← Back

/

Radiology, Nuclear Medicine and Imaging

Medical imaging and image-guided procedures, including diagnostic radiology, interventional radiology, nuclear medicine, radiation therapy planning, and advances in imaging technology

4 papers in this specialization

Paperzilla Papers in Radiology, Nuclear Medicine and Imaging

Low-Risk Differentiated Thyroid Cancer and Radioiodine Remnant Ablation: A Systematic Review of the Literature

A review of studies found no evidence that radioactive iodine treatment after surgery benefits low-risk differentiated thyroid cancer patients. The results for intermediate-risk patients are mixed, with some studies suggesting a benefit and others showing no effect. This highlights the need for better quality evidence, such as from randomized controlled trials, to guide decisions about radioactive iodine treatment in these patients.

UNETR: Transformers for 3D Medical Image Segmentation

This paper introduces UNETR, a novel transformer-based architecture for 3D medical image segmentation. UNETR leverages the power of transformers to capture global multi-scale information and achieves state-of-the-art performance on several benchmarks, including the BTCV and MSD datasets.

Standardized cardiovascular magnetic resonance imaging (CMR) protocols: 2020 update

This paper presents updated guidelines for standardized cardiovascular magnetic resonance (CMR) imaging protocols, covering various aspects from patient preparation and safety considerations to specific disease-oriented protocols. It emphasizes the use of advanced imaging techniques like T1 and T2 mapping, as well as stress perfusion and late gadolinium enhancement, while acknowledging the limitations posed by vendor-specific sequences and resource availability.

★

☆

s12968-020-00607-1.pdf

Jul 14, 07:01 AM

TimeFlow: Longitudinal Brain Image Registration and Aging Progression Analysis

TimeFlow, a novel framework for longitudinal brain MRI registration, allows for future brain image prediction and aging progression analysis without relying on segmentation. Leveraging a time-conditioned U-Net architecture, it overcomes limitations of existing methods by eliminating the need for explicit smoothness constraints and enabling extrapolation from limited temporal data.