Magnetic Resonance Imaging Reports Research Articles

Survival in dogs with meningoencephalomyelitis of unknown etiology with and without lesions detected by magnetic resonance imaging.

The prognosis of individual dogs with meningoencephalomyelitis of unknown etiology (MUE) remains difficult to predict. MUE cases with no lesions detected by magnetic resonance imaging (MRI) occur, but it is unknown whether this finding is associated with prognosis. MUE cases without detectable lesions on MRI have a better outcome than cases with detectable lesions. Study included 73 client-owned dogs with MUE presenting to Purdue University Veterinary Hospital from 2010 to 2020. Retrospective study. Dogs with a clinical diagnosis of MUE were identified by medical record search. MRI reports were reviewed for presence or absence of lesions consistent with MUE. Clinical findings at presentation, treatment, disease-specific survival, and outcomes including rates of remission and relapse were compared between cases with normal MRI or abnormal MRI. Overall, 54 dogs (74%) were classified as abnormal MRI, and 19 dogs (26%) were classified as normal MRI cases. Death caused by MUE occurred in 1/19 (5%) normal MRI dogs and 18/54 (33%) abnormal MRI dogs (P = .016). Median survival was >107 months in both groups, but survival was significantly longer in the normal MRI group (P = .019). On multivariate analysis, abnormal MRI was significantly related to death (hazard ratio, 7.71; 95% confidence interval 1.03-58.00, P = .0470), whereas significant relationships with death were not identified for either the use of secondary immunosuppressive medications or cerebrospinal fluid nucleated cell count. MUE dogs with no detectable lesions on MRI have reduced disease-related death compared with dogs with abnormal MRI. The presence or absence of MRI lesions in MUE dogs is prognostically relevant.

Unravelling pituitary tumours in medically treated patients with acromegaly: the impact of systematic MRI reassessment.

In acromegaly, the primary tumor is usually found during magnetic resonance imaging (MRI) of the pituitary gland. A remnant tumor after surgery is, however, harder to depict. When a tumor is missed, the remaining option is usually lifelong pharmacological treatment. To identify tumors by reassessment of all available MRI scans in pharmacologically treated patients, operated or not, and to compare our results with the routine MRI reports. Adult patients diagnosed with acromegaly and managed at a tertiary care center between 2005 and 2021 and currently on pharmacological treatment were included. MRI scans were evaluated in a standardized manner and classified independently by a radiologist and an endocrinologist into "certain," "suspected," or "no tumor." In case of disagreement, consensus was achieved with a senior neuroradiologist. The results were compared using the clinical radiologists' routine MRI reports. We identified certain and suspected tumors in 29/74 and 36/74 patients, respectively. No tumor was identified in nine patients. In five of these, no MRI contrast agent was given. Discrepancy between our results and the routine MRI reports was found in 31/74 patients (P = 0.01). In 22 patients, the routine reports described no tumor while we identified certain tumors in 2/22 patients and suspected tumors in 13/22 patients. In most patients with pharmacologically treated acromegaly, we identified a certain or suspected pituitary tumor. These findings were more frequent compared to the routine MRI reports. Based on our results, patients will be considered for a change in long-term treatment modality.

Using GPT-4 for LI-RADS feature extraction and categorization with multilingual free-text reports.

The Liver Imaging Reporting and Data System (LI-RADS) offers a standardized approach for imaging hepatocellular carcinoma. However, the diverse styles and structures of radiology reports complicate automatic data extraction. Large language models hold the potential for structured data extraction from free-text reports. Our objective was to evaluate the performance of Generative Pre-trained Transformer (GPT)-4 in extracting LI-RADS features and categories from free-text liver magnetic resonance imaging (MRI) reports. Three radiologists generated 160 fictitious free-text liver MRI reports written in Korean and English, simulating real-world practice. Of these, 20 were used for prompt engineering, and 140 formed the internal test cohort. Seventy-two genuine reports, authored by 17 radiologists were collected and de-identified for the external test cohort. LI-RADS features were extracted using GPT-4, with a Python script calculating categories. Accuracies in each test cohort were compared. On the external test, the accuracy for the extraction of major LI-RADS features, which encompass size, nonrim arterial phase hyperenhancement, nonperipheral 'washout', enhancing 'capsule' and threshold growth, ranged from .92 to .99. For the rest of the LI-RADS features, the accuracy ranged from .86 to .97. For the LI-RADS category, the model showed an accuracy of .85 (95% CI: .76, .93). GPT-4 shows promise in extracting LI-RADS features, yet further refinement of its prompting strategy and advancements in its neural network architecture are crucial for reliable use in processing complex real-world MRI reports.

Axial Spondyloarthritis: Does Magnetic Resonance Imaging Classification Improve Report Interpretation.

