Insights into Imaging

Poor responders to chemoradiotherapy (CRT) for locally advanced rectal cancer (LARC) can still have a good prognosis if the treatment strategy is changed in time. However, no reliable predictor of early-treatment response has been identified. The purpose of this study was to investigate the role of T2 relaxation time in magnetic resonance imaging (MRI) for the early prediction of a pathological response to CRT in LARC. A total of 123 MRIs were performed on 41 LARC patients immediately before, during, and after CRT. The corresponding tumor volume, T2 relaxation time, and apparent diffusion coefficient (ADC) values at different scan time points were obtained. The Mann–Whitney U test was used to compare the T2 relaxation time between pathological good responders (GR) and non-good responders (non-GR). The area under the curve (AUC) value was used to quantify the diagnostic ability of each parameter in predicting tumor response to CRT. Twenty-one (51%) and 20 (49%) were GRs and non-GRs, respectively. T2 relaxation time showed an excellent intraclass correlation coefficient (ICC) of > 0.85 at three-time points. It was significantly lower in the GR group than in the non-GR group during and after CRT. The early T2 decrease had a high AUC of 0.91 in differentiating non-GRs and GRs, similar to 0.90 of the T2 value after CRT. T2 relaxation time may help predict treatment response to CRT for LARC earlier, rather than having to wait until the end of CRT, thereby alleviating the physical burden for patients with no good response.

Convolutional neural network (CNN), a class of artificial neural networks that has become dominant in various computer vision tasks, is attracting interest across a variety of domains, including radiology. CNN is designed to automatically and adaptively learn spatial hierarchies of features through backpropagation by using multiple building blocks, such as convolution layers, pooling layers, and fully connected layers. This review article offers a perspective on the basic concepts of CNN and its application to various radiological tasks, and discusses its challenges and future directions in the field of radiology. Two challenges in applying CNN to radiological tasks, small dataset and overfitting, will also be covered in this article, as well as techniques to minimize them. Being familiar with the concepts and advantages, as well as limitations, of CNN is essential to leverage its potential in diagnostic radiology, with the goal of augmenting the performance of radiologists and improving patient care. • Convolutional neural network is a class of deep learning methods which has become dominant in various computer vision tasks and is attracting interest across a variety of domains, including radiology. • Convolutional neural network is composed of multiple building blocks, such as convolution layers, pooling layers, and fully connected layers, and is designed to automatically and adaptively learn spatial hierarchies of features through a backpropagation algorithm. • Familiarity with the concepts and advantages, as well as limitations, of convolutional neural network is essential to leverage its potential to improve radiologist performance and, eventually, patient care.

This is a condensed summary of an international multisociety statement on ethics of artificial intelligence (AI) in radiology produced by the ACR, European Society of Radiology, RSNA, Society for Imaging Informatics in Medicine, European Society of Medical Imaging Informatics, Canadian Association of Radiologists, and American Association of Physicists in Medicine. AI has great potential to increase efficiency and accuracy throughout radiology, but also carries inherent pitfalls and biases. Widespread use of AI-based intelligent and autonomous systems in radiology can increase the risk of systemic errors with high consequence, and highlights complex ethical and societal issues. Currently, there is little experience using AI for patient care in diverse clinical settings. Extensive research is needed to understand how to best deploy AI in clinical practice. This statement highlights our consensus that ethical use of AI in radiology should promote well-being, minimize harm, and ensure that the benefits and harms are distributed among stakeholders in a just manner. We believe AI should respect human rights and freedoms, including dignity and privacy. It should be designed for maximum transparency and dependability. Ultimate responsibility and accountability for AI remains with its human designers and operators for the foreseeable future. The radiology community should start now to develop codes of ethics and practice for AI which promote any use that helps patients and the common good and should block use of radiology data and algorithms for financial gain without those two attributes.

The purpose of the study was to investigate differences in ultrasound shear wave speed (SWS) between uninjured and injured limbs following hamstring strain injury (HSI) at time of injury (TOI), return to sport (RTS), and 12 weeks after RTS (12wks). This observational, prospective, cross-sectional design included male and female collegiate athletes who sustained an HSI. SWS imaging was performed at TOI, RTS, and 12wks with magnetic resonance imaging. SWS maps were acquired by a musculoskeletal-trained sonographer at the injury location of the injured limb and location-matched on the contralateral limb. The average SWS from three 5 mm diameter Q-boxes on each limb were used for analysis. A linear mixed effects model was performed to determine differences in SWS between limbs across the study time points. SWS was lower in the injured limb compared to the contralateral limb at TOI (uninjured – injured limb difference: 0.23 [0.05, 0.41] m/s, p = 0.006). No between-limb differences in SWS were observed at RTS (0.15 [-0.05, 0.36] m/s, p = 0.23) or 12wks (-0.11 [-0.41, 0.18] m/s, p = 0.84). The SWS in the injured limb of collegiate athletes after HSI was lower compared to the uninjured limb at TOI but not at RTS or 12 weeks after RTS. Hamstring strain injury with structural disruption can be detected by lower injured limb shear wave speed compared to the uninjured limb. Lack of between-limb differences at return to sport may demonstrate changes consistent with healing. Shear wave speed may complement traditional ultrasound or MRI for monitoring muscle injury. • Ultrasound shear wave speed can non-invasively measure tissue elasticity in muscle injury locations. • Injured limb time of injury shear wave speeds were lower versus uninjured limb but not thereafter. • Null return to sport shear wave speed differences may correspond to structural changes associated with healing. • Shear wave speed may provide quantitative measures for monitoring muscle elasticity during recovery.

Laparoscopic nephron-sparing surgery (L-NSS) is increasingly performed to treat localised renal lesions. However, the associated morbidity is non-negligible, with a rate of major complications approaching 10 %. This paper provides an overview of indications, surgical techniques and results of L-NSS; explains the incidence, risk factors and manifestations of postoperative complications; discusses the preferred multidetector computed tomography (CT) acquisition techniques; illustrates the appearance of normal postoperative images following L-NSS; and reviews, with example images, the most common and unusual iatrogenic complications. These include haematuria, haemorrhage, vascular injuries, infections and urinary leaks. Most emphasis is placed on CT, which provides rapid, reliable triage and follow-up of iatrogenic complications after L-NSS, identifying occurrences that require transarterial embolisation or repeated surgery. Multidetector CT allows precise assessment of the surgical resection site; detection of pneumoperitoneum and subcutaneous emphysema; quantification of retroperitoneal blood; and identification of active bleeding, pseudoaneurysms, arterio-venous fistulas, abscess collections and extravasated urine. • Laparoscopic nephron-sparing surgery (NSS) is increasingly performed to treat renal lesions. • Radiologists are increasingly requested to investigate suspected post-surgical NSS complications. • Post-NSS complications include haemorrhage, haematuria, vascular injuries, infections and urinary leaks. • Multidetector CT allows choice between conservative treatment, transarterial embolisation or surgery.