Dataset Information

Kinetic Curve Type Assessment for Classification of Breast Lesions Using Dynamic Contrast-Enhanced MR Imaging.

ABSTRACT:

Objective

The aim of this study was to employ a kinetic model with dynamic contrast enhancement-magnetic resonance imaging to develop an approach that can efficiently distinguish malignant from benign lesions.

Materials and methods

A total of 43 patients with 46 lesions who underwent breast dynamic contrast enhancement-magnetic resonance imaging were included in this retrospective study. The distribution of malignant to benign lesions was 31/15 based on histological results. This study integrated a single-compartment kinetic model and dynamic contrast enhancement-magnetic resonance imaging to generate a kinetic modeling curve for improving the accuracy of diagnosis of breast lesions. Kinetic modeling curves of all different lesions were analyzed by three experienced radiologists and classified into one of three given types. Receiver operating characteristic and Kappa statistics were used for the qualitative method. The findings of the three radiologists based on the time-signal intensity curve and the kinetic curve were compared.

Results

An average sensitivity of 82%, a specificity of 65%, an area under the receiver operating characteristic curve of 0.76, and a positive predictive value of 82% and negative predictive value of 63% was shown with the kinetic model (p = 0.017, 0.052, 0.068), as compared to an average sensitivity of 80%, a specificity of 55%, an area under the receiver operating characteristic of 0.69, and a positive predictive value of 79% and negative predictive value of 57% with the time-signal intensity curve method (p = 0.003, 0.004, 0.008). The diagnostic consistency of the three radiologists was shown by the κ-value, 0.857 (p<0.001) with the method based on the time-signal intensity curve and 0.826 (p<0.001) with the method of the kinetic model.

Conclusions

According to the statistic results based on the 46 lesions, the kinetic modeling curve method showed higher sensitivity, specificity, positive and negative predictive values as compared with the time-signal intensity curve method in lesion classification.

SUBMITTER: Yang SN

PROVIDER: S-EPMC4824432 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:PurposeTo assess the accuracy of the shutter-speed approach compared with standard approach dynamic contrast material-enhanced magnetic resonance (MR) imaging pharmacokinetic analysis for breast cancer diagnosis.Materials and methodsThis study was approved by the institutional review board and was HIPAA compliant. Informed consent was obtained from 89 high-risk women (age range, 28-83 years) who had 92 suspicious lesions with negative findings at mammography (but visible at MR imaging). Each underwent a research dynamic contrast-enhanced MR imaging examination just prior to a clinical MR imaging-guided interventional procedure. Tumor region of interest (ROI) averaged and (for some) pixel-by-pixel dynamic contrast-enhanced time-course data, together with mean arterial input function, were subjected to serial standard and shutter-speed approach analyses to extract pharmacokinetic parameters, including rate constant for passive contrast reagent transfer between plasma and interstitium (K(trans)) and interstitial space volume fraction, or v(e). Pathologic findings were used as reference standards. Diagnostic accuracy was assessed with receiver operating characteristic analyses.ResultsThe pathologic analyses revealed 20 malignant and 72 benign lesions. Positive predictive value of the institutional clinical breast MR imaging protocol was 22%. At 100% sensitivity, ROI-averaged shutter-speed approach K(trans) had significantly (P = .008) higher diagnostic specificity than standard approach K(trans): 86.1% versus 77.8%. The difference in the ROI-averaged K(trans) parameter value, or ΔK(trans) (≡ K(trans) [shutter-speed approach] - K(trans) [standard approach]), had even higher specificity (88.9%). Combined use of ROI analysis and pixel-by-pixel mapping of ΔK(trans) achieved 98.6% specificity at 100% sensitivity.ConclusionThe use of the shutter-speed dynamic contrast-enhanced MR imaging method has the potential to improve breast cancer diagnostic accuracy and reduce putatively unnecessary biopsy procedures that yield benign pathologic findings.Supplemental materialhttp://radiology.rsna.org/lookup/suppl/doi:10.1148/radiol.11102413/-/DC1.

