Radiology

MRI Pacinian corpuscles in the forefoot may reflect diabetic sensorimotor polyneuropathy severity, with counts correlating to nerve conduction measures.

MRI-derived LACI improves sudden cardiac death risk prediction in hypertrophic cardiomyopathy, adding value beyond ESC guideline models.

Amid a mounting workforce crisis, radiologists have highlighted issues with AI implementation in hospitals at the Health + AI Tech Show 2026.

Pregnancy imaging for acute abdominal pain varies widely across UK services, with gaps in access and pathways prompting calls for national guidance.

Magnetic resonance neurography (MRN) adds value in complex neuropathy cases, particularly when nerve damage is difficult to localise or in cases with proximal nerve involvement.

Epicardial adipose tissue index on cardiac MRI may help identify high-risk HFpEF patients and predict heart failure readmission or death.

LV myocardial strain reference values from CT show sex-based differences in healthy adults, offering clinicians clearer guidance for interpreting cardiac imaging.

A mobile head CT system with integrated shielding reduced staff radiation exposure to minimal or undetectable levels in a neurointensive care study.

CEM demonstrates high accuracy in breast cancer imaging, aligning closely with MRI for tumour size and disease extent assessment.

5.0 T SWI improves brain imaging clarity compared with 3.0 T, enhancing visualisation of cerebral veins and deep grey matter structures in healthy adults.

Limited light exposure in radiology may drive burnout, disrupt sleep, and affect workforce sustainability.

AI fracture detection shows modest gains in identifying child abuse fractures in children under five, highlighting both potential and current limitations.

England delivered 50% fewer PET-CT scans than guidelines recommend in 2021, with gaps for prostate cancer, neuroendocrine tumours, and dementia.

AI-assisted chest X-rays are enhancing TB detection in LMICs, helping clinicians identify cases earlier where imaging access and expertise are limited.

New vessel wall MRI detects more supracardiac atherosclerosis plaques than CTA, improving stroke evaluation and identifying embolic sources.

Targeted MRI accurately assesses new bone pain in stable multiple myeloma, guiding management and reducing unnecessary whole-body scans.

Synthetic radiographs are often hard to distinguish from real X-rays, challenging radiologists and AI while offering training and research opportunities.

IVUS guidance reduces costs and rehospitalisations in lower limb venous stenting, with mixed effects for thrombectomy and thrombolysis procedures.

MULTIPLEX MRI may help identify glioma grade and molecular subtypes in a single scan, offering new insights into tumour biology.

Young adults with rare spinal AVMs show strong recovery after targeted endovascular treatment, guided by MRI and CT imaging of the lesions.

Discordant AI mammography results can sharply increase patient anxiety, reduce trust in radiologists, and raise second-opinion requests, but context eases concern.

Suspended lead suits dramatically reduced radiation exposure for interventional echocardiographers during left atrial appendage occlusion procedures.

Fetal cardiac imaging at 34–38 weeks identified newborns with transposition of the great arteries at high risk of severe hypoxia that would need urgent postnatal intervention.

Analysis of fibroid subtypes reveals unique ultrasound patterns and clinical profiles, helping differentiate usual-type from variant uterine fibroids.

AI tools may help detect vertebral fractures on CT scans, matching expert accuracy and improving osteoporosis care.

The proportion of CT scans performed in patients aged 60–90 has risen over 10 years at a hospital, highlighting concerns about repeated radiation exposure.

Fetoscopic repair of spina bifida reduces brain edema and hindbrain bulging, but postoperative aqueductal obstruction remains a risk, study shows.

A multi-centre study shows AI-assisted CT can reduce reading time and improve consistency in tumour measurements during cancer follow-up imaging.

The PARADIGM study will test whether artificial intelligence can match radiologists in detecting clinically significant prostate cancer on MRI across an international cohort.

ECR 2026 data show that structured reporting systems help improve radiation safety in healthcare X ray practices across Finland.

Half-dose gadopiclenol boosts breast MRI contrast while reducing gadolinium exposure, offering equivalent or better imaging than full-dose gadobutrol.

Age and smoking status influence CT-based body composition in lung cancer screening, with muscle loss and fat gain increasing with age in a large NELSON cohort analysis.

New imaging research links glucose dysregulation with progression of cerebral small vessel disease, suggesting metabolic biomarkers may help identify patients at risk of worsening brain damage.

New deep learning model combining X-ray and MRI data shows promise in increasing efficiency of cervical spondylosis diagnosis.

An estimated 72% to 94% of cases were overdiagnosed in the USA between 1991 and 2019, according to a 2026 study.

