The purpose of this study was to investigate the potential of using clinically provided spine label annotations stored in a single institution image archive as training data for deep learning-based ...vertebral detection and labeling pipelines. Lumbar and cervical magnetic resonance imaging cases with annotated spine labels were identified and exported from an image archive. Two separate pipelines were configured and trained for lumbar and cervical cases respectively, using the same setup with convolutional neural networks for detection and parts-based graphical models to label the vertebrae. The detection sensitivity, precision and accuracy rates ranged between 99.1–99.8, 99.6–100, and 98.8–99.8% respectively, the average localization error ranges were 1.18–1.24 and 2.38–2.60 mm for cervical and lumbar cases respectively, and with a labeling accuracy of 96.0–97.0%. Failed labeling results typically involved failed S1 detections or missed vertebrae that were not fully visible on the image. These results show that clinically annotated image data from one image archive is sufficient to train a deep learning-based pipeline for accurate detection and labeling of MR images depicting the spine. Further, these results support using deep learning to assist radiologists in their work by providing highly accurate labels that only require rapid confirmation.
Full text
Available for:
NUK, OBVAL, SBMB, SBNM, UL, UM, UPUK, VSZLJ
Summary Background People with type 2 diabetes are at risk of cognitive impairment and brain atrophy. We aimed to compare the effects on cognitive function and brain volume of intensive versus ...standard glycaemic control. Methods The Memory in Diabetes (MIND) study was done in 52 clinical sites in North America as part of Action to Control Cardiovascular Risk in Diabetes (ACCORD), a double two-by-two factorial parallel group randomised trial. Participants (aged 55–80 years) with type 2 diabetes, high glycated haemoglobin A1c (HbA1c ) concentrations (>7·5%; >58 mmol/mol), and a high risk of cardiovascular events were randomly assigned to receive intensive glycaemic control targeting HbA1c to less than 6·0% (42 mmol/mol) or a standard strategy targeting HbA1c to 7·0–7·9% (53–63 mmol/mol). Randomisation was via a centralised web-based system and treatment allocation was not masked from clinic staff or participants. We assessed our cognitive primary outcome, the Digit Symbol Substitution Test (DSST) score, at baseline and at 20 and 40 months. We assessed total brain volume (TBV), our primary brain structure outcome, with MRI at baseline and 40 months in a subset of participants. We included all participants with follow-up data in our primary analyses. In February, 2008, raised mortality risk led to the end of the intensive treatment and transition of those participants to standard treatment. We tested our cognitive function hypotheses with a mixed-effects model that incorporated information from both the 20 and 40 month outcome measures. We tested our MRI hypotheses with an ANCOVA model that included intracranial volume and factors used to stratify randomisation. This study is registered with ClinicalTrials.gov , number NCT00182910. Findings We consecutively enrolled 2977 patients (mean age 62·5 years; SD 5·8) who had been randomly assigned to treatment groups in the ACCORD study. Our primary cognitive analysis was of patients with a 20-month or 40-month DSST score: 1378 assigned to receive intensive treatment and 1416 assigned to receive standard treatment. Of the 614 patients with a baseline MRI, we included 230 assigned to receive intensive treatment and 273 assigned to receive standard treatment in our primary MRI analysis at 40 months. There was no significant treatment difference in mean 40-month DSST score (difference in mean 0·32, 95% CI −0·28 to 0·91; p=0·2997). The intensive-treatment group had a greater mean TBV than the standard-treatment group (4·62, 2·0 to 7·3; p=0·0007). Interpretation Although significant differences in TBV favoured the intensive treatment, cognitive outcomes were not different. Combined with the non-significant effects on other ACCORD outcomes, and increased mortality in participants in the intensive treatment group, our findings do not support the use of intensive therapy to reduce the adverse effects of diabetes on the brain in patients with similar characteristics to those of our participants. Funding US National Institute on Aging and US National Heart, Lung, and Blood Institute.
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UL, UM, UPCLJ, UPUK
Magnetic resonance fingerprinting: an overview Tippareddy, Charit; Zhao, Walter; Sunshine, Jeffrey L. ...
European journal of nuclear medicine and molecular imaging,
12/2021, Volume:
48, Issue:
13
Journal Article
Peer reviewed
Magnetic resonance fingerprinting (MRF) is an evolving quantitative MRI framework consisting of unique data acquisition, processing, visualization, and interpretation steps. MRF is capable of ...simultaneously producing multiple high-resolution property maps including T1, T2, M0, ADC, and T2* measurements. While a relatively new technology, MRF has undergone rapid development for a variety of clinical applications from brain tumor characterization and epilepsy imaging to characterization of prostate cancer, cardiac imaging, among others. This paper will provide a brief overview of current state of MRF technology including highlights of technical and clinical advances. We will conclude with a brief discussion of the challenges that need to be overcome to establish MRF as a quantitative imaging biomarker.
