Breast CEST-Dixon imaging shows potential to differentiate more aggressive from less aggressive cancers. Magn Reson Med 80:895-903, 2018. © 2018 International Society for Magnetic Resonance in Medicine.
Dose was escalated to the target dose of 40 Gy in 5 fractions, with the occurrence of only 1 dose-limiting toxicity. Patients felt cosmetic results improved within the first year after surgery and stereotactic body radiation therapy. Our results show minimal toxicity with excellent cosmesis; however, further follow-up is warranted in future studies. This study is the first to show the safety, tolerability, feasibility, and cosmesis results of a 5-fraction dose-escalated S-PBI treatment for early-stage breast cancer in the adjuvant setting.
Iodine 125 ((125)I) radioactive seed localization has emerged as a reliable and safe alternative to wire localization for guidance during the surgical resection of nonpalpable breast lesions. The breast imager has a responsibility to be familiar with the general principles of this evolving technique, including its advantages and disadvantages as well as the technical differences involved in placement of seeds versus traditional wire localization. Although placement of (125)I seeds is conceptually similar to wire placement, there are additional technical considerations and safety measures that need to be addressed and implemented when radioactive seeds are used. We draw from our experience with more than 1000 cases of radioactive seed localization since inception of our program in 2009 to provide illustrative examples of not only the proper technique of radioactive seed localization, but also mishaps that may occur during this procedure, along with practical suggestions to prevent these problems. We examine some of the difficulties that we have encountered during radioactive seed localization at our institution, including bone wax mimicking the seed, the inadvertent deployment of seeds, the need for multiple seeds or supplemental wires, problematic seed locations, and difficulty in surgical retrieval of the seed. Recognizing the potential pitfalls of radioactive seed localization and understanding the appropriate guidelines and precautions for the safe, secure handling and placement of radioactive seeds is essential for a successful radioactive seed localization program.
Breast implants are frequently encountered on breast imaging studies, and it is essential for any radiologist interpreting these studies to be able to correctly assess implant integrity. Ruptures of silicone gel-filled implants often occur without becoming clinically obvious and are incidentally detected at imaging. Early diagnosis of implant rupture is important because surgical removal of extracapsular silicone in the breast parenchyma and lymphatics is difficult. Conversely, misdiagnosis of rupture may prompt a patient to undergo unnecessary additional surgery to remove the implant. Mammography is the most common breast imaging examination performed and can readily depict extracapsular free silicone, although it is insensitive for detection of intracapsular implant rupture. Ultrasonography (US) can be used to assess the internal structure of the implant and may provide an economical method for initial implant assessment. Common US signs of intracapsular rupture include the "keyhole" or "noose" sign, subcapsular line sign, and "stepladder" sign; extracapsular silicone has a distinctive "snowstorm" or echogenic noise appearance. Magnetic resonance (MR) imaging is the most accurate and reliable means for assessment of implant rupture and is highly sensitive for detection of both intracapsular and extracapsular rupture. MR imaging findings of intracapsular rupture include the keyhole or noose sign, subcapsular line sign, and "linguine" sign, and silicone-selective MR imaging sequences are highly sensitive to small amounts of extracapsular silicone. RSNA, 2017.
We propose a novel BIRADS-SSDL network that integrates clinically-approved breast lesion characteristics (BIRADS features) into task-oriented semi-supervised deep learning (SSDL) for accurate diagnosis of ultrasound (US) images with a small training dataset. Breast US images are converted to BIRADS-oriented feature maps (BFMs) using a distance-transformation coupled with a Gaussian filter. Then, the converted BFMs are used as the input of an SSDL network, which performs unsupervised stacked convolutional auto-encoder (SCAE) image reconstruction guided by lesion classification. This integrated multi-task learning allows SCAE to extract image features with the constraints from the lesion classification task, while the lesion classification is achieved by utilizing the SCAE encoder features with a convolutional network. We trained the BIRADS-SSDL network with an alternative learning strategy by balancing the reconstruction error and classification label prediction error. To demonstrate the effectiveness of our approach, we evaluated it using two breast US image datasets. We compared the performance of the BIRADS-SSDL network with conventional SCAE and SSDL methods that use the original images as inputs, as well as with an SCAE that use BFMs as inputs. The experimental results on two breast US datasets show that BIRADS-SSDL ranked the best among the four networks, with a classification accuracy of around 94.23 ± 3.33% and 84.38 ± 3.11% on two datasets. In the case of experiments across two datasets collected from two different institutions/and US devices, the developed BIRADS-SSDL is generalizable across the different US devices and institutions without overfitting to a single dataset and achieved satisfactory results. Furthermore, we investigate the performance of the proposed method by varying the model training strategies, lesion boundary accuracy, and Gaussian filter parameters. The experimental results showed that a pre-training strategy can help to speed up model convergence during training but with no improvement of the classification accuracy on the testing dataset. The classification accuracy decreases as the segmentation accuracy decreases. The proposed BIRADS-SSDL achieves the best results among the compared methods in each case and has the capacity to deal with multiple different datasets under one model. Compared with state-of-the-art methods, BIRADS-SSDL could be promising for effective breast US computer-aided diagnosis using small datasets.
Purpose To enable high spatial and temporal breast imaging resolution via combined use of high field MRI, array coils, and forced current excitation (FCE) multi channel transmit. Materials and Methods A unilateral 16-channel receive array insert was designed for use in a transmit volume coil optimized for quadrature operation with dual-transmit RF shimming at 7T. Signal-to-noise ratio (SNR) maps, g-factor maps, and high spatial and temporal resolution in vivo images were acquired to demonstrate the utility of the coil architecture. Results The dual-transmit FCE coil provided homogeneous excitation and the array provided an increase in average SNR of 3.3 times (max 10.8, min 1.5) compared to the volume coil in transmit/receive mode. High resolution accelerated in vivo breast imaging demonstrated the ability to achieve isotropic spatial resolution of 0.5 mm within clinically relevant 90 s scan times, as well as the ability to perform 1.0 mm isotropic resolution imaging, 7 s per dynamics, with the use of bidirectional SENSE acceleration of up to R = 9. Conclusion The FCE design of the transmit coil easily accommodates the addition of a sixteen channel array coil. The improved spatial and temporal resolution provided by the high-field array coil with FCE dual-channel transmit will ultimately be beneficial in lesion detection and characterization.
- Magnetic resonance imaging detects malignancies undetected by other imaging modalities but also detects a wide variety of benign lesions. Benign and malignant lesions identified by MRI share similar morphologic and kinetic features, necessitating biopsy for histologic confirmation.
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.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.