Generative Adversarial Networks for Brain MRI Synthesis: Impact of Training Set Size on Clinical Application.

We evaluated the impact of training set size on generative adversarial networks (GANs) to synthesize brain MRI sequences. We compared three sets of GANs trained to generate pre-contrast T1 (gT1) from post-contrast T1 and FLAIR (gFLAIR) from T2. The baseline models were trained on 135 cases; for this study, we used the same model architecture but a larger cohort of 1251 cases and two stopping rules, an early checkpoint (early models) and one after 50 epochs (late models). We tested all models on an independent dataset of 485 newly diagnosed gliomas. We compared the generated MRIs with the original ones using the structural similarity index (SSI) and mean squared error (MSE). We simulated scenarios where either the original T1, FLAIR, or both were missing and used their synthesized version as inputs for a segmentation model with the original post-contrast T1 and T2. We compared the segmentations using the dice similarity coefficient (DSC) for the contrast-enhancing area, non-enhancing area, and the whole lesion. For the baseline, early, and late models on the test set, for the gT1, median SSI was .957, .918, and .947; median MSE was .006, .014, and .008. For the gFLAIR, median SSI was .924, .908, and .915; median MSE was .016, .016, and .019. The range DSC was .625-.955, .420-.952, and .610-.954. Overall, GANs trained on a relatively small cohort performed similarly to those trained on a cohort ten times larger, making them a viable option for rare diseases or institutions with limited resources.

Does having a rotator cuff repair prior to reverse total shoulder arthroplasty influence the outcome?

AIMS: The number of rotator cuff repairs that are undertaken is increasing. Reverse shoulder arthroplasty (RSA) is the procedure of choice for patients with rotator cuff arthropathy. We sought to determine whether patients who underwent rotator cuff repair and subsequent RSA had different outcomes compared with a matched control group who underwent RSA without a previous rotator cuff repair. PATIENTS AND METHODS: All patients with a history of rotator cuff repair who underwent RSA between 2000 and 2015 with a minimum follow-up of two years were eligible for inclusion as the study group. Outcomes, including the American Shoulder and Elbow Surgeons (ASES) scores, were compared with a matched control group of patients who underwent RSA without having previously undergone rotator cuff repair. RESULTS: The study group included 45 patients. Their mean age was 69 years (sd 8.6) and 27 patients (60%) were women. The mean ASES score improved from 43.1 to 76.6 two years postoperatively, and to 66.9 five years postoperatively. There was no significant difference between the outcomes at two years in the two groups (all p ≥ 0.05), although there was significantly more improvement in ASES scores in the control group (44.5 vs 33.4; p = 0.01). However, there was no significant difference between ASES scores at two and five years when baseline ASES scores were matched in the two groups (p = 0.42 at two years; p = 0.35 at five years). CONCLUSION: Significant improvements in ASES scores were seen following RSA in patients who had previously undergone rotator cuff repair. They had higher baseline ASES scores than those who had not previously undergone this surgery. However, there was no significant difference in outcomes between the two groups, two years postoperatively. Previous rotator cuff repair does not appear to affect the early outcome after RSA adversely.

Comparative outcomes of outpatient and inpatient total shoulder arthroplasty: an analysis of the Medicare dataset.

AIMS: The aim of the present study was to compare the 30- and 90-day re-admission rates and complication rates of outpatient and inpatient total shoulder arthroplasty (TSA). PATIENTS AND METHODS: The United States Medicare Standard Analytical Files database was questioned to identify patients who had undergone outpatient or inpatient TSA between 2005 and 2012. Patient characteristics were compared between the two groups using chi-squared analysis. Multivariate logistic regression analysis was used to control for differences in baseline patient characteristics and to compare the two groups in terms of post-operative complications within 90 days and re-admission within 30 days and 90 days. RESULTS: A total of 123 347 Medicare subscribers underwent TSA between 2005 and 2012; 3493 (2.8%) had the procedure performed as an outpatient. A significantly greater proportion of patients who underwent TSA as inpatients were women, had a history of smoking, and had a greater incidence of medical comorbidity including diabetes, coronary artery disease, congestive heart failure, and chronic kidney disease (p < 0.05 for all). Re-admission rates were significantly higher for inpatients at both 30 days (0.83% versus 0.60%, p = 0.016, odds ratio 1.8) and 90 days (2.87% versus 2.04%, p < 0.001, odds ratio 1.8). Complications, including thromboembolic events (p < 0.001) and surgical site infection (p = 0.002), were significantly higher in inpatients. CONCLUSION: Patients who underwent TSA on an outpatient basis were overall younger and healthier than those who had inpatient surgery, which suggests that patient selection was taking place. After controlling for age, gender, and medical conditions, patients who underwent TSA as outpatients had lower rates of 30- and 90-day re-admission and a lower rate of complications than inpatients. Cite this article: Bone Joint J 2017;99-B:934-8.

