A new special report warns of LLM cybersecurity threats in radiology, highlighting the importance of implementing security measures to prevent malicious use. The report recommends robust security measures and best practices for safe integration of LLMs into healthcare.
A new study identified six distinct breast tissue patterns associated with an increased risk of invasive breast cancer, particularly among Black women. The radiomic phenotypes may help predict the likelihood of false-negative mammograms or interval cancer diagnoses.
A survey of patients who underwent breast cancer screening mammograms found most support AI's role as a second reader, but concerns about data privacy and bias persisted. Demographic factors such as education and racial background played a significant role in shaping patient perceptions of AI.
A new study published in Radiology found that higher blast exposure correlated with changes in brain function and structure, even when standard MRI exams appeared normal. The researchers discovered weaker connectivity in key areas and more severe symptoms such as memory problems and post-traumatic stress disorder.
Researchers developed a deep learning algorithm to denoise ultra-low dose CT scans, improving image quality and accuracy. The study found that this approach can diagnose pneumonia in immunocompromised patients using only 2% of the radiation dose of standard CT scans.
A new study uses advanced lung MRI to detect regional blood flow abnormalities in children and adolescents with long COVID, which may explain persistent symptoms. The researchers found reduced blood flow patterns that correlated with fatigue symptom severity, suggesting a possible biological basis for ongoing symptoms.
Researchers developed a robust AI model that automates segmenting of MRI images, reducing radiologist workload and improving consistency. The TotalSegmentator MRI model achieved high performance on various anatomical structures, with a Dice score of 0.839.
A study found racial and ethnic minorities are less likely to receive standard-of-care diagnostic imaging after abnormal screening mammograms compared to white patients. Minority groups were also less likely to have access to same-day biopsy services, despite similar availability of most diagnostic services.
A new study found that COVID-19 infection is associated with faster growth of plaque in the coronary arteries, leading to an increased risk of cardiovascular events. The study also revealed a higher incidence of high-risk plaques and coronary inflammation among infected patients.
A new deep learning model has been developed to detect and segment lung tumors on CT scans with high accuracy. The model achieved 92% sensitivity and 82% specificity in detecting lung tumors, outperforming physician-delineated volumes in some cases. However, further research is needed to improve the model's performance for larger tumors.
A deep learning model developed by researchers at San Diego State University can accurately diagnose chronic obstructive pulmonary disease (COPD) using a single inhalation lung CT scan. The study found that the model performed similarly to traditional two-phase CT measurements, with added clinical data improving accuracy.
The study found a significant increase in late-stage breast cancer diagnoses among U.S. women, with annual percentage increases of up to 2.9% between 2004 and 2021. The researchers attribute the trend to factors such as lack of national screening programs, inconsistent guidelines, and increasing obesity rates.
Researchers found structural and functional alterations in specific brain regions of individuals with opioid use disorder, including increased global connectivity. Women in the group had smaller medial prefrontal cortex volume compared to men.
Researchers have developed a new coronary risk score specifically for women, accurately predicting and categorizing the risk of major adverse cardiovascular events. The findings suggest that this novel approach can help identify high-risk women earlier, reducing the risk of heart attacks and sudden cardiac death.
A self-pay, AI-enhanced breast cancer screening program was offered to patients across 10 clinical practices, resulting in a 43% higher cancer detection rate among enrolled women. The AI-powered program attributed 21% of the increase in cancer detection, with higher-risk patients choosing to enroll more frequently.
A study published at the Radiological Society of North America annual meeting found that ultra-processed food intake is associated with higher amounts of intramuscular fat in thigh muscles. This increase in fat can lead to knee osteoarthritis, a condition affecting millions worldwide.
A new study found significantly higher injury rates among transgender women compared to cisgender women, with increased risk of head, facial, and chest injuries. The findings emphasize the need for timely IPV screening and support for vulnerable populations.
A study found that skeletal muscle loss, particularly in the temporalis muscle, is associated with an increased risk of developing Alzheimer's disease dementia. Participants with smaller temporalis muscle size were more than twice as likely to develop dementia over a median follow-up period of 5.8 years.
A minimally invasive procedure called Genicular artery embolization (GAE) has been shown to effectively relieve knee pain and improve quality of life in people with osteoarthritis. The study found that GAE can significantly reduce knee pain and improve quality of life, even in advanced cases.
A growing number of engineered stone countertop workers are developing silicosis, a long-term lung disease caused by inhaling crystalline silica dust. The disease can lead to respiratory failure, increased risk of infection, and chronic obstructive pulmonary disease.
Researchers have discovered a specific type of body fat linked to abnormal proteins in the brain that are hallmarks of Alzheimer's disease. Higher levels of visceral fat were found to be associated with increased amyloid accumulation and higher PET levels of hallmark pathologic proteins.
A study found that soccer heading causes structural brain abnormalities similar to those seen in CTE, affecting the frontal lobe and verbal learning. The research suggests repeated head impacts may harm brain health even without serious injury.
Researchers found lower pulmonary gas exchange may be associated with impaired cognitive function in patients with long COVID. The study also showed a potential treatment strategy using methods that target improved gas exchange, suggesting a causative relationship between cognitive dysfunction and lung dysfunction.
A new study reveals that concussions impact aperiodic brain activity in high school football players, leading to slowed cognitive function and worse post-concussion symptoms. The research highlights the importance of protective measures in contact sports and may lead to new treatments for concussion recovery.
Researchers identified acute effects of cigarette and e-cigarette smoking on vascular function, including decreased blood flow velocity and venous oxygen saturation. The study found that vaping with and without nicotine led to significant changes in vascular function, highlighting the potential harm of vaping even without nicotine.
