Articles tagged with Radiology
Researchers found that AI significantly improved the detection of lung nodules on chest X-rays, detecting actionable nodules at a rate of 0.59% compared to 0.25% without AI assistance. The study suggests that AI may work consistently across different populations.
A systematic review and meta-analysis found that MRI-based surveillance after surgical treatment can detect clinically occult local recurrences, potentially improving patient outcomes. The study included 19 studies and showed a significant association between high-intensity surveillance and the detection of local recurrences.
A deep learning model distinguishes colon carcinoma from acute diverticulitis in CT images, improving diagnostic performance and reducing false-negative and false-positive findings. The model's sensitivity and specificity were significantly improved, supporting radiologists in patient care.
A new study published in Radiology: Cardiothoracic Imaging shows that photon-counting detector CT can acquire high-quality images at lower contrast media volume than conventional CT scanners. The technology reduces the amount of contrast needed for CT angiography, improving image quality and reducing environmental impact.
Researchers created monthly infant brain atlases to track normative trends in brain development. The atlases reveal changes in brain structure, cortical geometry, and tissue contrast, shedding light on conditions like ADHD and dyslexia. The study aims to facilitate discovery of new insights into child cognition and social behavior.
Scientists at Göttingen University have created a novel approach to produce X-ray images in color from a single exposure, eliminating the need for focusing and scanning. This breakthrough method uses an X-ray color camera and a specially structured plate to capture the intensity pattern of fluorescing atoms in a sample.
Researchers used deep learning AI to analyze chest X-rays of patients with acute chest pain syndrome, identifying those at high risk for life-threatening conditions. The model significantly improved prediction accuracy beyond conventional clinical markers, allowing for more accurate triage and potential deferral of additional testing.
Ringlike peripheral high iodine concentration maps from dual-energy CT can help guide management in patients with known lung cancer and an indeterminate solitary nodule. This finding showed excellent interobserver agreement, high specificity, and independently predicted pulmonary metastasis.
A recent study found that AI failed to pass a radiology qualifying examination, with an average accuracy of 79.5%. However, the researchers suggest that further training and revision could lead to improved results.
A new study published in the American Journal of Roentgenology found significant cost savings by eliminating sedation during pediatric brain MRI examinations. The analysis calculated costs for sedated, nonsedated, and limited MRI exams at three US hospitals, revealing that labor costs were the largest expense category.
Jon A. Jacobson, MD, FACR, has been recognized as the 2023 ARRS Distinguished Educator for his exceptional contributions to radiological education and innovative educational activities. He is a renowned expert in musculoskeletal ultrasound and MRI, with over 250 peer-reviewed publications.
The American Roentgen Ray Society has awarded the 2023 ARRS Gold Medal to Dr. Bernard F. King, Jr., MD, FACR, FSAR, in recognition of his distinguished service to radiology. Dr. King has made significant contributions to the field, including the development of contrast media recognition and treatment protocols.
A team led by Abhinav Jha is developing a novel low-count quantitative single photon emission computed tomography (LC-QSPECT) method to measure the concentration of radiopharmaceutical material in tumors and vital organs. This technique aims to provide accurate measurements of absorbed dose from radiation therapy, helping plan treatmen...
Veterinary experts warn that AI algorithms used in radiology and imaging can provide faulty or incomplete diagnoses, posing risks to patient care. The absence of regulatory oversight for veterinary AI products increases concerns over accuracy, transparency, and potential harm.
A study published in the American Journal of Roentgenology found that reconstruction using BI64 kernel and 0.4-mm slice thickness yielded improved bronchial division identification and pulmonary fissure sharpness without loss in pulmonary vessel sharpness or pathology conspicuity.
Researchers have developed a new X-ray technology that visualizes lung tissue microstructure, providing additional information for accurate diagnosis. Dark-field X-ray images can differentiate between diseased and healthy lung tissue, potentially replacing computed tomography (CT) for repeated examinations.
For adrenal lesions evaluated by single-phase dual-energy CT, fat fraction had significantly higher sensitivity than virtual noncontrast attenuation at both clinically optimal threshold and traditional ≤10 HU threshold. Fat fraction-derived metrics can help definitively diagnose incidental adrenal lesions as adenomas.
Researchers found a strong association between hotter weather, more sunshine, and higher volumes of polytrauma CT scans. The study used machine learning algorithms to forecast daily polytrauma CT occurrence, predicting 73% of high-demand days and 83% of low-demand days.
A new machine learning fusion model has been developed to diagnose ovarian cancer more accurately by combining ultrasound and photoacoustic tomography imaging. The model achieved an accuracy of 90% in detecting ovarian lesions, outperforming previous methods.
