Search

BACKGROUND: Delusions affect approximately a third of Alzheimer’s disease(AD) patients and are associated with poor outcomes. Previous studies investigating the neuroanatomical correlates of delusions have yet to reach a consensus, with findings of reduced volume across all the lobes, particularly in frontal regions. The current study examined the grey matter(GM) differences associated with delusions in AD. METHODS: Using voxel-based morphometry(VBM), we assessed GM for 23 AD patients who developed delusions(AD+D) and 36 comparable AD patients who did not develop delusions(AD-D) at baseline and follow up. ANOVA was used to identify consistent differences between AD+D and AD-D patients across both timepoints(main effect of group), consistent changes from baseline to follow up(main effect of time) and differential changes between AD+D and AD-D over time(interaction of group and time). All data were obtained from the NACC database. RESULTS: The AD+D group had consistently lower frontal GM volume, although both groups showed decreased GM in fronto-temporal brain regions over time. An interaction was observed between delusions and longitudinal change, with AD+D patients having significantly elevated GM in predominantly temporal areas at baseline assessment, which had declined to become significantly lower than the AD-D group at follow-up. CONCLUSION: These findings suggest that although individuals with and without delusion show long-term GM decreases, there are specific volumetric markers that distinguish patients with delusion from those without, before and after the onset of delusions. Specifically, the decline of GM in temporal areas that were elevated prior to the onset of delusion may be involved in the manifestation of delusions.

Preoperative carbohydrate beverages have been shown to be beneficial in improving patient outcomes. There have been several investigations into the safety of maltodextrin as a preoperative carbohydrate. Although alternative preoperative carbohydrate sources have been proposed, there have been few investigations into the safety and gastric emptying of novel carbohydrate beverages. The present study aimed to compare the gastric emptying of phytoglycogen and maltodextrin to evaluate safety for use as presurgical carbohydrate beverages. In a quasi-experimental design, ten healthy participants orally consumed either a 12.5% maltodextrin or a 12.5% phytoglycogen solution. Gamma scintigraphy was used to evaluate gastric emptying at baseline at 45, 90, and 120 min. Serum insulin and serum glucose were measured at baseline at 15, 30, 45, 60, 90, and 120 min. Gastric volume was significantly lower in the phytoglycogen group at 45 min (p = 0.01) and 90 min (p = 0.01), but this difference lost significance at 120 min (p = 0.17). There were no significant differences between treatments for serum insulin or serum glucose at any time point. This study indicates that the gastric emptying of phytoglycogen is comparable to maltodextrin at 120 min after ingestion, opening the opportunity for the study of alternative carbohydrates for utilization as preoperative carbohydrates.

Artificial intelligence (AI) is rapidly moving from an experimental phase to an implementation phase in many fields, including medicine. The combination of improved availability of large datasets, increasing computing power, and advances in learning algorithms has created major performance breakthroughs in the development of AI applications. In the last 5 years, AI techniques known as deep learning have delivered rapidly improving performance in image recognition, caption generation, and speech recognition. Radiology, in particular, is a prime candidate for early adoption of these techniques. It is anticipated that the implementation of AI in radiology over the next decade will significantly improve the quality, value, and depth of radiology's contribution to patient care and population health, and will revolutionize radiologists' workflows. The Canadian Association of Radiologists (CAR) is the national voice of radiology committed to promoting the highest standards in patient-centered imaging, lifelong learning, and research. The CAR has created an AI working group with the mandate to discuss and deliberate on practice, policy, and patient care issues related to the introduction and implementation of AI in imaging. This white paper provides recommendations for the CAR derived from deliberations between members of the AI working group. This white paper on AI in radiology will inform CAR members and policymakers on key terminology, educational needs of members, research and development, partnerships, potential clinical applications, implementation, structure and governance, role of radiologists, and potential impact of AI on radiology in Canada.

