ObjectivesTo assess the feasibility of the mono-exponential, bi-exponential and stretched-exponential models in evaluating response of breast tumours to neoadjuvant chemotherapy (NACT) at 3 T.MethodsThirty-six female patients (median age 53, range 32–75 years) with invasive breast cancer undergoing NACT were enrolled for diffusion-weighted MRI (DW-MRI) prior to the start of treatment. For assessment of early response, changes in parameters were evaluated on mid-treatment MRI in 22 patients. DW-MRI was performed using eight b values (0, 30, 60, 90, 120, 300, 600, 900 s/mm2). Apparent diffusion coefficient (ADC), tissue diffusion coefficient (D
t), vascular fraction (ƒ), distributed diffusion coefficient (DDC) and alpha (α) parameters were derived. Then t tests compared the baseline and changes in parameters between response groups. Repeatability was assessed at inter- and intraobserver levels.ResultsAll patients underwent baseline MRI whereas 22 lesions were available at mid-treatment. At pretreatment, mean diffusion coefficients demonstrated significant differences between groups (p < 0.05). At mid-treatment, percentage increase in ADC and DDC showed significant differences between responders (49 % and 43 %) and non-responders (21 % and 32 %) (p = 0.03, p = 0.04). Overall, stretched-exponential parameters showed excellent repeatability.ConclusionDW-MRI is sensitive to baseline and early treatment changes in breast cancer using non-mono-exponential models, and the stretched-exponential model can potentially monitor such changes.Key points• Baseline diffusion coefficients demonstrated significant differences between complete pathological responders and non-responders.
• Increase in ADC and DDC at mid-treatment can discriminate responders and non-responders.
• The ƒ fraction at mid-treatment decreased in responders whereas increased in non-responders.
• The mono- and stretched-exponential models showed excellent inter- and intrarater repeatability.
• Treatment effects can potentially be assessed by non-mono-exponential diffusion models.
Electronic supplementary materialThe online version of this article (doi:10.1007/s00330-016-4630-x) contains supplementary material, which is available to authorized users.
Alterations in the geometry and function of the heart define well-established causes of cardiovascular disease. However, current approaches to the diagnosis of cardiovascular diseases often rely on subjective human assessment as well as manual analysis of medical images. Both factors limit the sensitivity in quantifying complex structural and functional phenotypes. Deep learning approaches have recently achieved success for tasks such as classification or segmentation of medical images, but lack interpretability in the feature extraction and decision processes, limiting their value in clinical diagnosis. In this work, we propose a 3D convolutional generative model for automatic classification of images from patients with cardiac diseases associated with structural remodeling. The model leverages interpretable task-specific anatomic patterns learned from 3D segmentations. It further allows to visualise and quantify the learned pathology-specific remodeling patterns in the original input space of the images. This approach yields high accuracy in the categorization of healthy and hypertrophic cardiomyopathy subjects when tested on unseen MR images from our own multi-centre dataset (100%) as well on the ACDC MICCAI 2017 dataset (90%). We believe that the proposed deep learning approach is a promising step towards the development of interpretable classifiers for the medical imaging domain, which may help clinicians to improve diagnostic accuracy and enhance patient risk-stratification.
B1(+) correction demonstrates a substantial effect on the results of quantitative dynamic contrast-enhanced analysis of breast tissue at 3 T, which propagates into the pharmacokinetic analysis of tumors that is dependent on whether the tumor is located in the right or left breast.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.