The interpretation of magnetic resonance imaging (MRI) reports is crucial for the diagnosis of axial spondyloarthritis, but the subjective nature of narrative reports can lead to varying interpretations. This study presents a validation of a novel MRI reporting system for the sacroiliac joint in clinical practice. A historical review was conducted on 130 consecutive patients referred by 2 rheumatologists for initial MRI assessment of possible axial spondyloarthritis. The original MRI reports were interpreted by the rheumatologists and the radiologist who originally read the images and then categorized according to the novel system. Two musculoskeletal radiologists then reinterpreted the original MRI scans using the new system, and the resulting reports were interpreted and categorized by the same rheumatologists. The quality of the new framework was assessed by comparing the interpretations of both reports. Ninety-two patients met the study criteria. The rheumatologists disagreed on the categorization of the original MRI reports in 12% of cases. The rheumatologists and original radiologists disagreed on the categorization of the initial report in 23.4% of cases. In contrast, there was 100% agreement between the rheumatologists and radiologists on the categorization of the new MRI report. The new MRI categorization system significantly improved the agreement between the clinician and radiologist in report interpretation. The system provided a standard vocabulary for reporting, reduced variability in report interpretation, and may therefore improve clinical decision-making.

From technical to understandable: Artificial Intelligence Large Language Models improve the readability of knee radiology reports.

The purpose of this study was to evaluate the effectiveness of an Artificial Intelligence-Large Language Model (AI-LLM) at improving the readability of knee radiology reports. Reports of 100 knee X-rays, 100 knee computed tomography (CT) scans and 100 knee magnetic resonance imaging (MRI) scans were retrieved. The following prompt command was inserted into the AI-LLM: 'Explain this radiology report to a patient in layman's terms in the second person:[Report Text]'. The Flesch-Kincaid reading level (FKRL) score, Flesch reading ease (FRE) score and report length were calculated for the original radiology report and the AI-LLM generated report. Any 'hallucination' or inaccurate text produced by the AI-LLM-generated report was documented. Statistically significant improvements in mean FKRL scores in the AI-LLM generated X-ray report (12.7 ± 1.0-7.2 ± 0.6), CT report (13.4 ± 1.0-7.5 ± 0.5) and MRI report (13.5 ± 0.9-7.5 ± 0.6) were observed. Statistically significant improvements in mean FRE scores in the AI-LLM generated X-ray report (39.5 ± 7.5-76.8 ± 5.1), CT report (27.3 ± 5.9-73.1 ± 5.6) and MRI report (26.8 ± 6.4-73.4 ± 5.0) were observed. Superior FKRL scores and FRE scores were observed in the AI-LLM-generated X-ray report compared to the AI-LLM-generated CT report and MRI report, p < 0.001. The hallucination rates in the AI-LLM generated X-ray report, CT report and MRI report were 2%, 5% and 5%, respectively. This study highlights the promising use of AI-LLMs as an innovative, patient-centred strategy to improve the readability of knee radiology reports. The clinical relevance of this study is that an AI-LLM-generated knee radiology report may enhance patients' understanding of their imaging reports, potentially reducing the responder burden placed on the ordering physicians. However, due to the 'hallucinations' produced by the AI-LLM-generated report, the ordering physician must always engage in a collaborative discussion with the patient regarding both reports and the corresponding images. Level IV.

Radiology Reports Do Not Accurately Portray the Severity of Cervical Neural Foraminal Stenosis.

Retrospective study. (1) To compare cervical magnetic resonance imaging (MRI) radiology reports to a validated grading system for cervical foraminal stenosis (FS) and (2) to evaluate whether the severity of cervical neural FS on MRI correlates to motor weakness or patient-reported outcomes. Radiology reports of cervical spine MRI are often reviewed to assess the degree of neural FS. However, research looking at the association between these reports and objective MRI findings, as well as clinical symptoms, is lacking. We retrospectively identified all adult patients undergoing primary 1 or 2-level anterior cervical discectomy and fusion at a single academic center for an indication of cervical radiculopathy. Preoperative MRI was assessed for neural FS severity using the grading system described by Kim and colleagues for each level of fusion, as well as adjacent levels. Neural FS severity was recorded from diagnostic radiologist MRI reports. Motor weakness was defined as an examination grade <4/5 on the final preoperative encounter. Regression analysis was conducted to evaluate whether the degree of FS by either classification was related to patient-reported outcome measure severity. A total of 283 patients were included in the study, and 998 total levels were assessed. There were significant differences between the MRI grading system and the assessment by radio-logists (P< 0.001). In levels with moderate stenosis, 28.9% were classified as having no stenosis by radiology. In levels with severe stenosis, 29.7% were classified as having mild-moderate stenosis or less. Motor weakness was found similarly often in levels of moderate or severe stenosis (6.9% and 9.2%, respectively). On regression analysis, no associations were found between baseline patient-reported outcome measures and stenosis severity assessed by radiologists or MRI grading systems. Radiology reports on the severity of cervical neural FS are not consistent with a validated MRI grading system. These radiology reports underestimated the severity of neural foraminal compression and may be inappropriate when used for clinical decision-making. Level III.