Rationale and ObjectivesRadiology residents acquire a diverse educational experience and skill set, including a general internship year, which may enable the direct management of patients. In order for radiology residents to define new scopes of practice, however, additional fellowship training may in certain instances be warranted.Materials and MethodsUsing the Canadian family medicine Enhanced Skills Program as a model, we conducted a Canada-wide survey of radiology residents to assess interest in additional fellowship training to expand their scope of practice.ResultsOur results indicate that a majority of residents (69.2%) would like to routinely see patients in clinic and more than half (52%) are willing to undergo an additional year of fellowship to enhance their skill set. The most popular choices for such fellowships were sports medicine (22.8%), emergency medicine (19.6%) and vascular medicine (18.5%). In addition, a majority (52.9%) of residents felt capable of offering incidentaloma clinics without additional training beyond their core radiology residency.ConclusionTraditional diagnostic and interventional radiology fellowships must be reconsidered to reflect the interests and capabilities of modern radiology trainees. Expansion of training options into the domain of direct patient management will likely prove popular among current residents.

Medical datasets are often highly imbalanced with over-representation of prevalent conditions and poor representation of rare medical conditions. Due to privacy concerns, it is challenging to aggregate large datasets between health care institutions. We propose synthesizing pathology in medical images as a means to overcome these challenges. We implement a deep convolutional generative adversarial network (DCGAN) to create synthesized chest X-rays based upon a modest sized labeled dataset. We used a combination of real and synthesized images to train deep convolutional neural networks (DCNNs) to detect pathology across five classes of chest X-rays. The comparative study of DCNNs trained with the combination of real and synthesized images showed that these networks can outperform similar networks trained solely with real images in pathology classification. This improved performance is largely attributable to the balancing of the dataset using DCGAN synthesized images, where classes that are lacking in example images are preferentially augmented.

In this paper, we propose a novel technique for sampling sequential images using a cylindrical transform in a cylindrical coordinate system for kidney semantic segmentation in abdominal computed tomography (CT). The images generated from a cylindrical transform augment a limited annotated set of images in three dimensions. This approach enables us to train contemporary classification deep convolutional neural networks (DCNNs) instead of fully convolutional networks (FCNs) for semantic segmentation. Typical semantic segmentation models segment a sequential set of images (e.g. CT or video) by segmenting each image independently. However, the proposed method not only considers the spatial dependency in the x-y plane, but also the spatial sequential dependency along the z-axis. The results show that classification DCNNs, trained on cylindrical transformed images, can achieve a higher segmentation performance value than FCNs using a limited number of annotated images., Comment: This paper is accepted for presentation at IEEE Global Conference on Signal and Information Processing (IEEE GlobalSIP), California, USA, 2018

Introduction: The purpose of this study was to determine the variety and prevalence of renal and non-renal abnormalities detected on multidetector computed tomography (MDCT) that precluded patients from donating a kidney.Methods: Institutional review board approval was obtained and the requirement for informed consent was waived. A retrospective, single-centre review of 701 patients (444 female, 257 male; age range 18–86 years; mean age 43.2±11.9 years) that underwent renal donor protocol MDCT was conducted. A systematic review of the CT report, records from multidisciplinary renal transplantation rounds, and electronic medical records was performed to determine which patients were approved or declined as live renal donors. If declined as a donor, CT-identified reasons were categorized as abnormalities of renal vasculature, renal parenchyma, collecting system, or extra-renal.Results: A total of 81 patients were excluded as renal donors on the basis of CT findings. Abnormalities of the collecting system accounted for the most frequent cause of exclusion (n=41), with asymptomatic renal calculi being detected in 39 patients. Complex vascular anatomy and vascular abnormalities resulted in the exclusion of 29 patients. Supernumerary arteries and early arterial branching resulted in the exclusion of 20 patients, while renal vein anomalies leading to exclusion were uncommon (n=2). Abnormalities of renal parenchyma resulted in the exclusion of nine patients. Three patients were diagnosed with autosomal dominant polycystic kidney disease, two patients had renal cell carcinoma, and two patients had areas of cortical scarring. A complex cystic lesion requiring surveillance imaging was encountered in one patient and a large area of renal infarction related to prior adrenalectomy was demonstrated in one patient. Extra-renal abnormalities leading to exclusion were limited to two patients with pulmonary nodules.Conclusions: MDCT plays a critical role in the preoperative assessment of potential renal donors by identifying contraindications to donor nephrectomy and providing accurate vascular mapping. This study is anticipated to be informative for those involved in the workup of potential living renal donors by quantifying the incidence and reasons for donor exclusion identified on CT.