In-hospital mortality risk of intertrochanteric hip fractures: a comprehensive review of the US Medicare database from 2005 to 2010.

PURPOSE: Intertrochanteric hip fracture is a common injury in the Medicare population. Very little is known about the in-hospital mortality risk of intertrochanteric hip fractures and associated demographics for the US Medicare population. The purpose of this study is to determine the in-hospital mortality rate of closed intertrochanteric hip fractures and to evaluate demographic factors influencing an increased mortality risk. METHODS: The PearlDiver Medicare database from 2005 to 2010 was queried for closed intertrochanteric hip fractures. Stratified sampling was conducted by creating subset for individuals with a death discharge from inpatient facilities. Statistical analysis was performed where appropriate. RESULTS: Throughout 2005-2010 there were a total of 1,138,142 intertrochanteric hip fractures. There were 19,385 deaths during the initial hospital stay, yielding a mortality rate of 1.70%. There was a 1.83% mortality rate for patients 75 and older and patients over the age of 84 comprised the majority of deaths at 58%. The mortality rate was lower for females (1.39%) than for males (2.56%) (p < 0.0002). CONCLUSION: We found in the Medicare database that there is a relatively low rate of in-hospitality mortality associated with intertrochanteric hip fractures; this rate is lower than previously reported. We report a 1.70% in-hospital mortality using a complete Medicare dataset. Based on previous reporting for short term and one-year mortality risk, the present study suggests that mortality risk is greatest after patients have been released from the hospital. More attention should be paid to understanding and attenuating the mortality associated with intertrochanteric hip fractures after the acute hospital phase.

Biomechanical testing of the reconstructed ulnar collateral ligament: a systematic review of the literature.

PURPOSE: The purpose was to perform a systematic review of the literature investigating biomechanical studies of ulnar collateral ligament reconstruction (UCLR) techniques to summarize the most commonly analyzed methods of fixation (at both the ulna and humerus), the degree of elbow flexion at the time of fixation, graft characteristics, and modes of failure with these techniques. MATERIALS AND METHODS: A systematic review was performed. All cadaveric biomechanical studies that tested a reconstruction method for UCLR were included. Descriptive statistics were calculated for each study and parameter/variable analyzed. RESULTS: Twenty-three studies were included with a total of 397 elbows in 242 cadavers (mean age 54.8 ± 20 years, range 16-96). The majority of studies (65 %) used a palmaris longus graft. The docking technique (37.2 %) was the most commonly tested reconstruction method. Significant heterogeneity between studies precluded assimilation of specific techniques (each of the 23 studies utilized a unique technique). Fixation was performed at 30°-90° of elbow flexion. The most common mode of failure was suture failure (51 %), followed by midsubstance rupture (27.00 %), and bone tunnel fracture (14.00 %). No significant differences were observed amongst techniques for all measures analyzed. CONCLUSION: This study found the docking technique to be the most commonly tested technique, while the mode of reconstruction failure was most commonly at the suture interface. If the graft failed at the bone interface, it was most likely to occur at the ulna. Surgeon preference and comfort level with a specific technique should dictate choice.

Bioceramic bone graft substitute for treatment of unicameral bone cysts.