A Dual-Gene Reporter-Amplifier Architecture for Enhancing the Sensitivity of Molecular MRI by Water Exchange.

The development of genetic reporters for magnetic resonance imaging (MRI) is essential for investigating biological functions in vivo. However, current MRI reporters have low sensitivity, making it challenging to create significant contrast against the tissue background, especially when only a small fraction of cells express the reporter. To overcome this limitation, we developed an approach for amplifying the sensitivity of molecular MRI by combining a chemogenetic contrast mechanism with a biophysical approach to increase water diffusion through the co-expression of a dual-gene construct comprising an organic anion transporting polypeptide, Oatp1b3, and a water channel, Aqp1. We first show that the expression of Aqp1 amplifies MRI contrast in cultured cells engineered to express Oatp1b3. We demonstrate that the contrast amplification is caused by Aqp1-driven increase in water exchange, which provides the gadolinium ions internalized by Oatp1b3-expressing cells with access to a larger water pool compared with exchange-limited conditions. We further show that our methodology allows cells to be detected using approximately 10-fold lower concentrations of gadolinium than that in the Aqp1-free scenario. Finally, we show that our approach enables the imaging of mixed-cell cultures containing a low fraction of Oatp1b3-labeled cells that are undetectable on the basis of Oatp1b3 expression alone.

Deep learning segmentation of the choroid plexus from structural magnetic resonance imaging (MRI): validation and normative ranges across the adult lifespan

BackgroundThe choroid plexus functions as the blood-cerebrospinal fluid (CSF) barrier, plays an important role in CSF production and circulation, and has gained increased attention in light of the recent elucidation of CSF circulation dysfunction in neurodegenerative conditions. However, methods for routinely quantifying choroid plexus volume are suboptimal and require technical improvements and validation. Here, we propose three deep learning models that can segment the choroid plexus from commonly-acquired anatomical MRI data and report performance metrics and changes across the adult lifespan.MethodsFully convolutional neural networks were trained from 3D T1-weighted, 3D T2-weighted, and 2D T2-weighted FLAIR MRI using gold-standard manual segmentations in control and neurodegenerative participants across the lifespan (n = 50; age = 21–85 years). Dice coefficients, 95% Hausdorff distances, and area-under-curve (AUCs) were calculated for each model and compared to segmentations from FreeSurfer using two-tailed Wilcoxon tests (significance criteria: p < 0.05 after false discovery rate multiple comparisons correction). Metrics were regressed against lateral ventricular volume using generalized linear models to assess model performance for varying levels of atrophy. Finally, models were applied to an expanded cohort of adult controls (n = 98; age = 21–89 years) to provide an exemplar of choroid plexus volumetry values across the lifespan.ResultsDeep learning results yielded Dice coefficient = 0.72, Hausdorff distance = 1.97 mm, AUC = 0.87 for T1-weighted MRI, Dice coefficient = 0.72, Hausdorff distance = 2.22 mm, AUC = 0.87 for T2-weighted MRI, and Dice coefficient = 0.74, Hausdorff distance = 1.69 mm, AUC = 0.87 for T2-weighted FLAIR MRI; values did not differ significantly between MRI sequences and were statistically improved compared to current commercially-available algorithms (p < 0.001). The intraclass coefficients were 0.95, 0.95, and 0.96 between T1-weighted and T2-weighted FLAIR, T1-weighted and T2-weighted, and T2-weighted and T2-weighted FLAIR models, respectively. Mean lateral ventricle choroid plexus volume across all participants was 3.20 ± 1.4 cm3; a significant, positive relationship (R2 = 0.54-0.60) was observed between participant age and choroid plexus volume for all MRI sequences (p < 0.001).ConclusionsFindings support comparable performance in choroid plexus delineation between standard, clinically available, non-contrasted anatomical MRI sequences. The software embedding the evaluated models is freely available online and should provide a useful tool for the growing number of studies that desire to quantitatively evaluate choroid plexus structure and function (https://github.com/hettk/chp_seg).

Age, Motion, Medical, and Psychiatric Associations With Incidental Findings in Brain MRI