Project description:BackgroundThe purpose of this study was to develop a deep learning-based system with a cascade feature pyramid network for the detection and classification of breast lesions in dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI).MethodsThis retrospective study enrolled 191 consecutive patients with pathological confirmed breast lesions who underwent preoperative magnetic resonance imaging (MRI) at the Second Affiliated Hospital of Xi'an Jiaotong University. Patients were randomly divided into a training set comprising 153 patients with 126 malignant and 27 benign lesions and a validation set containing 38 patients with 31 malignant and 7 benign lesions under 5-fold cross-validation. Two radiologists annotated the location and classification of all lesions. After augmentation with pseudo-color image fusion, MRI images were fed into the developed cascade feature pyramid network system, feature pyramid network, and faster region-based convolutional neural network (CNN) for breast lesion detection and classification, respectively. The performance on large (>2 cm) and small (≤2 cm) breast lesions was further evaluated. Average precision (AP), mean AP, F1-score, sensitivity, and false positives were used to evaluate different systems. Cohen's kappa scores were calculated to assess agreement between different systems, and Student's t-test and the Holm-Bonferroni method were used to compare multiple groups.ResultsThe cascade feature pyramid network system outperformed the other systems with a mean AP and highest sensitivity of 0.826±0.051 and 0.970±0.014 (at 0.375 false positives), respectively. The F1-score of the cascade feature pyramid network in real detection was also superior to that of the other systems at both the slice and patient levels. The mean AP values of the cascade feature pyramid network reached 0.779±0.152 and 0.790±0.080 in detecting large and small lesions, respectively. Especially for small lesions, the cascade feature pyramid network achieved the best performance in detecting benign and malignant breast lesions at both the slice and patient levels.ConclusionsThe deep learning-based system with the developed cascade feature pyramid network has the potential to detect and classify breast lesions on DCE-MRI, especially small lesions.

Project description:To evaluate the effects of breast compression on breast cancer masses, contrast material enhancement of glandular tissue, and quality of magnetic resonance (MR) images in the identification and characterization of breast lesions.This was a HIPAA-compliant, institutional review board-approved retrospective study, with waiver of informed consent. Images from 300 MR imaging examinations in 149 women (mean age ± standard deviation, 51.5 years ± 10.9; age range, 22-76 years) were evaluated. The women underwent diagnostic MR imaging (no compression) and MR-guided biopsy (with compression) between June 2008 and February 2013. Breast compression was expressed as a percentage relative to the noncompressed breast. Percentage enhancement difference was calculated between noncompressed- and compressed-breast images obtained in early and delayed contrast-enhanced phases. Breast density, lesion type (mass vs non-masslike enhancement [NMLE]), lesion size, percentage compression, and kinetic curve type were evaluated. Linear regression, receiver operating characteristic (ROC) curve analysis, and ? test were performed.Mean percentage compression was 31.3% ± 9.2 (range, 5.8%-53.2%). Percentage enhancement was higher in noncompressed- versus compressed-breast studies in early (146% ± 66 vs 107% ± 42, respectively; P < .001) and delayed (158% ± 68 vs 107% ± 42, respectively; P = .1) phases. Among breast lesions, 12% (seven of 59) were significantly smaller when compressed, which led to underestimation of TNM classification (P < .001). Breast masses (n = 35) showed significantly higher early percentage enhancement (157% ± 71) than lesions with NMLE (n = 15, 120% ± 40; P = .02) and a percentage enhancement difference (47.5% ± 64 vs 17% ± 28, respectively; P = .023). Kinetic curve performance for identifying invasive cancer decreased after compression (area under ROC curve = 0.53 vs 0.71, respectively; P = .02). Breast compression resulted in complete loss of enhancement of nine of 210 lesions (4%).Breast compression during biopsy affected breast lesion detection, lesion size, and dynamic contrast-enhanced MR imaging interpretation and performance. Limiting the application of breast compression is recommended, except when clinically necessary.

Project description:We sought to determine the antiatherosclerotic properties of pioglitazone using multimethod noninvasive imaging techniques.Inflammation is an essential component of vulnerable or high-risk atheromas. Pioglitazone, a peroxisome proliferator-activated receptor-gamma agonist, possesses potent anti-inflammatory properties. We aimed to quantify noninvasively the anti-inflammatory effects of pioglitazone on atheroma using (18)F-fluorodeoxyglucose ((18)F-FDG) positron emission tomography (PET)/computed tomography (CT) and dynamic contrast-enhanced (DCE) magnetic resonance imaging (MRI).Atherosclerotic plaques were induced in the aorta of 15 New Zealand white rabbits by a combination of a hyperlipidemic diet and 2 balloon endothelial denudations. Nine rabbits continued the same diet, whereas 6 rabbits received pioglitazone (10 mg/kg orally) in addition to the diet. Twelve animals underwent (18)F-FDG-PET/CT, and 15 animals underwent DCE-MRI at baseline, 1 month, and 3 months after treatment initiation. Concomitantly, serum metabolic parameters were monitored. After imaging was completed, aortic histologic analysis and correlation analysis were performed.The (18)F-FDG-PET/CT imaging detected an increase in average standardized uptake value in the control group (p < 0.01), indicating progressive inflammation, whereas stable standardized uptake values were observed in the treatment group, indicating no progression. The DCE-MRI analysis detected a significant decrease in the area under the curve for the pioglitazone group (p < 0.01). Immunohistologic examination of the aortas demonstrated a significant decrease in macrophage and oxidized phospholipid immunoreactivity in the pioglitazone group (p = 0.04 and p = 0.01, respectively) with respect to control animals, underlining the imaging results. Serum metabolic parameters showed no difference between groups. Strong positive correlations between standardized uptake value and macrophage density and between area under the curve and neovessels were detected (r(2) = 0.86 and p < 0.0001, and r(2) = 0.66 and p = 0.004, respectively).Both (18)F-FDG-PET/CT and DCE-MRI demonstrate noninvasively the anti-inflammatory effects of pioglitazone on atheroma. Both imaging methods seem suited to monitor inflammation in atherosclerosis.

