IGM, whereas plateau and wash-out types were more common in BC (p < 0.001). Independent predictors for BC were age, diffuse skin thickening and kinetic curve types. There was no significant difference in the diffusion characteristics. Based on these findings, MRI had a sensitivity, specificity and accuracy of 88%, 67.65%, and 78.32%, respectively, in differentiating IGM from BC.

CONCLUSIONS

In conclusion, for non-mass enhancement, MRI can rule out malignancy with a considerably high sensitivity; however, specificity is still low, as many IGM patients have overlapping findings. Final diagnosis should be complemented with histopathology whenever necessary.

Differentiation Between Granulomatous Lobular Mastitis and Breast Cancer Using Quantitative Parameters on Contrast-Enhanced Ultrasound

Objective

To investigate the Contrast-enhanced ultrasound (CEUS) imaging characteristics of granulomatous lobular mastitis (GLM) and the value of differentiating GLM from breast cancer.

Materials and methods

The study included 30 women with GLM (mean age 36.7 ± 5 years [SD]) and 58 women with breast cancer (mean age 48. ± 8 years [SD]) who were scheduled for ultrasound-guided tissue biopsy. All patients were evaluated with conventional US and CEUS prior to the biopsy. In both groups, the parameters of the quantitative and qualitative analysis of the CEUS were recorded and compared. The receiver-operating-characteristics curves (ROC) were created. Sensitivity, specificity, cut-off, and area under the curve (AUC) values were calculated.

Results

TTP values in GLM were statistically higher than in breast cancer (mean, 27.63 ± 7.29 vs. 20.10 ± 6.11), but WIS values were lower (mean, 0.16 ± 0.05 vs. 0.28 ± 0.17). Rich vascularity was discovered in 54.45% of breast cancer patients, but only 30.00% of GLM patients had rich vascularity. The AUC for the ROC test was 0.791 and 0.807, respectively. The optimal cut-off value for TTP was 24.5s, and the WIS cut-off value was 0.185dB/s, yielding 73.33% sensitivity, 84.48% specificity, and 86.21% sensitivity, 70% specificity respectively in the diagnosis of GLM. The lesion scores reduced from 4 to 3 with the addition of CEUS for the patients with GLM. However, the scores did not change for the patients with breast cancer.

Conclusion

CEUS could help distinguish GLM from breast cancer by detecting higher TTP and WIS values, potentially influencing clinical decision-making for additional biopsies.

Management of granulomatous lobular mastitis: an international multidisciplinary consensus (2021 edition)

Granulomatous lobular mastitis (GLM) is a rare and chronic benign inflammatory disease of the breast. Difficulties exist in the management of GLM for many front-line surgeons and medical specialists who care for patients with inflammatory disorders of the breast. This consensus is summarized to establish evidence-based recommendations for the management of GLM. Literature was reviewed using PubMed from January 1, 1971 to July 31, 2020. Sixty-six international experienced multidisciplinary experts from 11 countries or regions were invited to review the evidence. Levels of evidence were determined using the American College of Physicians grading system, and recommendations were discussed until consensus. Experts discussed and concluded 30 recommendations on historical definitions, etiology and predisposing factors, diagnosis criteria, treatment, clinical stages, relapse and recurrence of GLM. GLM was recommended as a widely accepted definition. In addition, this consensus introduced a new clinical stages and management algorithm for GLM to provide individual treatment strategies. In conclusion, diagnosis of GLM depends on a combination of history, clinical manifestations, imaging examinations, laboratory examinations and pathology. The approach to treatment of GLM should be applied according to the different clinical stage of GLM. This evidence-based consensus would be valuable to assist front-line surgeons and medical specialists in the optimal management of GLM.

The Value of Whole-Tumor Histogram and Texture Analysis Using Intravoxel Incoherent Motion in Differentiating Pathologic Subtypes of Locally Advanced Gastric Cancer

Purpose

To determine if whole-tumor histogram and texture analyses using intravoxel incoherent motion (IVIM) parameters values could differentiate the pathologic characteristics of locally advanced gastric cancer.

Methods

Eighty patients with histologically confirmed locally advanced gastric cancer who received surgery in our institution were retrospectively enrolled into our study between April 2017 and December 2018. Patients were excluded if they had lesions with the smallest diameter < 5 mm and severe image artifacts. MR scanning included IVIM sequences (9 b values, 0, 20, 40, 60, 100, 150,200, 500, and 800 s/mm²) used in all patients before treatment. Whole tumors were segmented by manually drawing the lesion contours on each slice of the diffusion-weighted imaging (DWI) images (with b=800). Histogram and texture metrics for IVIM parameters values and apparent diffusion coefficient (ADC) values were measured based on whole-tumor volume analyses. Then, all 24 extracted metrics were compared between well, moderately, and poorly differentiated tumors, and between different Lauren classifications, signet-ring cell carcinomas, and other poorly cohesive carcinomas using univariate analyses. Multivariate logistic analyses and multicollinear tests were used to identify independent influencing factors from the significant variables of the univariate analyses to distinguish tumor differentiation and Lauren classifications. ROC curve analyses were performed to evaluate the diagnostic performance of these independent influencing factors for determining tumor differentiation and Lauren classifications and identifying signet-ring cell carcinomas. The interobserver agreement was also conducted between the two observers for image quality evaluations and parameter metric measurements.