Women experience unique changes in brain structure after a second pregnancy, MRIs revealed in a 2026 cohort study.

Integrated PET and MRI scans can distinguish a new type of dementia, LATE, from Alzheimer’s, a 2026 observational study has found.

Fused Deposition Modelling 3D printing has become a practical, accessible, and cost-effective tool in surgical settings, a 2026 systematic review has found.

AI can predict postoperative breast cancer recurrence using preoperative mammograms, a 2026 study has found.

A new AI model was able to read and diagnose brain MRI scans in seconds, researchers at the University of Michigan have found.

Mammographic surveillance and adjunct imaging techniques can lower morality and enhance early detection among breast cancer survivors.

AI-prefilled structured reporting improves diagnostic accuracy in chest radiograph reporting compared with free-text reporting, a 2026 prospective reader study has found.

MRI-based radiomics models may help predict tumour deposit burden in rectal cancer, with a fusion AI approach outperforming radiologists in preoperative risk assessment.

Computer-aided detection chest X-ray performed comparably to experienced radiologists for pulmonary tuberculosis diagnosis, supporting its potential role in low-resource, high-burden clinical settings.

A deep-learning serial CT biomarker accurately predicts overall survival in immunotherapy-treated NSCLC, outperforming RECIST in large real-world and trial cohorts.

Medical societies warn that flawed studies may deter patients from lung cancer screening despite clear survival benefits.

Ultra-high-field 5 T MRI delivers improved image quality and diagnostic accuracy compared with 3 T MRI for prostate cancer, according to a prospective intra-individual study.

New stainless steel neural probes promise safer deep brain sensing, reducing fracture risk while supporting high-resolution recordings for research and surgery.

The new analysis draws on the prospective MALRIC study, including patients treated between 2009 and 2024 across 10 hospitals

Recreational substance use is associated with a broad range of neuroimaging findings, and radiologists play a key role in recognising patterns that influence emergency management and prognosis.

A machine learning lung cancer risk model showed improved accuracy over traditional methods, with potential to enhance patient selection for radiology based lung cancer screening.

Axillary ultrasound and MRI guide SLNB decisions in early-stage breast cancer, allowing safe surgical de-escalation in selected clinically node-negative patients.

The device, referred to as hPAM TUT, has a rigid body measuring 90 mm in length and just 17 mm in diameter, making it suitable for handheld use.

CT is one of the most widely used diagnostic imaging tools, but accurately estimating patient radiation dose remains challenging because of wide variation in body size and anatomy.

A proof-of-concept trial shows novel sarcoidosis PET/CT tracer detects inflammatory activity, supporting further clinical evaluation.

Sequential embolisation and renal cryoablation shows high efficacy and low severe adverse events in patients with renal masses over 3 cm.

More than 386,000 people waited longer than six weeks for testing, accounting for around one in five patients on diagnostic waiting lists.

A novel form of chest X-ray imaging could help clinicians detect pneumothorax more quickly and with greater confidence.

These findings indicate that the low-dose CT component of PET/CT, already performed as part of routine prostate cancer care, can reliably provide information relevant to bone health.

This combined model achieved a high level of accuracy, suggesting that imaging data alongside basic clinical information could be effective for screening.

A study has found that the way medical images are prepared before analysis can have a significant impact on the performance of deep learning models.

Retrospective study in 755 children shows how a deep-learning algorithm can support clinicians correctly diagnose paediatric elbow fractures, although further developments are required.

A prospective study finds that using both diffusion-weighted MRI and spectral Doppler ultrasound significantly improves classification of benign and malignant salivary gland tumours.

Learn how a pilot trial shows low-frequency electrical stimulation with exercise enhances local bone outcomes in premenopausal women.

PS-OCT reveals distinct scleral fibre changes in pathologic myopia, offering new clues to prevent vision loss.

Phase 1 trial shows single-fraction stereotactic breast irradiation is safe and effective

A prospective multicenter study shows an MRI-based AI tool can accurately distinguish PD, MSA and PSP, with results confirmed by neuropathology.

AI missed nearly one-third of breast cancer cases in a new study, but diffusion-weighted MRI detected most of these overlooked tumours, especially in dense breast tissue.

A new study finds many balloon sinus dilation patients undergo the procedure without a recent preoperative CT scan, raising concerns about safety and adherence to guidelines.

MRI-based nested habitat radiomics improves prediction of progression-free survival in spinal tumours, highlighting aggressive tumour subregions.