Researchers at USC's Keck School of Medicine are collaborating with Polaris Dawn to develop a novel method for collecting X-rays in outer space. They aim to harness ambient radiation to create images using analog X-ray film, which could improve medical care on long-duration spaceflights.
A large-scale international study found that patients diagnosed with lung cancer at an early stage via CT screening have a 20-year survival rate of 80 percent. This is compared to the average five-year survival rate for all lung cancer patients, which is 18.6 percent.
Researchers evaluated modified CEUS using perfluorobutane for diagnosing hepatocellular carcinoma in high-risk patients. The study found no significant difference in sensitivity, specificity, or accuracy between modified CEUS and CT/MRI LI-RADS v2018.
Researchers from the University of Ottawa found an association between marijuana smoking and chronic damage to airways, including higher rates of paraseptal emphysema and airway inflammatory changes. The study suggests that marijuana smoking may lead to additional lung damage beyond what is seen in tobacco smokers.
The IOF and ESCEO position paper recommends retaining the current WHO diagnostic criteria for osteoporosis, which has served the field well. The paper suggests that distinguishing between diagnostic and intervention thresholds can help improve treatment of patients at high risk of fractures.
A multicenter study found that left ventricular global longitudinal strain is a significant independent predictor of all-cause mortality and/or heart-failure hospitalization in patients with ischemic or nonischemic dilated cardiomyopathy. The study used cardiac MRI feature tracking to calculate six myocardial strain parameters.
The American Roentgen Ray Society (ARRS) named Dr. Sarah Kamel of Thomas Jefferson University Hospital and Dr. Ankur Goyal of the All India Institute of Medical Sciences as Melvin Figley and Lee Rogers Fellows in Radiology Journalism, respectively. They will receive expert instruction in scientific writing and communication through the...
Dual-energy CT (DECT) metrics provide insight into physiologic changes after lung cancer surgery, beyond PFTs and conventional CT. Lung perfusion ratio increases were greater after lobectomy than limited resection.
An AI-based model has been developed to assist radiologists in detecting and identifying leadless implanted electronic devices (LLIEDs) on chest X-ray images. The model achieved high detection and classification accuracy, even with suboptimal image quality, and showed promise for real-world deployment.
A study published in American Journal of Roentgenology found that a PACS tool improved completion rates of clinically necessary follow-up imaging. The tool identified socioeconomically disadvantaged patients at increased risk of failure, and referrer agreements played a crucial role in improving recommendations.
A deep learning algorithm trained on adult populations shows improved sensitivity for diagnosing thyroid nodules in children, with higher accuracy than traditional ACR TI-RADS. The study suggests the potential benefits of this new approach for evaluating thyroid nodules in younger patients.
A study by the Radiological Society of North America found that health insurance companies often pay more than necessary for radiology services, leading to higher premiums and out-of-pocket costs. The study also suggests that radiologists can play a key role in delivering high-quality, low-cost care to patients.
A new report from a joint task force recommends improved access to interventional radiologists in small and rural areas. The task force focused on strategies for recruitment and retention, including training opportunities, financial models, and community development.
A study of over 4 million claims found that Black women had less access to new mammography technology, even when receiving mammograms at the same institution. The disparities in digital breast tomosynthesis use among Black and white women began to subside after 2016.
Researchers have developed a novel, non-invasive way to measure blood flow to the brains of newborn children at the bedside. This method uses real-time ultrasound color flow technique and 3D sampling to capture flow in a completely painless and safe manner.
A study using gadoxetate disodium-enhanced MRI found associations between imaging characteristics and hepatocellular adenoma (HCA) subtypes. The algorithm identified common HCA subtypes with high accuracy, including β-catenin exon 3 mutations.
A new study found that fewer Asian, Black, and Hispanic patients qualify for anti-amyloid monoclonal antibody treatments, which may slow Alzheimer's progression. The treatment's effectiveness depends on the presence of amyloid plaques, with lower incidence in minorities.
A study published in the American Journal of Roentgenology found that large specialized referral centers for complex IVC filter retrievals achieved a 100% success rate and low adverse event rates. Prolonged filter dwell time was identified as an independent predictor of adverse events.
A prospective multicenter case-control study found that third-trimester fetuses exposed to opioids in utero exhibited smaller brain biometric measurements and altered fetal physiology. The study suggests a possible link between prenatal opioid exposure and postnatal clinical outcomes.
A prospective study found that patient decision aids improve patients' self-perceived understanding of image-guided procedures and satisfaction with the consent conversation. Patients who received a PDA before their visit reported significantly greater understanding and felt more listened to by their clinician.
The contrast-enhanced in-phase Dixon sequence demonstrated higher sensitivity and reader confidence for detecting breast biopsy clips on MRI. The sequence also showed a higher contrast-to-noise ratio without affecting positive predictive value.