Results

For diagnosing tumor differentiation, the ADC_median, pure diffusion coefficient median (Dslow_median), and pure diffusion coefficient entropy (Dslow_entropy) showed the greatest AUCs: 0.937, 0.948, and 0.850, respectively, and no differences were found between the three metrics, P>0.05). The 95th percentile perfusion factor (FP _P95th) was the best metric to distinguish diffuse-type GCs vs. intestinal/mixed (AUC=0.896). The ROC curve to distinguish signet-ring cell carcinomas from other poorly cohesive carcinomas showed that the Dslow_median had AUC of 0.738. For interobserver reliability, image quality evaluations showed excellent agreement (interclass correlation coefficient [ICC]=0.85); metrics measurements of all parameters indicated good to excellent agreement (ICC=0.65-0.89), except for the Dfast metric, which showed moderate agreement (ICC=0.41-0.60).

Conclusions

The whole-tumor histogram and texture analyses of the IVIM parameters based on the biexponential model provided a non-invasive method to discriminate pathologic tumor subtypes preoperatively in patients with locally advanced gastric cancer. The metric FP _P95th derived from IVIM performed better in determining Lauren classifications than the mono-exponential model.

Non-Mass Enhancements on DCE-MRI: Development and Validation of a Radiomics-Based Signature for Breast Cancer Diagnoses

Purpose

We aimed to assess the additional value of a radiomics-based signature for distinguishing between benign and malignant non-mass enhancement lesions (NMEs) on dynamic contrast-enhanced breast magnetic resonance imaging (breast DCE-MRI).

Methods

In this retrospective study, 232 patients with 247 histopathologically confirmed NMEs (malignant: 191; benign: 56) were enrolled from December 2017 to October 2020 as a primary cohort to develop the discriminative models. Radiomic features were extracted from one post-contrast phase (around 90s after contrast injection) of breast DCE-MRI images. The least absolute shrinkage and selection operator (LASSO) regression model was adapted to select features and construct the radiomics-based signature. Based on clinical and routine MR features, radiomics features, and combined information, three discriminative models were built using multivariable logistic regression analyses. In addition, an independent cohort of 72 patients with 72 NMEs (malignant: 50; benign: 22) was collected from November 2020 to April 2021 for the validation of the three discriminative models. Finally, the combined model was assessed using nomogram and decision curve analyses.

Results

The routine MR model with two selected features of the time-intensity curve (TIC) type and MR-reported axillary lymph node (ALN) status showed a high sensitivity of 0.942 (95%CI, 0.906 - 0.974) and low specificity of 0.589 (95%CI, 0.464 - 0.714). The radiomics model with six selected features was significantly correlated with malignancy (P<0.001 for both primary and validation cohorts). Finally, the individual combined model, which contained factors including TIC types and radiomics signatures, showed good discrimination, with an acceptable sensitivity of 0.869 (95%CI, 0.816 to 0.916), improved specificity of 0.839 (95%CI, 0.750 to 0.929). The nomogram was applied to the validation cohort, reaching good discrimination, with a sensitivity of 0.820 (95%CI, 0.700 to 0.920), specificity of 0.864 (95%CI,0.682 to 1.000). The combined model was clinically helpful, as demonstrated by decision curve analysis.

Conclusions

Our study added radiomics signatures into a conventional clinical model and developed a radiomics nomogram including radiomics signatures and TIC types. This radiomics model could be used to differentiate benign from malignant NMEs in patients with suspicious lesions on breast MRI.

Combining volumetric apparent diffusion coefficient histogram analysis with vesical imaging reporting and data system to predict the muscle invasion of bladder cancer

OBJECTIVE

The objective of this study was to explore whether volumetric apparent diffusion coefficient (ADC) histogram analysis can provide additional value to Vesical Imaging Reporting and Data System (VI-RADS) in differentiating muscle-invasive bladder cancer (MIBC) from non-muscle-invasive bladder cancer (NMIBC).