PURPOSE: To review the outcome of 12 patients who underwent debridement and injection of bioceramic for unicameral bone cyst (UBC). The resorption rate of the bioceramic was estimated by both traditional and novel methods. METHODS: Records of 10 males and 2 females aged 6 to 34 years who underwent debridement and injection of bioceramic for UBC and were followed up for a mean of 41 (range, 26-57) months were reviewed. Functional outcome was assessed using the selfcompleted Musculoskeletal Tumor Society (MSTS) questionnaire. Radiological outcome was assessed using both original and modified Neer Outcome Rating System. The resorption rate of the bioceramic was estimated using both traditional and novel (ImageJ) methods. RESULTS: The mean MSTS score was 29.7 (range, 28-30) indicating excellent functional outcome. Of the 12 patients, 9 achieved complete healing and 3 had a residual cyst of 1%, 11%, and 52%. The last was considered a local recurrence, and the patient underwent repeat percutaneous injection of the bioceramic 1.5 years later and remained disease-free 4 years later. The mean resorption rate was 29% faster when estimated using the traditional rather than the ImageJ method (0.47 vs. 0.33 cm3/day, p=0.02). In the patient with recurrence, the resorption rate was faster than the average (0.68 vs. 0.33 cm3/day). CONCLUSION: A single percutaneous injection of the bioceramic for UBC achieved good functional and radiological outcome while avoiding donor-site morbidity.

Imaging Informatics: 25 Years of Progress.

The science and applications of informatics in medical imaging have advanced dramatically in the past 25 years. This article provides a selective overview of key developments in medical imaging informatics. Advances in standards and technologies for compression and transmission of digital images have enabled Picture Archiving and Communications Systems (PACS) and teleradiology. Research in speech recognition, structured reporting, ontologies, and natural language processing has improved the ability to generate and analyze the reports of imaging procedures. Informatics has provided tools to address workflow and ergonomic issues engendered by the growing volume of medical image information. Research in computeraided detection and diagnosis of abnormalities in medical images has opened new avenues to improve patient care. The growing number of medical-imaging examinations and their large volumes of information create a natural platform for "big data" analytics, particularly when joined with high-dimensional genomic data. Radiogenomics investigates relationships between a disease's genetic and gene-expression characteristics and its imaging phenotype; this emerging field promises to help us better understand disease biology, prognosis, and treatment options. The next 25 years offer remarkable opportunities for informatics and medical imaging together to lead to further advances in both disciplines and to improve health.

Impact of Software Modeling on the Accuracy of Perfusion MRI in Glioma.

BACKGROUND AND PURPOSE: Relative cerebral blood volume, as measured by T2*-weighted dynamic susceptibility-weighted contrast-enhanced MRI, represents the most robust and widely used perfusion MR imaging metric in neuro-oncology. Our aim was to determine whether differences in modeling implementation will impact the correction of leakage effects (from blood-brain barrier disruption) and the accuracy of relative CBV calculations as measured on T2*-weighted dynamic susceptibility-weighted contrast-enhanced MR imaging at 3T field strength. MATERIALS AND METHODS: This study included 52 patients with glioma undergoing DSC MR imaging. Thirty-six patients underwent both non-preload dose- and preload dose-corrected DSC acquisitions, with 16 patients undergoing preload dose-corrected acquisitions only. For each acquisition, we generated 2 sets of relative CBV metrics by using 2 separate, widely published, FDA-approved commercial software packages: IB Neuro and nordicICE. We calculated 4 relative CBV metrics within tumor volumes: mean relative CBV, mode relative CBV, percentage of voxels with relative CBV > 1.75, and percentage of voxels with relative CBV > 1.0 (fractional tumor burden). We determined Pearson (r) and Spearman (ρ) correlations between non-preload dose- and preload dose-corrected metrics. In a subset of patients with recurrent glioblastoma (n = 25), we determined receiver operating characteristic area under the curve for fractional tumor burden accuracy to predict the tissue diagnosis of tumor recurrence versus posttreatment effect. We also determined correlations between rCBV and microvessel area from stereotactic biopsies (n = 29) in 12 patients. RESULTS: With IB Neuro, relative CBV metrics correlated highly between non-preload dose- and preload dose-corrected conditions for fractional tumor burden (r = 0.96, ρ = 0.94), percentage > 1.75 (r = 0.93, ρ = 0.91), mean (r = 0.87, ρ = 0.86), and mode (r = 0.78, ρ = 0.76). These correlations dropped substantially with nordicICE. With fractional tumor burden, IB Neuro was more accurate than nordicICE in diagnosing tumor versus posttreatment effect (area under the curve = 0.85 versus 0.67) (P < .01). The highest relative CBV-microvessel area correlations required preload dose and IB Neuro (r = 0.64, ρ = 0.58, P = .001). CONCLUSIONS: Different implementations of perfusion MR imaging software modeling can impact the accuracy of leakage correction, relative CBV calculation, and correlations with histologic benchmarks.

