The implementation of AI-enabled solutions, either in synergy with human users or as a standalone tool, have exhibited wide performance variability in different clinical contexts and scenarios [1,2]. This highlights the importance of careful planning for the integration of AI in clinical workflows, of continued post-market surveillance, and of simultaneously investing on AI literacy building within healthcare organizations [3,4]. Different challenges exist when implementing AI solutions in clini…
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.
-
Response to the Letter to the Editor: Diagnostic performance of abbreviated breast MRI in differentiating intraductal papilloma from ductal secretion
We thank you for the opportunity to respond to the letter submitted regarding our recently published manuscript entitled: “Diagnostic performance of abbreviated breast MRI in differentiating intraductal papilloma from ductal secretion”.
-
Complementing interpretable machine learning with synergistic analytical strategies for thyroid cancer recurrence prediction
This correspondence critically examines the methodology of Schindele et al. (2025) on thyroid cancer recurrence prediction. While their interpretable XGBoost model achieved a high predictive accuracy of 95.8% and a 0.947 AUROC, it is crucial to recognize that this predictive power does not justify the reliability of its derived feature importance rankings. As widely acknowledged in the literature, high predictive accuracy does not guarantee unbiased or reliable feature attribution. We underscore…
-
A novel segmentation-based deep learning model for enhanced scaphoid fracture detection
The scaphoid is a biomechanically significant and the most frequently fractured carpal bone [1–3]. Up to 20 % of scaphoid fractures do not appear on initial plain radiographs and are considered occult [4–6]. When a scaphoid fracture is clinically suspected but not evident on initial radiographs, further imaging studies with computed tomography (CT), cone-beam computed tomography (CBCT) or magnetic resonance imaging (MRI) or immobilization followed by repeat radiographs after two weeks are necess…
-
Evaluating the efficacy of 8Spheres microsphere combined with Iodine-125 seeds in the treatment of advanced lung squamous cell carcinoma presenting with hemoptysis: A retrospective study
Hemoptysis is a critical complication in advanced lung squamous cell carcinoma (LUSC). This study aims to evaluate the effectiveness and safety of 8Spheres microspheres embolization combined with iodine-125 seed implantation for managing hemoptysis in patients with advanced LUSC.
-
Comment on the diagnostic impact of BPE in contrast-enhanced mammography: is BPE in CEM really that problematic?
I read your recent article with interest by Bechyna et al. [1] examining the diagnostic implications of background parenchymal enhancement (BPE) in contrast-enhanced mammography (CEM). It can be considered a pilot and important study on CEM, which has recently started showing its strong effects. Your study emphasizes the significant negative effect of moderate to marked BPE on the diagnostic performance of CEM, especially in premenopausal women with dense breasts. However, there are a few import…
-
The influence of anatomical shape variations of wrist bones on bone orientation values in CT scans
Wrist injuries, like scapholunate ligament (SL) lesions, might cause abnormal kinematics of the wrist joint, which progress to osteoarthritis if left untreated. However, due to the complexity of the wrist joint, composed of eight carpal bones, two forearm bones, and multiple extrinsic and intrinsic ligaments, an accurate diagnosis of pathology is often challenging [1–4].
-
Accuracy evaluation of an automatic landmarking method based on 3D segmented CT scans of full lower limbs with knee osteoarthritis
Standards of care in orthopaedic lower limb surgery have evolved substantially over recent decades. Improvements in surgical tools, implants and accurate pre-operative planning have enhanced surgeon precision. Navigation systems[1,2], patient-specific instrumentation[3] and 3-Dimensional (3D) image-based planning[4] were developed in the late 1990 s and have been used primarily for joint replacements and osteotomies. Moreover, knee arthroplasty (KA) has seen a paradigm shift over the last decade…
-
Long-term outcomes of carotid artery stenting for carotid near-occlusion
Carotid artery near-occlusion (CANO) is a severe form of internal carotid artery (ICA) stenosis marked by marked narrowing of the post-stenotic distal ICA[1,2]. Imaging features of CANO include marked post-stenotic narrowing of the distal ICA, often accompanied by reduced flow, delayed contrast transit, and the presence of intracranial collateral circulation[1]. The degree of distal ICA narrowing in CANO can vary, ranging from a threadlike, markedly narrowed distal ICA in cases with full collaps…