A recent study found that levothyroxine use is associated with greater loss of total body bone mass and density, even in participants with normal TSH levels. The research suggests that adults taking levothyroxine should discuss their treatment with their healthcare provider and regularly monitor their thyroid function tests.
A study published in Radiology found that radiologists and physicians who received local explanations from artificial intelligence (AI) systems performed better and made faster diagnoses than those who received global explanations. The researchers also found that AI advice was trusted more quickly when it provided local explanations.
A 10-year study found that short-term exposure to heat and air pollution was associated with an increase in medical imaging exams, particularly for chest, neuro, and musculoskeletal imaging. High temperatures had a greater impact on X-ray utilization, while high particulate matter levels were linked to increased CT scans.
Radiologists who had access to questionnaire information during MRI interpretation achieved a near statistically perfect agreement with spine specialists. This improved diagnostic accuracy and confidence, avoiding diagnostic discrepancies between radiologists and spine specialists.
A 10-year study found that digital breast tomosynthesis increases cancer detection rates and reduces advanced cancers compared to conventional 2D digital mammography. DBT detected more aggressive cancers at an earlier stage than digital mammography, leading to improved cancer detection and lower recall rates.
The study found that ChatGPT-4 Vision performed well on text-based radiology exam questions but struggled with image-related questions, achieving an accuracy of 65.3%. The model's performance varied across subspecialties and images, with limitations in interpreting certain types of radiologic images.
A uniform lexicon has been developed and endorsed by a multi-medical society panel to describe observations on ultrasound during the first trimester of pregnancy. The new terms aim to minimize bias, clarify findings, and respect patient preferences.
A commercial AI tool was effective in excluding pathology and had lower rates of critical misses on chest X-rays compared to radiologists. The study found that the AI tool could autonomously report more than half of all normal chest X-rays without decreasing standard of care.
A recent study published in Radiology found that solid lung nodules were present in approximately 42% of non-smoking adults, with prevalence increasing with age. The study suggests that current nodule follow-up guidelines may lead to unnecessary follow-up examinations for low-risk individuals.
A study found that CT scans can identify individuals at high risk of type 2 diabetes through automated analysis of various body components. The index of visceral fat showed the highest predictive performance for diabetes.
A new deep learning model detects clinically significant prostate cancer on MRI with performance comparable to experienced abdominal radiologists. The model's prediction can be used as an adjunct to improve diagnostic performance and reduce false positives.
Despite progress, radiology's gender gap persists due to structural biases. RSNA's efforts have increased female representation in faculty and leadership positions.
The Special Report explores expert perspectives on deploying AI in radiology, emphasizing the need for trust, reproducibility, explainability, and accountability. Radiologists must be able to trust in AI systems' design and receive adequate training, while establishing clear guidelines regarding clinical accountability.
A study published in Radiology found that brain connectivity on MRI scans can predict the progression of brain atrophy in patients with mild Parkinson's disease. The researchers used MRI data to generate a map of the brain's neural connections, which showed potential in predicting gray matter atrophy accumulation over time.
A study published in Radiology found that AI significantly improved breast cancer detection and reduced false-positive findings. The AI system detected more cancers while lowering the rate of unnecessary recalls, reducing radiologist workload by 33.4%.
A study published in Radiology found that low-level light therapy increased resting-state functional connectivity in patients with moderate traumatic brain injuries. The treatment showed improved brain connectivity within the first two weeks, but its long-term effects are still unclear.
A study of nearly 5,000 mammograms found that AI algorithms produced false positive results more frequently in Black and older patients, while Asian patients and younger patients were less likely to be flagged as suspicious. Healthcare institutions should consider the patient population they serve before purchasing an AI algorithm.
A new study found that progressive lung abnormalities in smokers are associated with increased frequency and severity of acute respiratory diseases, highlighting the need for closer monitoring and earlier intervention.
A new study found that GPT-4 and Google Gemini performed poorly in breast imaging classification, with a high percentage of discordant assignments that could impact patient management. The researchers emphasize the need for regulation of these large language models in high-level medical scenarios.
Radiologists are proposing actions to reduce their department's greenhouse gas emissions and become more resilient to the effects of climate change. They suggest implementing a coordinated approach, establishing sustainability teams, and optimizing scanner usage to reduce emissions by up to 33% for MRI and CT scans.
A study found that GPT-4 matched the performance of radiologists in detecting errors in radiology reports, with an accuracy rate of 82.7%. The use of GPT-4 resulted in lower mean correction cost per report than the most cost-efficient radiologist.
A new study found that a non-invasive imaging test can help identify patients with coronary artery blockage or narrowing who need a revascularization procedure. The test, called CT-FFR, uses CTA images and AI algorithms to model coronary blood flow, reducing the need for invasive procedures.
A new consensus statement provides recommendations for improving diagnosis of deep endometriosis through additional pelvic ultrasounds and imaging. The guidelines aim to reduce diagnostic delay and enhance detection of the condition, which affects 10% of women of reproductive age.
A recent study has found increased coronary vessel wall thickness in asymptomatic individuals living with HIV, which is significantly associated with impaired diastolic function. The study also highlights the impact of HIV on developing subclinical coronary artery disease and its effects on heart function.
A 10-center study found that microwave ablation is associated with similar progression-free survival rates to surgery but with fewer complications and a greater potential for preserving thyroid function. The procedure resulted in less blood loss, shorter incision length, and shorter procedure and hospitalization durations.