The RSNA Image Sharing Network.

In the era of health information exchanges, there are trade-offs to consider when sharing a patient's medical record among all providers that a patient might choose. Exchange among in-network partners on the same electronic medical records (EMR) and other integrated information systems is trivial. The patient identifier is common, as are the relevant departmental systems, to all providers. Difficulties arise when patient records including images (and reports) must be shared among different networks and even with the patients themselves. The National Institutes of Health (NIH) challenged Radiological Society of North America (RSNA) to develop a transport method that could supersede the need for physical media (for patients or other providers), replace point-to-point private networks among providers, and enable image exchange on an ad hoc basis between arbitrary health networks without long legal delays. In concert with the evolving US health care paradigm, patient engagement was to be fundamental. With Integrating Healthcare Enterprise's (IHE's) help, the challenge has been met with an operational system.

Computer-aided diagnosis improves detection of small intracranial aneurysms on MRA in a clinical setting.

BACKGROUND AND PURPOSE: MRA is widely accepted as a noninvasive diagnostic tool for the detection of intracranial aneurysms, but detection is still a challenging task with rather low detection rates. Our aim was to examine the performance of a computer-aided diagnosis algorithm for detecting intracranial aneurysms on MRA in a clinical setting. MATERIALS AND METHODS: Aneurysm detectability was evaluated retrospectively in 48 subjects with and without computer-aided diagnosis by 6 readers using a clinical 3D viewing system. Aneurysms ranged from 1.1 to 6.0 mm (mean = 3.12 mm, median = 2.50 mm). We conducted a multireader, multicase, double-crossover design, free-response, observer-performance study on sets of images from different MRA scanners by using DSA as the reference standard. Jackknife alternative free-response operating characteristic curve analysis with the figure of merit was used. RESULTS: For all readers combined, the mean figure of merit improved from 0.655 to 0.759, indicating a change in the figure of merit attributable to computer-aided diagnosis of 0.10 (95% CI, 0.03-0.18), which was statistically significant (F(1,47) = 7.00, P = .011). Five of the 6 radiologists had improved performance with computer-aided diagnosis, primarily due to increased sensitivity. CONCLUSIONS: In conditions similar to clinical practice, using computer-aided diagnosis significantly improved radiologists' detection of intracranial DSA-confirmed aneurysms of ≤6 mm.

Analysis of hemodynamics and aneurysm occlusion after flow-diverting treatment in rabbit models.

BACKGROUND AND PURPOSE: Predicting the outcome of flow diversion treatment of cerebral aneurysms remains challenging. Our aim was to investigate the relationship between hemodynamic conditions created immediately after flow diversion and subsequent occlusion of experimental aneurysms in rabbits. MATERIALS AND METHODS: The hemodynamic environment before and after flow-diversion treatment of elastase-induced aneurysms in 20 rabbits was modeled by using image-based computational fluid dynamics. Local aneurysm occlusion was quantified by using a voxelization technique on 3D images acquired 8 weeks after treatment. Global and local voxel-by-voxel hemodynamic variables were used to statistically compare aneurysm regions that later thrombosed to regions that remained patent. RESULTS: Six aneurysms remained patent at 8 weeks, while 14 were completely or nearly completely occluded. Patent aneurysms had statistically larger neck sizes (P = .0015) and smaller mean transit times (P = .02). The velocity, vorticity, and shear rate were approximately 2.8 times (P < .0001) larger in patent regions-that is, they had larger "flow activity" than regions that progressed to occlusion. Statistical models based on local hemodynamic variables were capable of predicting local occlusion with good precision (84% accuracy), especially away from the neck (92%-94%). Predictions near the neck were poorer (73% accuracy). CONCLUSIONS: These results suggests that the dominant healing mechanism of occlusion within the aneurysm dome is related to slow-flow-induced thrombosis, while near the neck, other processes could be at play simultaneously.