Few investigations have evaluated rates of brain-based magnetic resonance imaging (MRI) incidental findings (IFs) in large lifespan samples, their stability over time, or their associations with health outcomes. To examine rates of brain-based IFs across the lifespan, their persistence, and their associations with phenotypic indicators of behavior, cognition, and health; to compare quantified motion with radiologist-reported motion and evaluate its associations with IF rates; and to explore IF consistency across multiple visits. This cross-sectional study included participants from the Nathan Kline Institute-Rockland Sample (NKI-RS), a lifespan community-ascertained sample, and the Healthy Brain Network (HBN), a cross-sectional community self-referred pediatric sample focused on mental health and learning disorders. The NKI-RS enrolled participants (ages 6-85 years) between March 2012 and March 2020 and had longitudinal participants followed up for as long as 4 years. The HBN enrolled participants (ages 5-21 years) between August 2015 and October 2021. Clinical neuroradiology MRI reports were coded for radiologist-reported motion as well as presence, type, and clinical urgency (category 1, no abnormal findings; 2, no referral recommended; 3, consider referral; and 4, immediate referral) of IFs. MRI reports were coded from June to October 2021. Data were analyzed from November 2021 to February 2023. Rates and type of IFs by demographic characteristics, health phenotyping, and motion artifacts; longitudinal stability of IFs; and Euler number in projecting radiologist-reported motion. A total of 1300 NKI-RS participants (781 [60.1%] female; mean [SD] age, 38.9 [21.8] years) and 2772 HBN participants (976 [35.2%] female; mean [SD] age, 10.0 [3.5] years) had health phenotyping and neuroradiology-reviewed MRI scans. IFs were common, with 284 of 2956 children (9.6%) and 608 of 1107 adults (54.9%) having IFs, but rarely of clinical concern (category 1: NKI-RS, 619 [47.6%]; HBN, 2561 [92.4%]; category 2: NKI-RS, 647 [49.8%]; HBN, 178 [6.4%]; category 3: NKI-RS, 79 [6.1%]; HBN, 30 [1.1%]; category 4: NKI-RS: 12 [0.9%]; HBN, 6 [0.2%]). Overall, 46 children (1.6%) and 79 adults (7.1%) required referral for their IFs. IF frequency increased with age. Elevated blood pressure and BMI were associated with increased T2 hyperintensities and age-related cortical atrophy. Radiologist-reported motion aligned with Euler-quantified motion, but neither were associated with IF rates. In this cross-sectional study, IFs were common, particularly with increasing age, although rarely clinically significant. While T2 hyperintensity and age-related cortical atrophy were associated with BMI and blood pressure, IFs were not associated with other behavioral, cognitive, and health phenotyping. Motion may not limit clinical IF detection.

MRI-Based Kinetic Heterogeneity Evaluation in the Accurate Access of Axillary Lymph Node Status in Breast Cancer Using a Hybrid CNN-RNN Model.

Accurate evaluation of the axillary lymph node (ALN) status is needed for determining the treatment protocol for breast cancer (BC). The value of magnetic resonance imaging (MRI)-based tumor heterogeneity in assessing ALN metastasis in BC is unclear. To assess the value of deep learning (DL)-derived kinetic heterogeneity parameters based on BC dynamic contrast-enhanced (DCE)-MRI to infer the ALN status. Retrospective. 1256/539/153/115 patients in the training cohort, internal validation cohort, and external validation cohorts I and II, respectively. 1.5 T/3.0 T, non-contrast T1-weighted spin-echo sequence imaging (T1WI), DCE-T1WI, and diffusion-weighted imaging. Clinical pathological and MRI semantic features were obtained by reviewing histopathology and MRI reports. The segmentation of the tumor lesion on the first phase of T1WI DCE-MRI images was applied to other phases after registration. A DL architecture termed convolutional recurrent neural network (ConvRNN) was developed to generate the KHimage (kinetic heterogeneity of DCE-MRI image) score that indicated the ALN status in patients with BC. The model was trained and optimized on training and internal validation cohorts, tested on two external validation cohorts. We compared ConvRNN model with other 10 models and the subgroup analyses of tumor size, magnetic field strength, and molecular subtype were also evaluated. Chi-squared, Fisher's exact, Student's t, Mann-Whitney U tests, and receiver operating characteristics (ROC) analysis were performed. P < 0.05 was considered significant. The ConvRNN model achieved area under the curve (AUC) of 0.802 in the internal validation cohort and 0.785-0.806 in the external validation cohorts. The ConvRNN model could well evaluate the ALN status of the four molecular subtypes (AUC = 0.685-0.868). The patients with larger tumor sizes (>5 cm) were more susceptible to ALN metastasis with KHimage scores of 0.527-0.827. A ConvRNN model outperformed traditional models for determining the ALN status in patients with BC. 3 TECHNICAL EFFICACY: Stage 2.

Are clinically unimportant findings qualified as benign in lumbar spine imaging reports? A content analysis of plain X-ray, CT and MRI reports.

Lumbar spine diagnostic imaging reports may cause patient and clinician concern when clinically unimportant findings are not explicitly described as benign. Our primary aim was to determine the frequency that common, benign findings are reported in lumbar spine plain X-ray, computed tomography (CT) and magnetic resonance imaging (MRI) reports as either normal for age or likely clinically unimportant. We obtained 600 random de-identified adult lumbar spine imaging reports (200 X-ray, 200 CT and 200 MRI) from a large radiology provider. Only reports requested for low back pain were included. From the report text, one author extracted each finding (e.g., 'broad-based posterior disc bulge') and whether it was present or absent (e.g., no disc bulge) until data saturation was reached, pre-defined as a minimum of 50 reports and no new/similar findings in the last ten reports within each imaging modality. Two authors independently judged whether each finding was likely clinically unimportant or important. For each likely clinicially unimportant finding they also determined if it had been explicitly reported to be benign (expressed as normal, normal for age, benign, clinically unimportant or non-significant). Data saturation was reached after coding 262 reports (80 X-ray, 82 CT, 100 MRI). Across all reports we extracted 3,598 findings. Nearly all reports included at least one clinically unimportant finding (76/80 (95%) X-ray, 80/82 (98%) CT, 99/100 (99%) MRI). Over half of the findings (n = 2,062, 57%; 272 X-Ray, 667 CT, 1123 MRI) were judged likely clinically unimportant. Most likely clinically unimportant findings (90%, n = 1,854) were reported to be present on imaging (rather than absent) and of those only 18% (n = 331) (89 (35%) X-ray, 93 (16%) CT and 149 (15%) MRI) were explicitly reported as benign. Lumbar spine imaging reports frequently include findings unlikely to be clinically important without explicitly qualifying that they are benign.