AI chest X-ray risk scores predict respiratory death risk, increasing 16% per five-year lung-risk score rise.

MRI radiomics model uses pituitary scans to accurately predict growth hormone deficiency in children, providing a non-invasive diagnostic alternative.

A major study exposes high ghost scan rates in trauma ultrasound, prompting urgent calls for improved quality control and clinician training.

Retinal inner layer thinning detected by OCT offers a promising biomarker for monitoring multiple sclerosis progression.

AI algorithm has achieved remarkable accuracy in detecting cardiac amyloidosis, outperforming traditional diagnostic methods.

A PET imaging study reveals widespread synaptic AMPAR increases in long COVID patients with cognitive impairment.

Learn how new trial data show CT colonography offers similar colorectal cancer detection to biennial FIT.

Microwave ablation is a safe, effective treatment for benign thyroid nodules and AFTNs.

Discover how a recent cohort study links medical imaging radiation to increased childhood haematologic cancer risk.

Explore how supplementing DBT with molecular breast imaging significantly improves invasive breast cancer detection in women with dense breasts.

PET-MRE significantly reduced SUVMax and inflammation markers in small bowel Crohn’s disease.

Discover how LAVImin measured by cardiac MRI outperformed traditional risk factors in predicting adverse outcomes after heart attack, offering a new biomarker for patient risk assessment.

Increased tissue magnetic susceptibility in the entorhinal cortex and putamen is significantly associated with the onset of mild cognitive impairment and cognitive decline in cognitively unimpaired older adults.

A deep learning algorithm has shown promise in improving the risk assessment of lung nodules on CT scans, potentially reducing false positives in lung cancer screening.

Endovascular thrombectomy (EVT) is an established treatment for basilar artery occlusion (BAO) strokes, a rare but often devastating subtype of stroke…

Central tumour location linked to worse survival in lung cancer, suggesting need for closer post-operation monitoring.

Understand how an AI model uses chest CT scans to diagnose COPD with 95% accuracy.

Understand how chamber volume ratios on MRI predict cardiac risk, even when heart chambers fall within normal size limits.

Learn how a new scan shows promise for staging invasive lobular breast cancer with 100% specificity in small feasibility study.

Understand how microwave ablation effective and safe for hyperthyroidism, especially with benign nodules

Ultrasound is an effective screening tool for steatotic liver disease, provided that inter- and intra-observer comparisons are incorporated to enhance diagnostic consistency.

Discover how AI detects 32% of missed breast cancers, pointing to improved screening accuracy.

New MRI model predicts early treatment response in TNBC, supporting faster, more tailored clinical decisions.

Visual amyloid-PET scans show 86% agreement with quantitative reads, confirming reliability in real-world dementia care.

Learn how PET study shows safe, targeted brain delivery of intranasal insulin, a critical step for future Alzheimer’s treatments.

Understand how diabetes may raise bone density over time, but fracture risk remains.

A recent prospective study has directly compared different diffusion MRI parameters over the course of NAC to assess their ability to predict pathologic complete response.

Understand how MRI sign accurately identifies Parkinson’s in complex cases, offering a radiation-free alternative to SPECT scans.

Learn how acupuncture offers long-term relief for postconcussion symptoms in mild traumatic brain injury.

Learn how visual training improves radiology accuracy and efficiency, with peripheral focus proving more impactful than central vision methods.

Understand how DVA halves radiation dose in lower limb angiography without compromising image clarity.

Learn how a safety bundle in assisted birth cut trauma and NICU rates, improving outcomes without raising neonatal complications.

Discover how a new study suggests stem cell embolisation is safe and effective for knee osteoarthritis treatment.

Pathologic complete response was accurately predicted by post-treatment MRI in patients with early triple-negative breast cancer having undergone neoadjuvant chemoimmunotherapy.

Study links air pollution to heart scarring, even in healthy individuals, raising concerns for cardiovascular risk management.

Covert cerebrovascular disease may significantly raise fall risk in older adults, even in the absence of symptoms.

MRI-based scoring after cardiac arrest may help predict neurological recovery at discharge in survivors of OHCA.

Learn how CEM after DBT detects more early cancers in high-risk women, though with higher false positive rates.

Understand how MRI-derived RV strain predicts outcomes in HFpEF, offering improved risk stratification over traditional metrics.

Discover how PSMA PET/CT before salvage radiotherapy boosts survival in prostate cancer recurrence.

Understand how POCUS highly accurate for hip effusion diagnosis in children, when performed by trained emergency physicians.

