Articles tagged with Radiology
Researchers have developed an intelligent metamaterial that boosts the energy emitted by a patient's body, increasing signal-to-noise ratio and improving MRI imaging. The technology reduces scan time and cost, making high-quality imaging more accessible to patients worldwide.
Researchers observed reversed halo signs in most cases of CT-based septic pulmonary embolism diagnosis related to IV substance use disorder. The sign is an early and reliable imaging finding.
Researchers developed a neural network, PatchFCN, trained on 4,396 CT scans to detect brain hemorrhage abnormalities with accuracy similar to human experts. The algorithm achieved high accuracy and pixel-level delineation, classifying abnormalities into different pathological subtypes.
A new AI algorithm developed by UCSF and UC Berkeley outperformed two out of four expert radiologists in detecting tiny brain hemorrhages on head scans. The algorithm achieved exam-level accuracy, tracing detailed outlines of abnormalities within the brain's three-dimensional structure.
A new AI tool, trained on a large dataset of mammography images, accurately identified breast cancer with 90% accuracy when combined with radiologist analysis. The study suggests that AI can augment human radiologists' diagnoses, reducing false-positive and false-negative results.
The AI system uses a huge database of x-ray images to identify collapsed lungs with 75% accuracy, outperforming medical specialists who diagnose fewer than 50%. Researchers plan to integrate the technology into a software system and apply it to other conditions, reducing treatment delays and improving patient outcomes.
The American Journal of Roentgenology review article details common imaging manifestations of vaping-associated lung injury, including hypersensitivity pneumonitis, diffuse alveolar hemorrhage, and organizing pneumonia. The article emphasizes the importance of recognizing these patterns in radiologists to prompt clinical teams to ask a...
A study of academic radiology departments found that there is no uniform approach to handling outside imaging studies, with some requiring formal reports and others allowing them in the institutional PACS. This lack of consensus may impact patient care and radiation safety.
The multi-society statement focuses on three areas: data, algorithms, and practice. It emphasizes the importance of ethical use of AI in radiology, ensuring benefits and harms are distributed fairly among stakeholders. Radiologists will need to acquire new skills to work effectively with AI tools.
A new AI system can accurately identify key findings in chest X-rays of pneumonia patients in just 10 seconds, significantly outperforming current clinical practice. This ultra-quick detection enables physicians to confirm a pneumonia diagnosis faster, allowing for timely treatment and reducing delays for severely ill patients.
A study published in the American Journal of Roentgenology found that diagnostic radiologists with lifetime ABR certificates were significantly less likely to participate in Maintenance of Certification (MOC) programs. Participation rates were only 13.9% among those with lifetime certificates, compared to nearly universal participation...
Nonphysician providers, including nurse practitioners and physician assistants, rarely render diagnostic imaging services, with most being radiography and fluoroscopy. Despite growing involvement in imaging-guided procedures, NPPs still represent a small fraction of all diagnostic imaging interpretations.
A large study of almost 200,000 patients found that premedicating with antihistamines and switching contrast media can reduce recurrent allergic reactions. Genetic predisposition may also play a role in reaction to CT contrast agents, according to the researchers.
A lack of government action on NHS staffing undermines efforts to diagnose cancer early, with nearly half of all cancers diagnosed in England at stage 3 or 4. The health service needs an extra 1,700 radiologists and nearly 2,000 therapeutic radiographers to improve staff efficiency.
A new machine learning-based model evaluates immunohistochemical characteristics in patients with suspected thyroid nodules, achieving excellent performance for individualized noninvasive prediction. The model improves thyroid nodule diagnoses and helps identify papillary thyroid cancers earlier.
A recent study found that vaping impairs vascular function, reducing blood flow and increasing arterial stiffness in healthy adults. The study used MRI scans to measure the effects of e-cigarette inhalation on the body's vascular system.
The article provides a comprehensive overview of gender affirmation surgical therapies encountered in diagnostic imaging. It defines normal postsurgical anatomy and describes select complications using a multidisciplinary, multimodality approach. Key findings include the importance of regular prostate cancer screening for trans-female ...
A multidisciplinary team at Kaiser Permanente developed a complete imaging history definition, including four prompts for date, location, pain, and concern. This led to a significant increase in orders containing all four components, from 16% to 52%, and improved mean character count of entered histories.
A study found that smoking impairs the effectiveness of embolization treatment for pulmonary arteriovenous malformations (PAVMs) in patients with HHT. Smokers experienced a higher rate of PAVM persistence after treatment compared to non-smokers, with those who smoked more than 20 pack-years facing a fivefold increased risk.