Impact of the 2014 NHLBI Hydroxyurea Guidelines on Cerebrovascular Events in Sickle Cell Disease

Background: Cerebrovascular disease (CVD) is a common cause of morbidity in patients with sickle cell disease (SCD). An estimated 37% of children with SCD have silent cerebral infarcts (SCI) by age 14, leading to increased stroke risk and neurocognitive impairment. In the DISPLACE trial (Kanter et al, 2021), 2.9% of patients had abnormal cerebral blood flow as measured on transcranial Doppler ultrasound (TCD). The 2014 National Heart, Lung, and Blood Institute (NHLBI) guidelines recommend hydroxyurea (HU) for all patients with SCD-SS and SCD-Sβ 0 starting at age 9 months, a significant shift from prior practice. In patients who have already developed TCD abnormalities, HU is protective against stroke and CVD progression. The real-world impact of universal early HU on the development of CVD, particularly on SCI, is not well defined . We hypothesize that the 2014 guidelines led to an increase in HU use and a reduction in the initial development of CVD. Methods: We performed a single center, retrospective cohort study investigating trends in HU prescriptions and CVD before and after the publication of the 2014 NHLBI guidelines. Patients with SCD-SS and SCD-Sβ 0 who initiated hematology care at the Children's Hospital of Philadelphia between 2009 and 2023 were identified using an institutional registry. Incident HU was defined as the first outpatient HU prescription in patients with a minimum of two orders. Incident CVD was defined as a composite outcome consisting of the first documentation of: abnormal TCD, brain magnetic resonance imaging (MRI) with silent infarct, or overt ischemic stroke. Patients with known CVD before care initiation, prematurity &amp;lt; 28 weeks gestational age, or non-SCD neurologic disease were excluded. For time-to-HU analysis, patients with HU prescriptions before enrollment at our center also were excluded. Patients were followed from their first hematology visit until the outcome of interest or censoring. Censoring criteria included loss to follow up, transition of care, or bone marrow transplant. All brain MRI and TCD reports were manually reviewed. Ages at HU initiation pre- and post- guideline were compared using the Wilcoxon rank-sum test . Times to 1) HU and 2) CVD were summarized using Kaplan-Meier estimators, with the guideline publication date of September 1, 2014 incorporated as a time-varying exposure. Unadjusted associations were calculated using a Cox proportional hazard regression. Results: A total of 446 patients met inclusion criteria and contributed to analyses. Median age at baseline was 2.4 months, and median duration of follow-up was 5.0 years (IQR: 2.3-8.7). Most patients (430/446, 96%) had SCD SS, and 51% were male. The proportion of patients ever prescribed HU increased with time (Fig 1), with a peak prevalence of 82% in 2022. Of 385 patients not exposed to HU at baseline, 277 (72%) were prescribed HU during study follow-up. Incident HU prescriptions significantly increased after NHLBI guideline publication (HR 3.40, CI: 2.50 to 4.63, p &amp;lt; 0.001) with an older age at HU initiation before (median 4.0 y, IQR 2.5-7.7) than after (median 2.9 y, IQR 1.1-5.8) publication of the guideline (p = 0.002). Over a total of 2,395 person-years of follow-up, 66 incident CVD outcomes occurred in this longitudinal cohort, corresponding to a cumulative incidence of 14.8%. These events included 53 SCI, 10 abnormal TCD, and 3 overt ischemic strokes. CVD outcome occurrence was lower after guideline publication than before (HR 0.58, CI: 0.32-1.05), though the association did not reach statistical significance (p = 0.071, Fig. 2). Discussion: After the NHLBI guideline publication, HU prescriptions increased significantly at our center.Our data suggest that the 2014 guideline-directed care may have led to a greater than 40% reduction in CVD, though this reduction was not statistically significant in our small sample size. Despite guideline recommendations and strong evidence for the role of HU in reducing pain and acute chest syndrome, few patients across the United States are prescribed HU therapy. Adherence and care continuity challenges further lessen the number of patients benefitting from HU. Evidence of efficacy of early and universal HU in CVD prevention would strengthen guideline recommendations and improve clinical care in patients with SCD. Ongoing analysis includes an expanded cohort and adjustment for clinical covariates to better examine the direct impact of early HU on CVD.