Optimal presentation modes for detecting brain tumor progression.

BACKGROUND AND PURPOSE: A common task in radiology interpretation is visual comparison of images. The purpose of this study was to compare traditional side-by-side and in-place (flicker) image presentation modes with advanced methods for detecting primary brain tumors on MR imaging. MATERIALS AND METHODS: We identified 66 patients with gliomas and 3 consecutive brain MR imaging examinations (a "triplet"). A display application that presented images in side-by-side mode with or without flicker display as well as display of image subtraction or automated change detection information (also with and without flicker display) was used by 3 board-certified neuroradiologists. They identified regions of brain tumor progression by using this display application. Each case was reviewed using all modes (side-by-side presentation with and without flicker, subtraction with and without flicker, and change detection with and without flicker), with results compared via a panel rating. RESULTS: Automated change detection with or without flicker (P < .0027) as well as subtraction with or without flicker (P < .0027) were more sensitive to tumor progression than side-by-side presentation in cases where all 3 raters agreed. Change detection afforded the highest interrater agreement, followed by subtraction. Clinically determined time to progression was longer for cases rated as nonprogressing by using subtraction images and change-detection images both with and without flicker display mode compared with side-by-side presentation. CONCLUSIONS: Automated change detection and image subtraction, with and without flicker display mode, are superior to side-by-side image comparison.

Clinical and radiographic spectrum of pathologically confirmed tumefactive multiple sclerosis.

Atypical imaging features of multiple sclerosis lesions include size >2 cm, mass effect, oedema and/or ring enhancement. This constellation is often referred to as 'tumefactive multiple sclerosis'. Previous series emphasize their unifocal and clinically isolated nature, however, evolution of these lesions is not well defined. Biopsy may be required for diagnosis. We describe clinical and radiographic features in 168 patients with biopsy confirmed CNS inflammatory demyelinating disease (IDD). Lesions were analysed on pre- and post-biopsy magnetic resonance imaging (MRI) for location, size, mass effect/oedema, enhancement, multifocality and fulfilment of Barkhof criteria. Clinical data were correlated to MRI. Female to male ratio was 1.2 : 1, median age at onset, 37 years, duration between symptom onset and biopsy, 7.1 weeks and total disease duration, 3.9 years. Clinical course prior to biopsy was a first neurological event in 61%, relapsing-remitting in 29% and progressive in 4%. Presentations were typically polysymptomatic, with motor, cognitive and sensory symptoms predominating. Aphasia, agnosia, seizures and visual field defects were observed. At follow-up, 70% developed definite multiple sclerosis, and 14% had an isolated demyelinating syndrome. Median time to second attack was 4.8 years, and median EDSS at follow-up was 3.0. Multiple lesions were present in 70% on pre-biopsy MRI, and in 83% by last MRI, with Barkhof criteria fulfilled in 46% prior to biopsy and 55% by follow-up. Only 17% of cases remained unifocal. Median largest lesion size on T2-weighted images was 4 cm (range 0.5-12), with a discernible size of 2.1 cm (range 0.5-7.5). Biopsied lesions demonstrated mass effect in 45% and oedema in 77%. A strong association was found between lesion size, and presence of mass effect and/or oedema (P < 0.001). Ring enhancement was frequent. Most tumefactive features did not correlate with gender, course or diagnosis. Although lesion size >5 cm was associated with a slightly higher EDSS at last follow-up, long-term prognosis in patients with disease duration >10 years was better (EDSS 1.5) compared with a population-based multiple sclerosis cohort matched for disease duration (EDSS 3.5; P < 0.001). Given the retrospective nature of the study, the precise reason for biopsy could not always be determined. This study underscores the diagnostically challenging nature of CNS IDDs that present with atypical clinical or radiographic features. Most have multifocal disease at onset, and develop RRMS by follow-up. Although increased awareness of this broad spectrum may obviate need for biopsy in many circumstances, an important role for diagnostic brain biopsy may be required in some cases.