Project description:BACKGROUND:Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) has been shown to be a promising technique for assessing lung lesions. However, DCE-MRI often suffers from motion artifacts and insufficient imaging speed. Therefore, highly accelerated free-breathing DCE-MRI is of clinical interest for lung exams. PURPOSE:To test the performance of rapid free-breathing DCE-MRI for simultaneous qualitative and quantitative assessment of pulmonary lesions using Golden-angle RAdial Sparse Parallel (GRASP) imaging. STUDY TYPE:Prospective. POPULATION:Twenty-six patients (17 males, mean age?=?55.1?±?14.4) with known pulmonary lesions. FIELD STRENGTH/SEQUENCE:3T MR scanner; a prototype fat-saturated, T1 -weighted stack-of-stars golden-angle radial sequence for data acquisition and a Cartesian breath-hold volumetric-interpolated examination (BH-VIBE) sequence for comparison. ASSESSMENT:After a dual-mode GRASP reconstruction, one with 3-second temporal resolution (3s-GRASP) and the other with 15-second temporal resolution (15s-GRASP), all GRASP and BH-VIBE images were pooled together for blind assessment by two experienced radiologists, who independently scored the overall image quality, lesion delineation, overall artifact level, and diagnostic confidence of each case. Perfusion analysis was performed for the 3s-GRASP images using a Tofts model to generate the volume transfer coefficient (Ktrans ) and interstitial volume (Ve ). STATISTICAL TESTS:Nonparametric paired two-tailed Wilcoxon signed-rank test; Cohen's kappa; unpaired Student's t-test. RESULTS:15s-GRASP achieved comparable image quality with conventional BH-VIBE (P?>?0.05), except for the higher overall artifact level in the precontrast phase (P?=?0.018). The Ktrans and Ve in inflammation were higher than those in malignant lesions (Ktrans : 0.78?±?0.52?min-1 vs. 0.37?±?0.22?min-1 , P?=?0.020; Ve : 0.36?±?0.16 vs. 0.26?±?0.1, P?=?0.177). Also, the Ktrans and Ve in malignant lesions were also higher than those in benign lesions (Ktrans : 0.37?±?0.22?min-1 vs. 0.04?±?0.04?min-1 , P?=?0.001; Ve : 0.26?±?0.12 vs. 0.10?±?0.00, P?=?0.063). DATA CONCLUSION:This feasibility study demonstrated the performance of high spatiotemporal resolution free-breathing DCE-MRI of the lung using GRASP for qualitative and quantitative assessment of pulmonary lesions. LEVEL OF EVIDENCE:2 Technical Efficacy: Stage 1 J. MAGN. RESON. IMAGING 2018;48:459-468.

Project description:PurposeTo assess the performance of computer-extracted dynamic contrast material-enhanced (DCE) magnetic resonance (MR) imaging kinetic and morphologic features in the differentiation of invasive versus noninvasive breast lesions and metastatic versus nonmetastatic breast lesions.Materials and methodsIn this institutional review board-approved HIPAA-compliant study, in which the requirement for informed patient consent was waived, breast MR images were retrospectively collected. The images had been obtained with a 1.5-T MR unit by using a gadodiamide-enhanced T1-weighted spoiled gradient-recalled acquisition in the steady state sequence. The breast MR imaging database contained 132 benign, 71 ductal carcinoma in situ (DCIS), and 150 invasive ductal carcinoma (IDC) lesions. Fifty-four IDC lesions were associated with metastasis-positive lymph nodes (LNs), and 64 IDC lesions were associated with negative LNs. Lesion segmentation and extraction of morphologic and kinetic features were automatically performed by a laboratory-developed computer workstation. Features were first selected by using stepwise linear discriminant analysis and then merged by using Bayesian neural networks. Lesion classification performance was assessed with receiver operating characteristic analysis.ResultsDifferentiation of DCIS from IDC lesions yielded an area under the receiver operating characteristic curve (AUC) of 0.83 +/- 0.03 (standard error). AUCs were 0.85 +/- 0.02 for differentiation between IDC and benign lesions and 0.79 +/- 0.03 for differentiation between DCIS and benign lesions. Differentiation between IDC lesions associated with positive LNs and IDC lesions associated with negative LNs yielded an AUC of 0.82 +/- 0.04. AUCs were 0.86 +/- 0.03 for differentiation between IDC lesions associated with positive LNs and benign lesions and 0.83 +/- 0.03 for differentiation between IDC lesions associated with negative LNs and benign lesions.ConclusionComputer-aided diagnosis of breast DCE MR imaging-depicted lesions was extended from the task of discriminating between malignant and benign lesions to the prognostic tasks of distinguishing between noninvasive and invasive lesions and discriminating between metastatic and nonmetastatic lesions, yielding MR imaging-based prognostic markers.Supplemental materialhttp://radiology.rsna.org/lookup/suppl/doi:10.1148/radiol.09090838/-/DC1.

