Search

Electronic medical record (EMR) systems provide easy access to radiology reports and offer great potential to support quality improvement efforts and clinical research. Harnessing the full potential of the EMR requires scalable approaches such as natural language processing (NLP) to convert text into variables used for evaluation or analysis. Our goal was to determine the feasibility of using NLP to identify patients with Type 1 Modic endplate changes using clinical reports of magnetic resonance (MR) imaging examinations of the spine. Identifying patients with Type 1 Modic change who may be eligible for clinical trials is important as these findings may be important targets for intervention. Four annotators identified all reports that contained Type 1 Modic change, using N = 458 randomly selected lumbar spine MR reports. We then implemented a rule-based NLP algorithm in Java using regular expressions. The prevalence of Type 1 Modic change in the annotated dataset was 10%. Results were recall (sensitivity) 35/50 = 0.70 (95% confidence interval (C.I.) 0.52–0.82), specificity 404/408 = 0.99 (0.97–1.0), precision (positive predictive value) 35/39 = 0.90 (0.75–0.97), negative predictive value 404/419 = 0.96 (0.94–0.98), and F1-score 0.79 (0.43–1.0). Our evaluation shows the efficacy of rule-based NLP approach for identifying patients with Type 1 Modic change if the emphasis is on identifying only relevant cases with low concern regarding false negatives. As expected, our results show that specificity is higher than recall. This is due to the inherent difficulty of eliciting all possible keywords given the enormous variability of lumbar spine reporting, which decreases recall, while availability of good negation algorithms improves specificity. [ABSTRACT FROM AUTHOR]

Background and Purpose: Recurrent or residual back pain after vertebroplasty (VP) is common, and many patients with these symptoms are evaluated by MRI. The purpose of this report is to describe post-treatment MRI findings after successful VP., Methods: We identified all patients who underwent VP at our institution and later presented with back pain and had a spine MRI. From these patients, we identified a cohort with uncomplicated procedures, in whom back pain at the site of the VP was not the dominant pain source at follow-up. Using the pre- and postoperative MRIs and the VP conventional radiographs, we assessed such features as interval height loss and changes in marrow edema in treated vertebrae., Results: Thirty patients (51 treated vertebrae) met our clinical criteria for uncomplicated VP. Nine (18%) of 51 vertebrae lost additional height after the procedure. Neither patient demographics nor variables associated with the procedure itself, including volume of cement injected, correlated with additional height loss after VP. Moderate or severe marrow edema was demonstrated in 62% of vertebrae on preoperative MRI and in 33% of vertebrae on follow-up MRI. Twenty-one percent of vertebrae had new areas of marrow edema on follow-up. Twenty-two percent of vertebrae imaged >6 months after VP had moderate or severe edema., Conclusion: Progressive and persistent edema and interval height loss after successful VP are common and should not be interpreted as sufficient evidence of ongoing pathology at the treated vertebral level.

Background and Purpose: The detection and localization of spinal dural arteriovenous fistulas (AVFs) remain diagnostic challenges. This study tested the hypothesis that elliptic centric contrast-enhanced MR angiography (MRA) can be used to detect spinal dural AVFs, predict the level of fistulas, and reduce the radiation dose and volume of iodinated contrast material associated with conventional angiography., Methods: We examined 31 patients who presented with suspected spinal dural AVF between December 2000 and March 2004. All patients underwent MRA and conventional angiography. The effect of MRA on subsequent conventional angiography was assessed by analyzing total fluoroscopy time and volume of iodinated contrast material used., Results: At angiography, spinal dural AVFs were diagnosed in 22 of 31 patients, and MRA depicted an AVF in 20 of the 22 patients. MRA findings correctly predicted a negative angiogram in seven of nine cases. Of the 20 true-positive MRA results, the level of the fistula was included in the imaging volume in 14. In 13 of these 14 cases, MRA results correctly predicted the side and the level of the fistula to within one vertebral level. Fluoroscopy time and the volume of contrast agent was reduced by more than 50% in the 13 patients with a spinal dural AVF in whom MRA prospectively indicated the correct level., Conclusion: Contrast-enhanced MRA can be used to detect spinal dural AVFs, predict the level of fistulas, and substantially reduce the radiation dose and volume of contrast agent associated with catheter spinal angiography.

In some patients with spontaneous spinal CSF leaks, leaks are numerous or tears are so large that extrathecal myelographic contrast material is seen at multiple levels during CT, making identification of their source impossible. This study introduces a dynamic CT myelographic technique that provides high temporal and spatial resolution. In this technical note, we describe the utility of this technique in four patients with challenging high-flow spinal CSF leaks.

Background and Purpose: The recently described posterior reversible encephalopathy syndrome (PRES) classically consists of reversible vasogenic edema in the posterior circulation territories, although conversion to irreversible cytotoxic edema has been described. We hypothesized that the extent of edema has prognostic implications and that diffusion-weighted MR imaging (DWI) can help predict the progression to infarction., Methods: Twenty-two patients with PRES and 18 control subjects were examined with isotropic DWI. Nineteen regions of interest (ROIs) were systematically placed, and apparent diffusion coefficients (ADCs) were computed and correlated with T2 and DWI signal intensity in each ROI., Results: T2 signal abnormalities were always present in territories of the posterior circulation. Anterior circulation structures were involved in 91% of patients. ADC values in areas of abnormal T2 signal were high. More extensive T2 signal abnormalities were seen in patients with a poor outcome than in patients who recovered. In six patients (27%), areas of high DWI signal intensity were seen with ADC values that were paradoxically normal, which we called pseudonormalized. Abnormal T2 signal intensity and high ADC values surrounded these areas. Follow-up images in two patients showed progression to infarction in pseudonormalized regions., Conclusion: Vasogenic edema in PRES involves predominantly the posterior circulation territories, but anterior circulation structures are also frequently involved. The extent of combined T2 and DWI signal abnormalities correlate with patient outcome. High DWI signal intensity and pseudonormalized ADC values are associated with cerebral infarction and may represent the earliest sign of nonreversibility as severe vasogenic edema progresses to cytotoxic edema.