Deep learning models have a large number of free parameters that must be estimated by efficient training of the models on a large number of training data samples to increase their generalization performance. In real-world applications, the data available to train these networks is often limited or imbalanced. We propose a sampling method based on the radial transform in a polar coordinate system for image augmentation to facilitate the training of deep learning models from limited source data. This pixel-wise transform provides representations of the original image in the polar coordinate system by generating a new image from each pixel. This technique can generate radial transformed images up to the number of pixels in the original image to increase the diversity of poorly represented image classes. Our experiments show improved generalization performance in training deep convolutional neural networks with radial transformed images., This paper is accepted for presentation at IEEE International Conference on Acoustics, Speech and Signal Processing (IEEE ICASSP), 2018

Pathfinding in hospitals is challenging for patients, visitors, and even employees. Many people have experienced getting lost due to lack of clear guidance, large footprint of hospitals, and confusing array of hospital wings. In this paper, we propose Halo; An indoor navigation application based on voice-user interaction to help provide directions for users without assistance of a localization system. The main challenge is accurate detection of origin and destination search terms. A custom convolutional neural network (CNN) is proposed to detect origin and destination search terms from transcription of a submitted speech query. The CNN is trained based on a set of queries tailored specifically for hospital and clinic environments. Performance of the proposed model is studied and compared with Levenshtein distance-based word matching., This paper is accepted for presentation at 2017 IEEE 28th Annual International Symposium on Personal, Indoor, and Mobile Radio Communications

Background There are limited data on the effects of trastuzumab on the right ventricle (RV). Therefore, we sought to evaluate the temporal changes in right ventricular (RV) structure and function as measured by cardiovascular magnetic resonance (CMR), and their relationship with left ventricular (LV) structure and function in breast cancer patients treated with trastuzumab. Methods Prospective, longitudinal, observational study involving 41 women with HER2+ breast cancer who underwent serial CMR at baseline, 6, 12, and 18 months after initiation of trastuzumab. A single blinded observer measured RV parameters on de-identified CMRs in a random order. Linear mixed models were used to investigate temporal changes in RV parameters. Results Of the 41 women (age 52 ± 11 years), only one patient experienced trastuzumab-induced cardiotoxicity. Compared to baseline, there were small but significant increases in the RV end-diastolic volume at 6 months (p = 0.002) and RV end-systolic volume at 6 and 12 months (p

Medical datasets are often highly imbalanced with over-representation of common medical problems and a paucity of data from rare conditions. We propose simulation of pathology in images to overcome the above limitations. Using chest X-rays as a model medical image, we implement a generative adversarial network (GAN) to create artificial images based upon a modest sized labeled dataset. We employ a combination of real and artificial images to train a deep convolutional neural network (DCNN) to detect pathology across five classes of chest X-rays. Furthermore, we demonstrate that augmenting the original imbalanced dataset with GAN generated images improves performance of chest pathology classification using the proposed DCNN in comparison to the same DCNN trained with the original dataset alone. This improved performance is largely attributed to balancing of the dataset using GAN generated images, where image classes that are lacking in example images are preferentially augmented., Comment: This paper is accepted for presentation at IEEE International Conference on Acoustics, Speech and Signal Processing (IEEE ICASSP), 2018