A new AI tool, CXR-risk, analyzes chest X-ray images to predict health and mortality, providing independent information that surpasses radiologists' readings. The study found that CXR-risk identified people at high risk for future heart attack, lung cancer, or death with accuracy.
A new study found increased CT use for suspected urolithiasis patients in emergency departments, with utilization rates increasing by 100.8% between 2006 and 2014. Geographic variation was also noted, with CT scans being more frequent in higher-income ZIP codes and urban hospitals.
A new machine learning approach for low-dose CT imaging has been shown to perform as well as, or better than, traditional iterative techniques in an overwhelming majority of cases. The method allows radiologists to fine-tune images according to clinical requirements, enabling faster and more accurate scans.
A new AI tool developed by researchers at Stanford University improves clinicians' ability to correctly identify brain aneurysms by highlighting areas of interest on scans. The HeadXNet algorithm reduces the 'miss' rate and increases consensus among clinicians, with promising results but further investigation needed.
A new study published in the Journal of Vascular and Interventional Radiology found significant growth in hemodialysis conduit angiography utilization, increasing by 1700% nationally from 2001 to 2015. This trend was particularly pronounced among nephrologists, with a 24.0% increase in procedures performed annually in office settings.
A new AI system has been developed that can detect lung cancer with high accuracy, potentially leading to earlier treatment. The system uses deep learning to analyze low-dose chest computed tomography (LDCT) scans and performs better than human radiologists in detecting malignant nodules.
Researchers evaluated clinical and radiological features of fibro-adipose vascular anomaly (FAVA), a soft-tissue vascular lesion. FAVA is characterized by unique clinico-radiological features, including a painful intramuscular lesion in extremities, progressive heterogeneous enhancement on MRI, and associated phlebectasia.
A recent study reveals that radiology residents in the US are not adequately trained to identify and report child abuse. The study, presented at the ARRS 2019 Annual Meeting, found that residents correctly identified only 37.6% of non-accidental trauma cases over a five-year period.
Jonathan S. Lewin, a renowned expert in interventional and intraoperative MR imaging, has received the 2019 ARRS Gold Medal for his distinguished contributions to radiology. With over 200 peer-reviewed scientific manuscripts published, he is a pioneer in advancing knowledge and skills in radiology.
The upgraded picture archiving and communication system (PACS) significantly reduces the time radiologists spend searching for prior studies, with average search times decreasing from 6.1 to 4.6 seconds. The new system also reduces total study reading time, with a decrease of 137 to 113 seconds per study.
Christine Menias, MD, is recognized for her dedication to radiology education, with a notable track record of instructional courses and awards. She has been named the 2019 ARRS Distinguished Educator for her inspirational teaching and commitment to improving competence and patient outcomes.
A survey of fourth-year radiology residents who took the 2018 ABR Core Exam found that those who passed perceived higher value in preparation resources. Residents who passed also had more study time off and higher USMLE step 1 scores compared to those who failed.
A new AI system called FocalNet has been developed to aid radiologists in detecting prostate cancer, achieving 80.5% accuracy in reading MRI scans. The system uses machine learning to evaluate tumors and predict their aggressiveness, providing diagnostic guidance to less-experienced radiologists.
A workshop published a roadmap for AI in medical imaging, highlighting key research themes and prioritizing foundational machine learning research. The report emphasizes the need for collaboration among professionals, funding agencies, and institutions to develop innovative imaging technologies.
A new study published in the American Journal of Roentgenology reviews historical rationale for patient shielding and concludes that its benefits outweigh its risks. The practice, which has been in place since 1976, poses little benefit while increasing radiation exposure and decreasing image quality.
Dr. Harry K. Genant received the 2019 International Osteoporosis Foundation Medal of Achievement for his innovative work in non-invasive and quantitative imaging methods, advancing knowledge on bone fragility and fracture risk assessment. This award recognizes his significant contributions to osteoporosis research.
A study found that lumbar spine MRI reports were written at a readability level significantly above the average reading ability of US adults, with only one report meeting basic education standards. The high complexity of these reports may limit patient understanding and access to their health information.
A new study found that patient factors, including distance from hospital, age, and health insurance status, significantly impact follow-up imaging rates. The study also showed disparities in follow-up rates among different age groups, with younger and older patients having lower completion rates.
Researchers evaluated the safety and effectiveness of uterine artery embolization (UAE) in patients with various autoimmune diseases. UAE was technically and clinically successful in all patients, with no significant difference in volume reduction rates between groups.
Researchers use virtual reality to guide catheters through blood vessels with electromagnetic sensors, reducing radiation exposure and increasing precision. The technology has the potential to increase access to interventional radiology treatments in rural areas.