Project description:Background and purposeAlthough resection of a tumor by trans-sphenoidal surgery is considered the criterion standard for successful surgical treatment of functional pituitary microadenoma, MR imaging occasionally fails to visualize and identify the tumor and supplementary imaging modalities are necessary. We tested the possibility of dynamic contrast-enhanced multisection CT of the pituitary gland accompanying image reconstruction of contrast agent dynamics to identify the localizations of microadenomas and compared the diagnostic performance with conventional pituitary-targeted MR imaging.Materials and methodsTwenty-eight patients with surgically confirmed functional pituitary microadenomas (including growth hormone-, adrenocorticotropic hormone-, and prolactin-secreting adenomas) who underwent pituitary-targeted dynamic contrast-enhanced multisection CT were retrospectively investigated. We undertook image reconstruction of the dynamics of the contrast agent around the pituitary gland in a voxelwise manner, visualizing any abnormality and enabling qualification of contrast dynamics within the tumor.ResultsFifteen cases were correctly diagnosed by MR imaging, while dynamic contrast-enhanced multisection CT correctly diagnosed 26 cases. The accuracy of localization was markedly better for adrenocorticotropic hormone-secreting microadenomas, increasing from 32% on MR imaging to 85% by dynamic contrast-enhanced multisection CT. Compared with the normal pituitary gland, adrenocorticotropic hormone-secreting adenoma showed the least difference in contrast enhancement of the different functional microadenomas. Images acquired at 45-60 seconds after contrast agent injection showed the largest difference in contrast enhancement between an adenoma and the normal pituitary gland.ConclusionsDynamic contrast-enhanced multisection CT combined with image reconstruction of the contrast-enhanced dynamics holds promise in detecting MR imaging-occult pituitary microadenomas.

Project description:BackgroundThe apparent diffusion coefficient (ADC) is a highly diagnostic factor in discriminating malignant and benign breast masses in diffusion-weighted magnetic resonance imaging (DW-MRI). The combination of ADC and other pictorial characteristics has improved lesion type identification accuracy. The objective of this study was to reassess the findings on an independent patient group by changing the magnetic field from 1.5-Tesla to 3.0-Tesla.MethodsThis retrospective study consisted of a training group of 234 female patients, including 85 benign and 149 malignant lesions, imaged using 1.5-Tesla MRI, and a test group of 95 female patients, including 19 benign and 85 malignant lesions, imaged using 3.0-Tesla MRI. The lesion of interest was segmented from the raw image and four sets of measurements describing the morphology, kinetics, DW-MRI, and texture of the pictorial properties of each lesion were obtained. Each lesion was characterized by 28 features in total. Three classical machine-learning algorithms were used to build prediction models on the training group, which evaluated the prognostic performance of the multi-sided features in three scenarios. To reduce information redundancy, five highly diagnostic factors were selected to obtain a compact yet informative characterization of the lesion status.ResultsThree classification models were built on the training of 1.5-Tesla patients and were tested on the independent 3.0-Tesla test group. The following results were found. i) Characterization of breast masses in a multi-sided way dramatically increased prediction performance. The usage of all features gave a higher performance in both sensitivity and specificity than any individual feature groups or their combinations. ii) ADC was a highly effective factor in improving the sensitivity in discriminating malignant from benign masses. iii) Five features, namely ADC, Sum Average, Entropy, Elongation, and Sum Variance, were selected to achieve the highest performance in diagnosis of the 3.0-Tesla patient group.ConclusionsThe combination of ADC and other multi-sided characteristics can increase the capability of discriminating malignant and benign breast lesions, even under different imaging protocols. The selected compact feature subsets achieved a high diagnostic performance and thus are promising in clinical applications for discriminating lesion type and for personalized treatment planning.