MATERIALS AND METHODS

80 patients were retrospectively reviewed with pathologically proven NMIBC (n = 53) or MIBC (n = 27). All patients underwent MRI including diffusion-weighted imaging (DWI) (b = 0, 800 s/mm²), and the VI-RADS score was evaluated based on DWI. Volumetric ADC histogram parameters were calculated from the volumetric of interest (VOI) on DWI, including the min ADC, mean ADC, median ADC, max ADC, 10th, 25th, 75th, 90th percentiles ADC, skewness, kurtosis, and entropy. The Mann-Whitney U-test was used to compare histogram parameters between NMIBC and MIBC. Receiver operating characteristic analysis was used to evaluate the diagnostic value of each significant parameter.

RESULTS

Among all parameters, the VI-RADS yield the highest Area Under the Curve (AUC, 0.88; sensitivity, 88.89%; specificity, 83.61%). MIBC had significantly lower min ADC, mean ADC, median ADC, 10th, 25th, 75th, and 90th percentiles ADC than NMIBC (p = 0.002, p < 0.001, p < 0.001, p = 0.003, p = 0.004, p < 0.001, p < 0.001). Skewness and kurtosis of MIBC were significantly higher than those of NMIBC (p < 0.001, p < 0.001). The combination of VI-RADS and skewness showed significantly higher AUC (AUC 0.923; 95% CI 0.847-0.969) than only with VI-RADS (AUC 0.880; 95% CI 0.793-0.940).

CONCLUSION

Volumetric ADC histogram analysis and VI-RADS are both useful methods in differentiating MIBC from NMIBC, and the volumetric ADC histogram analysis can provide additional value to VI-RADS.

Performances of Whole Tumor Texture Analysis Based on MRI: Predicting Preoperative T Stage of Rectal Carcinomas

Objective

To determine whether there is a correlation between texture features extracted from high-resolution T2-weighted imaging (HR-T2WI) or apparent diffusion coefficient (ADC) maps and the preoperative T stage (stages T1–2 versus T3–4) in rectal carcinomas.

Materials and Methods

One hundred and fifty four patients with rectal carcinomas who underwent preoperative HR-T2WI and diffusion-weighted imaging were enrolled. Patients were divided into training (n = 89) and validation (n = 65) cohorts. 3D Slicer was used to segment the entire volume of interest for whole tumors based on HR-T2WI and ADC maps. The least absolute shrinkage and selection operator (LASSO) was performed to select feature. The significantly difference was tested by the independent sample t-test and Mann-Whitney U test. The support vector machine (SVM) model was used to develop classification models. The correlation between features and T stage was assessed by Spearman’s correlation analysis. Multivariate logistic regression analysis was performed to identify independent predictors of tumor invasion. The performance of classifiers was evaluated by the receiver operating characteristic (ROC) curves.

Results

The wavelet HHH NGTDM strength (R_S = -0.364, P < 0.001) from HR-T2WI was an independent predictor of stage T3–4 tumors. The shape maximum 2D diameter column (R_S = 0.431, P < 0.001), log σ = 5.0 mm 3D first-order maximum (R_S = 0.276, P = 0.009), and log σ = 5.0 mm 3D first-order interquartile range (R_S = -0.229, P = 0.032) from ADC maps were independent predictors. In training cohorts, the classification models from HR-T2WI, ADC maps and the combination of two achieved the area under the ROC curves (AUCs) of 0.877, 0.902 and 0.941, with the accuracy of 79.78%, 89.86% and 89.89%, respectively. In validation cohorts, the three models achieved AUCs of 0.845, 0.881 and 0.910, with the accuracy of 78.46%, 83.08% and 87.69%, respectively.

Conclusions

Texture analysis based on ADC maps shows more potential than HR-T2WI in identifying preoperative T stage in rectal carcinomas. The combined application of HR-T2WI and ADC maps may help to improve the accuracy of preoperative diagnosis of rectal cancer invasion.

Risk Stratification of Ductal Carcinoma In Situ and Texture Analysis of Contrast-Enhanced Breast Magnetic Resonance Imaging

OBJECTIVE

The aim of this study was to investigate whether texture analysis of contrast-enhanced T1 weighted images could predict risk of ductal carcinoma in situ (DCIS).

METHODS

The study included 185 DCIS lesions that were classified as either low risk or non-low risk using surgical pathology records. All magnetic resonance imaging texture analyses were performed using postprocessing software, and texture-derived parameters were extracted.

RESULTS

The sphericity, compactness, and spherical disproportion were significantly different in the low-risk and non-low risk groups using the Van Nuys Prognostic Index (mean ± SD, 0.479 ± 0.189 vs 0.414 ± 0.176, 0.161 ± 0.159 vs 0.112 ± 0.134, and 2.569 ± 1.434 vs 2.934 ± 1.374, respectively; P < 0.05). In the univariate analyses, sphericity (odds ratio, 7.091; 95% confidence interval, 1.236-40.666; P = 0.028) and compactness (odds ratio, 9.267; 95% confidence interval, 1.125-76.360; P = 0.039) were significantly associated with a high probability of being low risk according to the Van Nuys Prognostic Index.