A new AI tool has been developed to detect pancreatic cancer on CT scans with high accuracy, achieving 90% sensitivity and 96% specificity. The tool's performance is comparable to that of radiologists, offering a promising supplement for early detection and improved treatment outcomes.
A study found that trained radiographers perform as well as radiologists in key areas of double reading mammograms, improving cancer detection rates. The results suggest that focused training and experience are key factors in reader performance, not just medical degree or broad radiology education.
Researchers at Mayo Clinic Comprehensive Cancer Center found that metabolic imaging, combined with traditional treatment response assessment methods, can provide critical information to guide therapy for pancreatic cancer patients. The use of positron emission tomography (PET) with 18-fluorodeoxyglucose (FDG) tracer adds significant pr...
Researchers evaluated 35 commercial breast biopsy markers for ultrasound color Doppler twinkling, finding that three markers exhibited actionable twinkling for ≥2 transducers. Higher surface roughness was associated with actionable twinkling in C1-6 and 9L transducers.
A large retrospective study found that visceral fat area from fully automated and normalized abdominal CT analysis predicts subsequent myocardial infarction or stroke in Black and White patients. The study suggests that body composition analysis using machine learning could be widely adopted to add prognostic utility to clinical practice.
The study found a significant increase in the use of single-encounter thoracoabdominopelvic CT examinations and minor injuries between 2011-18. In commercially insured patients, rates rose from 3.4 to 9.8 for single-encounter thoracoabdominopelvic CT per 1,000 trauma-related ED encounters.
A proof-of-concept study developed three machine learning models to predict posttreatment recurrence in early-stage hepatocellular carcinoma patients. The models achieved high accuracy using imaging data alone, while combining clinical data did not significantly improve performance.
Researchers developed a blood test tool that predicts patient outcomes after traumatic brain injury (TBI), with 87% accuracy for predicting death and 86% for predicting severe disability. The tests are diagnostic and prognostic, easy to administer, swift, and inexpensive.
A study published in the American Journal of Roentgenology found that combining deep-learning reconstruction with a subtraction technique yielded optimal diagnostic performance for detecting in-stent restenosis by coronary CT angiography. This approach reduced radiation exposure while maintaining high image quality.
A study found that electronic health record (EHR) order entry-based interventions significantly reduced contrast-enhanced CT utilization within a large health system during the global iodinated contrast media shortage. The number of patients undergoing contrast-enhanced CT examinations per day decreased by 12.0%, and the number of orde...
A retrospective study found that patients with multiple architectural distortions on digital breast tomosynthesis were more likely to have high-risk pathology, but not significantly different frequencies of malignancy. Biopsy of all areas may be warranted for these patients.
A retrospective study of molecular breast imaging (MBI) findings suggests that the MBI lexicon's lesion descriptors can inform interpretation and guide the incorporation into the BI-RADS Atlas. The study found high positivity rates for malignancy among mass lesions, those with marked uptake intensity, and certain distribution patterns.
A retrospective study found an AI algorithm for detecting incidental pulmonary embolus (iPE) on contrast-enhanced chest CT examinations had high NPV and moderate PPV, even identifying iPEs missed by radiologists. The tool's diagnostic performance did not show significant variation across study subgroups.
Researchers have created a noninvasive method to identify tumors causing hypertension by reducing radiation exposure. The new fluorine-based agent replaces an iodine-based agent and may allow for screening of patients with hypertension-linked aldosterone adenomas.
The American Journal of Roentgenology has dramatically increased its impact factor from 3.959 to 6.582 in 2021, according to Journal Citation Reports. The journal's new ranking places it in the 83rd percentile in the Radiology category, reflecting the high quality of work by authors and reviewers.
Researchers found brain ripples in all areas of the human cortex, synchronizing distant elements of memory upon recollection. The ripples involve brief, high-frequency oscillations that promote neural interactions and potentially support interaction between distant locations.
A multicenter study found that radiologists can accurately diagnose rickets and classic metaphyseal lesions (CMLs) with high interobserver agreement and diagnostic performance. The study highlighted the importance of considering age factors in diagnosis, as CMLs mostly occur in children younger than 6 months.
The global contrast media shortage has affected millions of diagnostic imaging examinations and requires immediate attention. Researchers propose several methods to conserve contrast media, including weight-based dosing and adjusting CT settings.
A 5-tiered CT scoring algorithm, including mass-to-cortex corticomedullary attenuation ratio and heterogeneity score, showed substantial inter-observer agreement and moderate diagnostic accuracy for clear-cell RCC. The algorithm may represent a clinically useful tool for diagnosis in small solid renal masses.
A study published in JAMA Oncology found significant disparities in breast biopsy delays among non-white women, with Black women experiencing the longest delays. Structural racism within healthcare facilities and screening site-specific factors contribute to these disparities.