High-resolution ultrasound improves diagnosis of tendon and nerve injuries in forearm lacerations, supporting better trauma care outcomes.

Understand how ultra-low-dose CT reduces radiation and maintains diagnostic accuracy for osteoporosis, for safer, repeated BMD scans.

Subclinical carotid plaques with calcification are more prone to develop dangerous haemorrhages, especially in older men.

AI-guided breast MRI could halve scan times without missing cancers, offering new efficiency for screening programmes.

Understand how multiparametric MRI radiomics predicts HER-2 status in breast cancer with high accuracy.

Learn how dynamic CT detects resistant postpartum haemorrhage, aiding timely interventions and better maternal care.

Study shows DBT improves breast cancer screening outcomes in high-risk women, particularly those with dense breast tissue.

New research shows cystic lung lesions with wall thickening carry a higher cancer risk, though most grow slowly.

Understand how CT metrics match FVC in predicting ALS survival, offering a spirometry alternative for patients with bulbar impairment.

AI accurately identifies osteoporosis from low-dose chest CTs, enabling opportunistic screening during routine imaging procedures.

New study shows radiomic features from mammograms enhance breast cancer prediction, particularly for false-negatives and interval cancers.

Researchers have established standardised global cutoffs for skeletal muscle index and density on abdominal CT in healthy young adults, aiding early sarcopenia detection.

Imaging findings help guide targeted medical therapy for coronary artery disease, improving survival in selected patient groups.

Read how a recent study showed widespread cardiopulmonary abnormalities in long COVID patients one year post-infection.

Understand how ultrasound-guided PDT improves first-puncture success and reduces bleeding risk compared to landmark guidance.

Understand how preoperative MRI does not boost survival in HER2-positive, HR-negative breast cancer

AI model improves heart disease risk prediction by analysing CT scans and identifying ischaemic heart volume automatically.

AI-accelerated knee MRI delivers sub–5-minute scans with high diagnostic accuracy, validated against arthroscopic surgical findings.

AI-driven bone tumour classification model rivals expert radiologists, introducing BTI-RADS 2.0 for structured malignancy risk.

A European study reveals that even low levels of brain amyloid detected by PET scans can signal future Alzheimer’s risk in individuals reporting memory concerns.

Most women support AI in mammography if radiologists remain involved, with education and race influencing acceptance levels.

CT pattern analysis improves diagnosis and predicts immunotherapy success in MSI-H colon cancer. Read more!

Understand how a deep learning model helps radiologists detect endometriosis.

PET/MRI shows higher sensitivity in cancer detection than PET/CT, offering safer imaging for gastric and oesophageal tumours.

UFMRI’s time to enhancement difference helps identify malignant foci, potentially reducing unnecessary breast biopsies.

Standalone AI outperforms radiologists in detecting second breast cancers in postmastectomy patients.

The study highlighted that structural factors, such as neighbourhood-level poverty and limited access to healthcare, continue to drive racial disparities in cancer outcomes.

New study has demonstrated that incorporating MRI-based radiomic features with clinical data significantly improves the prediction of disease upstaging in patients with DCIS.

AI can enhance the efficiency and accuracy of fetal anomaly ultrasound scans, reducing the workload on sonographers while maintaining high diagnostic performance.

New research highlights the significance of timing and tumour location when evaluating PSMA PET scans following radiotherapy in prostate cancer patients.

AI imaging has revealed that female patients are at higher risk of MACE.

Scientists in China report a rare case of brain inflammation linked to influenza, diagnosed using a F-18 FDG-PET/CT scan.

A domain-specific AI model shows high diagnostic accuracy in generating preliminary chest radiograph reports.

Deep-learning (DL) radiopathomics models can help predict the presence of vessels encapsulating tumour clusters (VETC) in hepatocellular carcinoma.

PET/CT scans using an experimental prostate-specific membrane antigen (PSMA) imaging agent can identify SC nodal metastasis in newly diagnosed prostate cancer patients.

The cross-sectional study examined the relationship between brain structure and verbal memory performance.

Yttrium-90 radioembolisation may offer improved survival for large hepatocellular carcinoma tumours compared to transarterial hepatic embolisation.

Results showed that the median radiation dose of ULDCT was only 1.95% of that of normal-dose CT, substantially lowering patient exposure.

A recent meta-analysis synthesised evidence from five clinical validation studies to evaluate AI’s role in CT image quality control.

Explore the rates of screening for breast and colorectal cancer post-pandemic.

A new study has found strong associations between chronic rhinosinusitis and chronic airflow limitation, lower airway symptoms, and smoking-related lung diseases.