CONCLUSIONS

Whole-lesion texture analysis may be helpful in identifying patients classified as having low-risk DCIS before surgery.

Granuloma after breast conserving surgery-a report of three cases

Granulomatous mastitis is a rare breast disease that is categorized as a benign tumor with chronic inflammation. Since the cause of the chronic inflammation is usually unknown, it is sometimes called idiopathic granulomatous mastitis (IGM). Although imaging modalities, such as ultrasound, magnetic resonance imaging and mammography can detect tumors, they are sometimes unable to differentiate between benign and malignant tumors. In such cases, biopsy is needed to make a correct diagnosis. We experienced three cases of IGM after breast conserving surgery in breast cancer patients in whom we needed to rule out recurrence of breast cancer. In our cases, tumorectomy was performed in two cases for pathological diagnosis, since neither biopsy nor cytology was able to reveal a conclusive pathological diagnosis. Our management of these three cases might suggest the appropriate management of granulomatous tumors after breast conserving surgery in breast cancer survivors.

Is There any Relationship Between Granulomatous Mastitis and Seasons? An Analysis of Seasonal Frequency, Clinical, and Radiologic Findings

Objective

Idiopathic granulomatous mastitis (IGM) is a rare, resistant, and recurrent benign disease of the breast. IGM can be clinically and radiologically confused with breast carcinoma, and core needle biopsy is needed to diagnose. The etiology and pathogenesis of IGM have not been fully explained. This premenopausal disease may be associated with pregnancy, breastfeeding, autoimmune processes, inflammation, and oral contraceptives. However, there is no study on whether there is a seasonal relationship.

Materials and Methods

From January 2015 to January 2020, the seasonal relationship of IGM was evaluated in 37 females aged between 25-49.

Results

Although all cases were distributed between September and May, there was no statistically significant result in the relationship with the season. US is the main modality in the diagnosis of this condition which only provides an accurate pre-diagnosis approach with the typical USG appearance features. Some MRI features may help us to distinguish IGM from breast malignities.

Conclusion

IGM is a rare chronic non-specific inflammatory lesion of the breast, which can be confused with benign and malignant breast diseases in both clinical and radiologic aspects. To understand the etiology of this condition better, the seasonal connection needs to be evaluated in larger patient groups.

The value of multiparametric histogram features based on intravoxel incoherent motion diffusion-weighted imaging (IVIM-DWI) for the differential diagnosis of liver lesions

Background

The present study analyzed whole-lesion histogram parameters from intravoxel incoherent motion diffusion-weighted imaging (IVIM-DWI) to explore the clinical value of IVIM histogram features in the differentiation of liver lesions.

Methods

In this retrospective study, 33 cases of hepatic hemangioma (HH), 22 cases of hepatic cysts (HC), and 34 cases of hepatocellular carcinoma (HCC) were underwent IVIM-DWI (b =0–600 s/mm²), which were confirmed pathologically and clinically. The data were processed by IVIM model to obtain the following quantitative indicators: perfusion fraction (f), slow diffusion coefficient (D), and pseudo-diffusion coefficient (or fast diffusion coefficient, D*). The region of interest in the largest solid part of the lesion was delineated for histogram analysis of the correlation between tissue image and lesion type. The relevant histogram parameters were obtained and statistically analyzed. The characteristic histogram parameters for HH, HC, and HCC were compared to find significantly different parameters. The diagnostic efficacies of these parameters for HH, liver cysts, and HCC were assessed using the receiver operating characteristic (ROC) curves.

Results

There were significant differences in the maximum diameter, maximum value, minimum value, mean, median, standard deviation, uniformity, skewness, kurtosis, volume, 10th percentile (P10) of D, and 90th percentile (P90) of D between the three groups (P<0.05). The maximum diameter, minimum value, entropy, and volume of D* differed significantly between the three groups (P<0.05). The maximum diameter, minimum value, mean, median, skewness, kurtosis, volume, P10, and P90 of f differed significantly between the three groups (P<0.05). The largest area under the ROC curve (AUC) for both D* and f was that of volume (AUC =0.883 for both). When 1438.802 was used as the volume cut-off, the sensitivity and specificity of volume in differentiating between HH and HC were 87.88 and 77.27, respectively, and the sensitivity and specificity of volume in differentiating between HC and HCC were 77.27 and 85.29.

Conclusions

A multiparametric histogram from IVIM-DWI magnetic resonance imaging (MRI) is an effective means of identifying HH, HC, and HCC that provides valuable reference information for clinical diagnosis.