Survey of rectal cancer MRI technique and reporting tumour descriptors in the UK: a multi-centre British Society of Gastrointestinal and Abdominal Radiology (BSGAR) audit

To evaluate variation in magnetic resonance imaging (MRI) technique and reporting of rectal cancer staging examinations across the UK. A retrospective, multi-centre audit was undertaken of imaging protocols and information documented within consecutive MRI rectal cancer reports between March 2020 and August 2021, which were compared against American and European guidelines. Inclusion criteria included histologically proven rectal adenocarcinoma and baseline staging MRI rectum only. Fully anonymised data from 924 MRI reports by 78 radiologists at 24 centres were evaluated. Thirty-two per cent of radiologists used template reporting, but these reports offered superior documentation of 13 out of 18 key tumour features compared to free-text reports including T-stage, relation to peritoneal reflection and mesorectal fascia (MRF), nodal status, and presence of extramural venous invasion (EMVI; p<0.027 in each). There was no significant differences in the remaining five features. Across all tumour locations, the tumour relationship to the MRF, the presence of EMVI, and the presence of tumour deposits were reported in 79.5%, 85.6%, and 44% of cases, respectively, and tumour, nodal, and distant metastatic stage documented in 94.4%, 97.7%, and 78.3%. In low rectal tumours, the relationship to the anal sphincter complex was reported in only 54.6%. Considerable variation exists in rectal cancer MRI acquisition and reporting in this sample of UK centres. Inclusion of key radiological features in reports must be improved for risk stratification and treatment decisions. Template reporting is superior to free-text reporting. Routine adoption of standardised radiology practices should now be considered to improve standards to facilitate personalised precision treatment for patients to improve outcomes.

Application of natural language processing to post-structuring of rectal cancer MRI reports

To evaluate a natural language processing (NLP) system for extracting structured information from the free-form text of rectal cancer magnetic resonance imaging (MRI) reports written in Chinese. A rule-based NLP model that could extract 11 key image features of rectal cancer was constructed using 358 MRI reports of rectal cancer written between 2015 and 2021. Fifty reports written before 2015 and 50 written after 2021 were used as test datasets, and the reference standard was determined by manual extraction of information by two radiologists. The length and reporting rate of image features in pre-2015 and post-2021 datasets, as well as the accuracy, precision, recall, and F1 score of feature extraction by the NLP system, were compared. The time required for the NLP to extract data was compared with that required by the radiologists. Reports written after 2021 had longer diagnostic impression sections than reports written before 2015. The reporting rate of key imaging features of rectal cancer was 36.55% before 2015 and 79.82% after 2021. The accuracy, precision, recall, and F1 score of NLP for correct extraction of values from reports were 93.82%, 95.63%, 87.06%, and 91.15%, respectively, for pre-2015 reports, and 92.55%, 98.53%, 94.15%, and 96.29%, respectively, for post-2021 reports. NLP generated all the structured information in <1 second. The NLP system with rule-based pattern matching achieved rapid and accurate structured processing of rectal cancer MRI reports. MRI reports with structured templates are more suitable for NLP-based extraction of information.

Clinical Value and Operational Risks of MRI in ICU patients - ARetrospective Analysis Performed at a University Medical Center.

Intensive care unit (ICU) patients have a high risk of developing complications when leaving the ICU for diagnostic procedures or therapeutic interventions. Our study examined the frequency of adverse events associated with magnetic resonance imaging (MRI) of intensive care patients and the extent of changes in therapy due to the MRI scan to weigh the risks associated with the scan against the potential benefits of an MR scan, using a change in therapy as an indicator of benefit. 4434 ICU patients (January to December 2015) were identified by Hospital Information System (SAP-R/3 IS-H, Walldorf, Germany), ICU patient data management system Metavision (iMDsoft, Israel), and Radiology Information System (Nexus.medRIS, Version 8.42, Nexus, Germany). All intensive care and medical records (HIS) and MRI reports (RIS) were matched and further evaluated in a retrospective case-to-case analysis for biometric data, mechanical ventilation, ICU requirements, planned postoperative vs. emergency diagnostic requirements, complications and impact on further diagnosis or therapy. Out of 4434 ICU patients, 322 ICU patients (7.3 %) underwent a total of 385 MRI examinations. 167 patients needed a total of 215 emergency scans, while 155 patients underwent 170 planned postoperative MRI exams. 158 (94.6 %) out of 167 emergency scan patients were ventilated under continuous intravenous medication and monitoring. In the planned postoperative group, only 6 (3.9 %) out of 155 were ventilated, but a total of 38 (24.5 %) were under continuous medication. 111 patients were accompanied by nurses only during MRI. Only one severe adverse event (0.3 %) was noted and was attributed to study preparation (n = 385). In 8MRI examinations (2.1 %), the examination was interrupted or cancelled due to the patients' condition. While all MRI examinations in the planned group were completed (n = 170, 100 %) (e. g., postoperative controls), only 207 out of 215 (96.3 %) could be performed for emergency diagnostic reasons. MRI influenced the clinical course with a change in diagnosis or therapy in 74 (19.2 %) of all 385 MRI examinations performed, and in the emergency diagnostic group it was 31.2 % (n = 67/215). Nearly 20 % of MRI examinations of ICU patients resulted in a change of therapy. With only one potentially life-threatening adverse event (0.3 %) during transport and the MRI examination, the risk seems to be outweighed by the diagnostic benefit. · The risk of adverse events associated with MRI scans in ICU patients is low.. · The rate of premature termination of ICU patients' MRI scans is low.. · Almost 20 % of ICU patients' MRI scans lead to a change of therapy..