Full text
Available for:
DOBA, EMUNI, FIS, FZAB, GEOZS, GIS, IJS, IMTLJ, IZUM, KILJ, KISLJ, MFDPS, NLZOH, NUK, OILJ, PILJ, PNG, SAZU, SBCE, SBJE, SBMB, SBNM, SIK, UILJ, UKNU, UL, UM, UPUK, VKSCE, VSZLJ, ZAGLJ
Predictive models and anecdotal articles suggest radiology practices were losing 50%-70% of their normal imaging volume during the COVID-19 pandemic. Using actual institutional data, we investigated ...the change in imaging utilization and revenue during this public health crisis.
Imaging performed within the 8-week span between March 8 and April 30, 2020 was categorized into the COVID-19 healthcare crisis timeframe. The first week of this date range and the 10 weeks prior were used to derive the normal practice expected volume. A rolling 7-day total value was used for volume tracking and comparison. Total imaging utilization was derived and organized by patient setting (outpatient, inpatient, emergency) and imaging modality (X-ray, CT, Mammography, MRI, Nuclear Medicine/PET, US). The three highest volume hospitals were analyzed. Revenue information was collected from the hospital billing system.
System-wide imaging volume decreased by 55% between April 7 and 13, 2020. Outpatient exams decreased by 68% relative to normal practice. Emergency exams decreased by 48% and inpatient exams declined by 31%. Mammograms and nuclear medicine scans were the most affected modalities, decreasing by 93% and 61%, respectively. The main campus hospital experienced less relative imaging volume loss compared to the other smaller and outpatient-driven hospitals. At its lowest point, the technical component revenue from main campus imaging services demonstrated a 49% negative variance from normal practice.
The trends and magnitude of the actual imaging utilization data presented will help inform evidence-based decisions for more accurate volume predictions, policy changes, and institutional preparedness for current and future pandemics.
Laser interstitial thermal therapy has been used as an ablative treatment for glioma; however, its development was limited due to technical issues. The NeuroBlate System incorporates several ...technological advances to overcome these drawbacks. The authors report a Phase I, thermal dose-escalation trial assessing the safety and efficacy of NeuroBlate in recurrent glioblastoma multiforme (rGBM).
Adults with suspected supratentorial rGBM of 15- to 40-mm dimension and a Karnofsky Performance Status score of ≥ 60 were eligible. After confirmatory biopsy, treatment was delivered using a rigid, gas-cooled, side-firing laser probe. Treatment was monitored using real-time MRI thermometry, and proprietary software providing predictive thermal damage feedback was used by the surgeon, along with control of probe rotation and depth, to tailor tissue coagulation. An external data safety monitoring board determined if toxicity at lower levels justified dose escalation.
Ten patients were treated at the Case Comprehensive Cancer Center (Cleveland Clinic and University Hospitals-Case Medical Center). Their average age was 55 years (range 34-69 years) and the median preoperative Karnofsky Performance Status score was 80 (range 70-90). The mean tumor volume was 6.8 ± 5 cm(3) (range 2.6-19 cm(3)), the percentage of tumor treated was 78% ± 12% (range 57%-90%), and the conformality index was 1.21 ± 0.33 (range 1.00-2.04). Treatment-related necrosis was evident on MRI studies at 24 and 48 hours. The median survival was 316 days (range 62-767 days). Three patients improved neurologically, 6 remained stable, and 1 worsened. Steroid-responsive treatment-related edema occurred in all patients but one. Three had Grade 3 adverse events at the highest dose.
NeuroBlate represents new technology for delivering laser interstitial thermal therapy, allowing controlled thermal ablation of deep hemispheric rGBM. CLINICAL TRIAL REGISTRATION NO.: NCT00747253 ( ClinicalTrials.gov ).
The workload of US radiologists has increased over the past two decades as measured through total annual relative value units (RVUs). This increase in RVUs generated suggests that radiologists’ ...productivity has increased. However, true productivity (output unit per input unit; RVU per time) is at large unknown since actual time required to interpret and report a case is rarely recorded. In this study, we analyzed how the time to read a case varies between radiologists over a set of different procedure types by retrospectively extracting reading times from PACS usage logs. Specifically, we tested two hypotheses that; i) relative variation in time to read per procedure type increases as the median time to read a procedure type increases, and ii) relative rankings in terms of median reading speed for individual radiologists are consistent across different procedure types. The results that, i) a correlation of -0.25 between the coefficient of variation and median time to read and ii) that only 12 out of 46 radiologists had consistent rankings in terms of time to read across different procedure types, show both hypotheses to be without support. The results show that workload distribution will not follow any general rule for a radiologist across all procedures or a general rule for a specific procedure across many readers. Rather the findings suggest that improved overall practice efficiency can be achieved only by taking into account radiologists’ individual productivity per procedure type when distributing unread cases.