A study presented at the Society of Interventional Radiology Annual Scientific Meeting found that training interventional radiologists to perform endovascular thrombectomies improves access to life-saving stroke care. The model expands access to quality, evidence-based care and reduces lifelong disability associated with stroke.
A new study shows that uterine fibroid embolization (UFE) is as effective as surgical treatment in treating uterine fibroids, with fewer post-procedure complications. UFE preserves the uterus and allows women to return to their normal lives sooner.
Researchers developed a device-sizing application for Google Home smart speakers to provide accurate medical information and recommendations during procedures. The technology aims to improve patient care by allowing physicians to concentrate on patient care while reducing device technicalities.
Researchers found statistically significant reductions in pain-rating scores and improvements in tendinosis and tear scores after transcatheter arterial embolization (TAE) treatment. The non-surgical procedure decreases abnormal blood flow to the injured area, alleviating inflammation and pain without invasive surgery or physical therapy.
A study published in the Journal of Neurology, Neurosurgery and Psychiatry found increased connectivity between motor regions and the multimodal integration network in patients with FND. Symptom severity also correlated with increased coupling between the insula and the amygdala, which improved response to treatment.
A new study found that AI systems can perform as accurately as radiologists in detecting breast cancer in digital mammography screening. This breakthrough suggests that AI may be a viable alternative to current methods, which are labor-intensive and facing staff shortages.
A new study explores an episodic bundled payment model for breast cancer screening reflecting the widespread adoption of DBT. The analysis found that non-DBT approaches to bundled payment models remain viable, but higher DBT-inclusive prices highlight the need to explore societal costs more broadly.
A study published in Radiology found that radiologists can identify patterns of injury indicative of intimate partner violence. The research suggests that medical images can offer early clues of this type of violence, particularly among women and those with certain socioeconomic factors.
A new study finds that women's brains are metabolically three years younger than men's of the same age, which may contribute to their greater mental sharpness in later years. The researchers used PET scans and machine-learning algorithms to measure brain metabolism and calculate each person's brain age.
A novel AI system can reduce average delay in reporting abnormal chest X-rays from 11 days to less than 3 days. The system uses computer vision and NLP algorithms to quickly identify critical findings, prioritizing exams for expert radiologist opinion.
A new AI system can interpret and prioritize abnormal chest X-rays with critical findings, reducing the backlog of exams and bringing urgently needed care to patients more quickly. The system distinguishes abnormal from normal chest X-rays with high accuracy and can perform this task significantly faster than human radiologists.
Researchers at the University of Cincinnati are studying the use of neuroimaging to pinpoint the risk factors of stroke recurrence, which affects nearly 800,000 people annually in the US. The study aims to identify patients at higher risk for recurrent strokes by assessing imaging for signs of small vessel disease in the brain.
A Massachusetts General Hospital team developed an AI system to diagnose and classify brain hemorrhages using CT scan images. The system can provide the basis for its decisions by highlighting key features of each subtype.
Researchers from Ben-Gurion University of the Negev developed a system to monitor outgoing commands from CT equipment, detecting anomalies and alerting operators before execution. A second study looked at the potential to tamper with mammogram results using AI-powered image manipulation.
Researchers found a link between shoulder muscle brightness on ultrasound and type 2 diabetes, with low glycogen levels suspected as the cause. The study's findings suggest that radiologists can use ultrasound images to predict diabetic status in nearly nine out of ten patients.
A new study published in Radiology uses deep learning to predict Alzheimer's disease from brain imaging, achieving 100% sensitivity in detecting the disease an average of more than six years prior to final diagnosis. The algorithm has the potential to complement existing tests and provide opportunities for early intervention.
A deep-learning model has been successfully used on real patients to assess dense breast tissue in mammograms with high accuracy. The model outperforms traditional prediction models, achieving a kappa score of 0.85 in clinical applications, and could lead to more consistent screening procedures nationwide.
A new AI algorithm measures breast density at a high level of accuracy, comparable to experienced mammographers. The system has been successfully implemented in routine clinical practice and has the potential to standardize and automate breast density assessment.
A deep learning approach can identify nuanced mammographic imaging features specific to recalled but benign (false-positive) mammograms, distinguishing them from those identified as malignant or negative. The study achieved an area under the curve (AUC) of 0.76-0.91, indicating high performance in detecting false recalls.
A new study found that abdominal aortic calcification measured by CT scans is a strong predictor of future cardiovascular events, outperforming the widely used Framingham risk score. The research suggests that detecting calcification via routine CT scans could help identify patients at high risk and enable early interventions.
Researchers developed a new MRI technique that analyzes overlooked data to reveal brain cell density and connectivity. This may aid in diagnosing brain conditions such as Alzheimer's disease, multiple sclerosis, traumatic brain injury, and autism.