Abstract 17893: Deep Learning for the Prediction of Myocardial Scarring Through Electrocardiograms

Introduction: Cardiac Magnetic Resonance Imaging (MRI) with Late Gadolinium enhancement (LGE), the gold standard for myocardial scarring evaluation, is limited by cost, time, and expertise required. The Electrocardiogram (ECG), in contrast, is a widely available and affordable diagnostic tool. We applied deep learning (DL) models to ECGs to directly predict the presence of LGE on MRI. Methods: Patients who underwent a cardiac MRI and had at least one ECG collected from 2012 to 2022 were considered. Patients were classified into positive and negative for myocardial scarring via their MRI reports, labeled by experts. Patients with non-specific scarring were excluded from the dataset. Patient ECGs most recent to their MRI were utilized as input for the model. MRI reports and ECG waveforms were collected from NYU Langone Health. We implemented a 34-layer Neural Network architecture, adapted from a state-of-the-art arrhythmia detection model, to learn from a 10 second ECG sampled at 250hz. We evaluated the model using area under the receiver-operating curve (AUROC), specificity, sensitivity, and precision. Results: 8813 patients with MRI and ECG were included. 1559 were excluded due to non-specific scarring. 2739 patients were positive for scarring. Patients were split into a 70% train, 10% validation, and 20% test dataset. The deep learning model obtained a test set AUROC of 0.80 [95%CI, 0.77 - 0.81]. At 20% false positive rate, we obtained a sensitivity of 0.63, specificity of 0.80, and precision of 0.64. Conclusions: This study suggests that our DL model can predict for the presence of LGE from a surface ECG. A future direction of this work would be to evaluate the use of DL for the prediction of LGE location, quantity and specific phenotypes of myocardial disease.

MRI in rectal cancer: An institutional real life analysis of technical parameters

MRI plays a key role in the preoperative staging of rectal cancers and choice of neoadjuvant radiochemotherapy. Yet, the acquisition and interpretation of rectum magnetic resonance imaging (MRI) turn out to be unequal, impacting patients'care. The present study aims at evaluating the quality of the acquisition of technical parameters of the rectal MRI performed by comparing them according to the various guidelines. The medical MRI reports of all consecutive patients with locally advanced rectal cancer treated in a curative intent, by preoperative RCT and completion surgery were retrospectively reviewed over two periods (January 2010-December 2014 and January 2018 and December 2020) according to international 2012 and 2016 ESGAR and 2017 SAR MRI recommendation reports. During the first period (69 MRI performed), 58% of these MRI abided by the recommendations and 75% of essential criteria could be found in 25.5% of MRI reportings. During the second period (73 MRI performed), the protocol used by 6.8% of MR images abided by the 2016 Society of Gastrointestinal and Abdominal Radiology(ESGAR) recommendations and 39.7% abided by the Society of Abdominal Radiology (SAR) recommendations. 75% of essential criteria could be found in 52.3% of MRI reportings and 90% of essential criteria could be found in 6.2% of MRI reportings. In an era of increasing individualized patient care and conservative treatment focused on tumour response and prognostic factors, the present study showed that compliance to MRI protocols and reporting guidelines needs improving to upgrade patient care.

Expression of clMagR/clCry4 protein in mBMSCs provides T2-contrast enhancement of MRI