Full text
Available for:
NUK, OBVAL, SBMB, SBNM, UL, UM, UPUK, VSZLJ
Inherent and acquired factors determine the integrated autonomic response to cardiovascular stressors. Excessive sympathoexcitation to ischemic stress is a major contributor to the potential for ...sudden cardiac death. To define fundamental aspects of cardiac-related autonomic neural network interactions within the thoracic cord, specifically as related to modulating sympathetic preganglionic (SPN) neural activity. Adult, anesthetized Yorkshire pigs (
= 10) were implanted with penetrating high-density microarrays (64 electrodes) at the T2 level of the thoracic spinal cord to record extracellular potentials concurrently from left-sided dorsal horn (DH) and SPN neurons. Electrical stimulation of the T2 paravertebral chain allowed for antidromic identification of SPNs located in the intermediolateral cell column (57 of total 1,760 recorded neurons). Cardiac stressors included epicardial touch, occlusion of great vessels to transiently alter preload/afterload, and transient occlusion of the left anterior descending coronary artery (LAD). Spatial/temporal assessment of network interactions was characterized by cross-correlation analysis. While some DH neurons responded solely to changes in preload/afterload (8.5 ± 1.9%) or ischemic stress (10.5 ± 3.9%), the majority of cardiovascular-related DH neurons were multimodal (30.2 ± 4.7%) with ischemia sensitivity being one of the modalities (26.1 ± 4.7%). The sympathoexcitation associated with transient LAD occlusion was associated with increased correlations from baseline within DH neurons (2.43 ± 0.61 to 7.30 ± 1.84%,
= 0.04) and between SPN to DH neurons (1.32 ± 0.78 to 7.24 ± 1.84%,
= 0.02). DH to SPN network correlations were reduced during great vessel occlusion. In conclusion, increased intrasegmental network coherence within the thoracic spinal cord contributes to myocardial ischemia-induced sympathoexcitation.
In an in vivo pig model, we demonstrate using novel high-resolution neural electrode arrays that increased intrasegmental network coherence within the thoracic spinal cord contributes to myocardial ischemia-induced sympathoexcitation.
High comorbidity among pediatric disorders characterized by behavioral and emotional dysregulation poses problems for diagnosis and treatment, and suggests that these disorders may be better ...conceptualized as dimensions of abnormal behaviors. Furthermore, identifying neuroimaging biomarkers related to dimensional measures of behavior may provide targets to guide individualized treatment. We aimed to use functional neuroimaging and pattern regression techniques to determine whether patterns of brain activity could accurately decode individual-level severity on a dimensional scale measuring behavioural and emotional dysregulation at two different time points.
A sample of fifty-seven youth (mean age: 14.5 years; 32 males) was selected from a multi-site study of youth with parent-reported behavioral and emotional dysregulation. Participants performed a block-design reward paradigm during functional Magnetic Resonance Imaging (fMRI). Pattern regression analyses consisted of Relevance Vector Regression (RVR) and two cross-validation strategies implemented in the Pattern Recognition for Neuroimaging toolbox (PRoNTo). Medication was treated as a binary confounding variable. Decoded and actual clinical scores were compared using Pearson's correlation coefficient (r) and mean squared error (MSE) to evaluate the models. Permutation test was applied to estimate significance levels.
Relevance Vector Regression identified patterns of neural activity associated with symptoms of behavioral and emotional dysregulation at the initial study screen and close to the fMRI scanning session. The correlation and the mean squared error between actual and decoded symptoms were significant at the initial study screen and close to the fMRI scanning session. However, after controlling for potential medication effects, results remained significant only for decoding symptoms at the initial study screen. Neural regions with the highest contribution to the pattern regression model included cerebellum, sensory-motor and fronto-limbic areas.
The combination of pattern regression models and neuroimaging can help to determine the severity of behavioral and emotional dysregulation in youth at different time points.
Full text
Available for:
DOBA, IZUM, KILJ, NUK, PILJ, PNG, SAZU, SIK, UILJ, UKNU, UL, UM, UPUK
Regular communication between technologists and radiologists is necessary for maintaining optimal diagnostic image quality throughout a radiology practice. In a large hospital system with multiple ...sites, this task becomes increasingly difficult without simultaneously causing significant disruptions in the clinical workflow and decreased throughput. Thus, establishing a system for quality control reporting that enables effective communication in a seamless and convenient manner is imperative. In this report, we describe the development of a new integrated system, in collaboration with our PACS vendor, with tools that allow for instant reporting of quality errors and dashboards providing real-time up-to-date quality data across our hospital system, directly accessible from PACS. To date, 8,167 quality reports have been logged in our new system with roughly 355 submissions per month. Early user engagement and consensus feedback among radiologists and technologists have been positive suggesting an overall improvement from prior systems. We hope this report can help inform other radiology enterprises seeking to improve quality control reporting within their clinical practice.
Full text
Available for:
NUK, SBMB, SBNM, UL, UM, UPUK, VSZLJ