Selecting the appropriate magnetic resonance imaging (MRI) protocol to elucidate a patient’s clinical diagnosis poses a significant challenge for radiologists, particularly for less experienced radiologists [1,2]. It is essential to choose relevant sequences to aid in diagnosing and initiating the correct therapy. This process can be time-consuming and is also subject to inter-reader variability, as different radiologists may have varying preferences for specific sequences [3]. The Radiology Req…
We scan the top radiology sources so you don’t have to.
From AI breakthroughs to imaging trends, we serve up real-time radiology insights.
-
Large language models in radiology reporting: Bridging semantics, education, and safety
We read with great interest the article by Lindholz et al., “Comparing large language models and text embedding models for automated classification of textual, semantic, and critical changes in radiology reports” [1]. The authors should be commended for addressing an important and often overlooked aspect of radiological practice: the systematic evaluation of changes between preliminary reports authored by residents and finalized versions approved by attending radiologists. Their study is highly …
-
AI radiomics predicts spatial glioma recurrence on preoperative MRI: a systematic review
AI models have been shown capable of predicting local and distant tumor recurrence in glioma patients from baseline MRI data. While the high odds ratios reported from the multicenter study are encouraging, the evidence comes mainly from small, single-center, retrospective cohorts, so larger prospective multicenter studies are needed before clinical adoption.
-
Synthetic MRI as a T2WI alternative in bp-MRI: comparable image quality and improved PI-RADS performance
Prostate cancer (PCa) is the second most prevalent malignant neoplasm among men globally, with its incidence rising significantly in China and Eastern European countries [1]. In recent years, multiparametric magnetic resonance imaging (mp-MRI), which includes T2-weighted imaging (T2WI), diffusion-weighted imaging (DWI), and dynamic contrast-enhanced imaging (DCE), has increasingly played a significant role in the non-invasive diagnosis of PCa [2]. To standardize MRI scan parameters and improve i…
-
Quantification of tumor heterogeneity based on fractal dimension for predicting the response to neoadjuvant chemotherapy in triple-negative breast cancer
Triple-negative breast cancer (TNBC) exhibits high heterogeneity, leading to variable responses to neoadjuvant chemotherapy (NAC) among patients. Noninvasive quantification of intratumoral heterogeneity (ITH) may be valuable in predicting treatment response. This study aims to investigate whether fractal dimension (FD) based on pre-treatment contrast-enhanced magnetic resonance imaging (MRI), combined with clinicopathological data, can predict NAC response in TNBC patients.
-
Metal artifact reduction from surgical clips for intracranial aneurysms in photon-counting detector CT angiography
Surgical clipping or endovascular coiling procedures are performed to treat ruptured and unruptured cerebral aneurysms based on the location shape, and size of the aneurysm [1,2]. After these treatments, long-term follow-up imaging every 6–12 months is recommended to monitor potential recurrence. CT angiography (CTA) or MR angiography (MRA) is typically used in such a follow-up [3,4]. However, it is difficult to evaluate blood vessels (including the aneurysm neck) with MRA after clipping, becaus…
-
Reference standard methodology in the clinical evaluation of AI chest X-ray algorithms for lung cancer detection: A systematic review
Lung cancer is associated with a poor prognosis and is the most common cause of cancer death worldwide [1,2]. Recognising and treating lung cancer at an earlier stage is associated with improved survival [3]. In the UK, lung cancer is frequently picked up on chest x-rays (CXR) either as part of an ‘urgent suspicion of cancer’ pathway or as an incidental finding. Whilst the former has national time targets for x-ray turnaround time, the latter is more susceptible to delays due to the high volume …
-
A novel multimodal framework combining habitat radiomics, deep learning, and conventional radiomics for predicting MGMT gene promoter methylation in Glioma: Superior performance of integrated models
The present study aimed to develop a noninvasive predictive framework that integrates clinical data, conventional radiomics, habitat imaging, and deep learning for the preoperative stratification of MGMT gene promoter methylation in glioma.
-
Development and validation of a radiomics model based on the ASPECTS framework using CT imaging for predicting malignant cerebral edema
Ischemic stroke is the most common subtype of stroke, associated with high morbidity, disability, and mortality rates [1,2]. Malignant cerebral edema (MCE) is a life-threatening complication occurring within 2–5 days after a large hemispheric infarction, leading to elevated intracranial pressure, rapid neurological deterioration, and brain herniation and a mortality rate of 40 %-80 % [3,4]. Early identification of patients at risk of developing MCE is critical to enable timely management strateg…
-
Spatial heterogeneity and distribution of CT-Based pulmonary vascular volumes in chronic thromboembolic pulmonary hypertension
Chronic thromboembolic pulmonary hypertension (CTEPH), classified as Group 4 in the World Symposium on Pulmonary Hypertension (WSPH) system, is a relatively uncommon form of pH that is characterized by persistent obstruction of the pulmonary vascular bed due to organized thrombus, neointima formation, and vascular webs[1,2]. Although there are a growing number of effective therapies for this condition (including surgical endarterectomy, endovascular methods, and pharmacologic agents), CTEPH rema…