Here, we propose for the first time the evaluation of magnetosensitive clMagR/clCry4 as a magnetic resonance imaging (MRI) reporter gene that imparts sensitivity to endogenous contrast in eukaryotic organisms. Using a lentiviral vector, we introduced clMagR/clCry4 into C57BL/6 mice-derived bone marrow mesenchymal stem cells (mBMSCs), which could specifically bind with iron, significantly affected MRI transverse relaxation, and generated readily detectable contrast without adverse effects in vivo. Specifically, clMagR/clCry4 makes mBMSCs beneficial for enhancing the sensitivity of MRI-R2 for iron-bearing granules, in which cells recruit exogenous iron and convert these stores into an MRI-detectable contrast; this is not achievable with control cells. Additionally, Prussian blue staining was performed together with ultrathin cell slices to provide direct evidence of natural iron-bearing granules being detectable on MRI. Hence, it was inferred that the sensitivity of MRI detection should be correlated with clMagR/clCry4 and exogenous iron. Taken together, the clMagR/clCry4 has great potential as an MRI reporter gene. Statement of significanceIn this study, we propose the evaluation of magnetosensitive clMagR/clCry4 as an MRI reporter gene, imparting detection sensitivity to eukaryotic mBMSCs for endogenous contrast. At this point, the clMagR and clCry4 were located within the cytoplasm and possibly influence each other. The clMagR/clCry4 makes mBMSCs beneficial for enhancing the sensitivity of MRI-R2 for iron-bearing granules, in which protein could specifically bind with iron and convert these stores into MRI-detectable contrast; this is not achieved by control cells. The viewpoint was speculated that the clMagR/clCry4 and exogenous iron were complementary to each other. Additionally, Prussian blue staining was performed together with TEM observations to provide direct evidence that the iron-bearing granules were sensitive to MRI.

Aktif Sakroiliitin Manyetik Rezonans Görüntülemesinde Gözlemciler Arasındaki Uyum

Introduction &#x0D; Axial spondyloarthritis has characteristic clinical features such as enthesitis, sacroiliitis and spondylitis, and extra-articular manifestations. Sacroiliitis (SI) occurs as a result of inflammation of the sacroiliac joint. Magnetic resonance imaging (MRI) of sacroiliac (SI) joints is used to detect early sacroiliitis. Sometimes, there can be variations in the interpretation of MRI findings of the SI joint among observers. Our aim was to investigate the inter-observer agreement among the observers.&#x0D; Materials and methods&#x0D; The study included the MRI results of 1150 patients who were diagnosed with active or chronic sacroiliitis based on the findings from sacroiliac MRIs, or whose MRI was deemed indicative of sacroiliitis by the rheumatologist. 1150 MRIs were re-evaluated by a different and expert radiologist.&#x0D; Results&#x0D; Out of the total 1150 patients investigated within the scope of this study. A statistically significant disparity emerged between the assessments provided by the expert radiologists and those obtained from outsourced radiologist evaluations.&#x0D; Conclusion&#x0D; The diagnosis of spondyloarthropathy may be delayed for some reasons. If the patient's clinic and MRI report are not consistent, the patient should not be removed from follow-up.

In Vivo Effect of Partial Ulnar Collateral Ligament Tear Location on Ulnohumeral Joint Gapping With Stress Ultrasound in Baseball Pitchers

Background: Stress ultrasonography (SUS) has provided clinicians with an alternative means of evaluating ulnar collateral ligament (UCL) injury by dynamically assessing ulnohumeral joint space gapping under applied valgus stress. However, 3 previous cadaveric studies have evaluated the biomechanical effect of partial UCL tears at different anatomic locations with conflicting results. Purpose/Hypothesis: The purpose of this study was to use in vivo data from patients with magnetic resonance (MR)–confirmed partial UCL tears to determine if anatomic partial tear location influenced the resultant stability of the elbow in terms of ulnohumeral joint gapping on SUS. It was hypothesized that no difference would be seen in the amount of ulnohumeral joint gapping based on MR-identified anatomic partial UCL tear location. Study Design: Cohort study; Level of evidence, 3. Methods: All patients diagnosed with elbow injury from 2015 to 2020 were screened to identify competitive baseball pitchers with a partial UCL tear who received an SUS. Partial UCL tear was diagnosed via surgeon and radiologist interpretation of elbow MR imaging (MRI) as well as confirmation at the time of surgery as documented in the operative report. Demographic, injury, and treatment data were collected from the clinic notes, MRI reports, and stress ultrasound reports. Ulnohumeral joint gapping on SUS was calculated as the difference between ulnohumeral joint gapping without valgus stress versus ulnohumeral joint gapping with valgus stress (delta) as performed by a musculoskeletal radiologist. Results: Overall, 60 male baseball pitchers (age, 19.2 ± 2.1 years), including 22 (37%) high school, 26 (43%) collegiate, and 12 (20%) professional pitchers, were evaluated. Regarding the location of partial UCL tears, 32 (53%) pitchers had proximal tears and 28 (47%) had distal tears. The mean delta value cohort was 2.1 ± 1.1 mm, and 34 pitchers (57%) had delta values >2.0 mm. Ulnohumeral joint gapping (high school [1.9 mm] vs collegiate [2.6 mm] vs professional [1.6 mm]; P = .004) and the percentage of pitchers with delta values >2.0 mm (45% vs 77% vs 33%; P = .017) differed based on level of competition. After controlling for age, body mass index, and level of competition in a multivariate linear regression, tear location was not related to ulnohumeral joint gapping (P = .499). Conclusion: No difference was observed in the amount of ulnohumeral joint gapping on SUS in symptomatic baseball pitchers with MR-identified partial UCL tears at differing anatomic locations. The findings of this study are contrary to previous cadaveric studies that have evaluated the effects of UCL tear characteristics on delta laxity.