App Review: Management Guide for Incidental Findings on CT and MRI

Radiology, Mobile Devices, and Internet of Things (IoT)

Radiology by its nature is intricately connected to the Internet and is at the forefront of technology in medicine. The past few years have seen a dramatic rise in Internet-based technology in healthcare, with imaging as a core application. Numerous Internet-based applications and technologies have made forays into medicine, and for radiology it is more seamless than in other clinical specialties. Many applications in the practice of radiology are Internet based and more applications are being added every day. Introduction of mobile devices and their integration into imaging workflow has reinforced the role played by the Internet in radiology. Due to the rapid proliferation of wearable devices and smartphones, IoT-enabled technology is evolving healthcare from conventional hub-based systems to more personalized healthcare systems. This article briefly discusses how the IoT plays a useful role in daily imaging workflow and current and potential future applications, how mobile devices can be integrated into radiology workflows, and the impact of the IoT on resident and medical student education, research, and patient engagement in radiology.

Deep Feature Selection and Causal Analysis of Alzheimer's Disease

Deep convolutional neural networks (DCNNs) have achieved great success for image classification in medical research. Deep learning with brain imaging is the imaging method of choice for the diagnosis and prediction of Alzheimer’s disease (AD). However, it is also well known that DCNNs are “black boxes” owing to their low interpretability to humans. The lack of transparency of deep learning compromises its application to the prediction and mechanism investigation in AD. To overcome this limitation, we develop a novel general framework that integrates deep leaning, feature selection, causal inference, and genetic-imaging data analysis for predicting and understanding AD. The proposed algorithm not only improves the prediction accuracy but also identifies the brain regions underlying the development of AD and causal paths from genetic variants to AD via image mediation. The proposed algorithm is applied to the Alzheimer’s Disease Neuroimaging Initiative (ADNI) dataset with diffusion tensor imaging (DTI) in 151 subjects (51 AD and 100 non-AD) who were measured at four time points of baseline, 6 months, 12 months, and 24 months. The algorithm identified brain regions underlying AD consisting of the temporal lobes (including the hippocampus) and the ventricular system.