Imaging of remyelination and neuronal health.

Remyelination of axons that have been demyelinated due to multiple sclerosis (MS) may be a critical step in restoring the damaged axons and reversing the disease process. While it is possible to establish the presence of remyelination with microscopy of tissue samples, it is important to have noninvasive or minimally invasive methods to measure remyelination in living animals and humans. Such tools are critical to establishing the efficacy of agents purported to promote or enhance remyelination. This chapter reviews the technology of imaging of the brain, its application to MS, and the current state of imaging techniques for measuring remyelination and the health of the associated neurons in the setting of MS.

Pattern of T2 hypointensity associated with ring-enhancing brain lesions can help to differentiate pathology.

Ring-enhancing lesions seen on MR images can occur with a variety of etiologies. Some ring-enhancing lesions have hypointense rims peripherally on T2-weighted MR images. In this study, we examined whether T2 hypointense rims were associated with specific pathologies. A search for ring-enhancing lesions on MR images obtained from 1996 to 2004 was performed, and revealed 221 patients with MRI findings of ring enhancement. The pattern of T2 hypointensity (arc or rim) corresponding with ring enhancement was recorded. In addition, we analyzed other imaging characteristics, including signal on diffusion-weighted images, central homogeneity on T2 and multiplicity of lesions. We then reviewed clinical data on the patients to ascertain the diagnosis for each examination. The most common associated pathologies in our study were gliomas (40%), metastases (30%), abscesses (8%) and multiple sclerosis (MS; 6%). Hypointense borders on T2-weighted images were present in 67% of lesions in the form of a rim in 40% and an arc in 60%. Abscesses had the highest percentage of hypointense rims. Metastases and gliomas more commonly had arcs, and MS lesions were divided between rims and arcs. Abscesses and MS lesions were more commonly homogeneous centrally, compared to gliomas and metastases. Additionally, abscesses were more often bright on diffusion imaging than the other pathologies. As expected, abscesses and MS lesions were usually multiple, whereas metastases were typically multiple in approximately 50% of the patients; gliomas were generally solitary. Trends in T2 hypointensity may aid in distinguishing among etiologies of ring-enhancing lesions, although there is overlap between the MR appearance of these various pathologies.

FLAIR histogram segmentation for measurement of leukoaraiosis volume.

The purposes of this study were to develop a method to measure brain and white matter hyperintensity (leukoaraiosis) volume that is based on the segmentation of the intensity histogram of fluid-attenuated inversion recovery (FLAIR) images and to assess the accuracy and reproducibility of the method. Whole-head synthetic image phantoms with manually introduced leukoaraiosis lesions of varying severity were constructed. These synthetic image phantom sets incorporated image contrast and anatomic features that mimicked leukoaraiosis found in real life. One set of synthetic image phantoms was used to develop the segmentation algorithm (FLAIR-histoseg). A second set was used to measure its accuracy. Test retest reproducibility was assessed in 10 elderly volunteers who were imaged twice. The mean absolute error of the FLAIR-histoseg method was 6.6% for measurement of leukoaraiosis volume and 1.4% for brain volume. The mean test retest coefficient of variation was 1.4% for leukoaraiosis volume and 0.3% for brain volume. We conclude that the FLAIR-histoseg method is an accurate and reproducible method for measuring leukoaraiosis and whole-brain volume in elderly subjects.

Requirements for an enterprise digital image archive.

This report describes several image archival problems facing the authors' department and the results of their attempt to define the requirements for an enterprise digital image archive. The problems identified include the costs of supporting multiple distinct archives, the increased complexity of supporting multiple archive interfaces, the differences in data handling policies and resulting variations in data integrity, and variability in support for nonimage data. The authors also describe the data collected including image volumes and trends and imaging device trends. Finally, the resulting specification for an enterprise digital image archive, including storage and retrieval performance and interface requirements are presented.