
Purpose: To evaluate the diagnostic performance of radiologists, with and without an artificial intelligence tool, at a tertiary cancer centre. Methods: The institutional review board considered this study at a tertiary cancer centre as a quality improvement project. Pre-biopsy ultrasound images of thyroid nodules obtained by a single interventional radiologist between July 2023 and April 2025 were retrospectively evaluated by five subspecialised diagnostic radiologists independently for thyroid imaging reporting and data system assessment, first without and then with the artificial intelligence tool. Standalone artificial intelligence assessments were also recorded. Diagnostic performance metrics were assessed, including conditional sensitivity, conditional specificity and conditional area under the curve. Results: Ninety nodules from 73 patients (50/73, 68%, female; mean age, 56 ± 17 years) were included. Pathology and genetic testing classified 46/90 nodules (51%) as benign, 31/90 (34%) as malignant and 13/90 (14%) as indeterminate. Compared to radiologist interpretation alone, artificial intelligence assistance yielded slight improvements in conditional specificity (44% [95% confidence interval: 22%, 69%] vs 38% [95% confidence interval: 15%, 65%]), high risk predictive value (58% [95% confidence interval: 37%, 78%] vs 57% [95% confidence interval: 35%, 77%]), area under the receiver operating characteristic curve (0.78 vs 0.75), overall test yield (42% [95% confidence interval: 30%, 53%] vs 38% [95% confidence interval: 27%, 49%]) in classifying benign and malignant nodules, though not statistically significant. Conclusion: Combined radiologist and artificial intelligence tool interpretations achieved similar diagnostic performance to interpretation by radiologists alone or Koios DS Thyroid alone, demonstrating utility of artificial intelligence–assisted interpretation in a specialised cancer centre with a high prevalence of rare and aggressive thyroid and non-thyroid malignancies.
Objective: To evaluate the diagnostic accuracy of ultrasound-derived peak vein velocity ratio for detecting central vein stenosis in the innominate vein resulting from long-term non-cuffed double-lumen haemodialysis catheters. Methods: A cross-sectional study was conducted on haemodialysis patients with right internal jugular vein catheters. All subjects underwent duplex ultrasound to calculate peak vein velocity ratio, defined as the ratio of peak velocity in the innominate vein to the pre-stenotic internal jugular vein. Venography was performed as the gold standard. Results: Of 50 patients analysed, innominate vein stenosis prevalence was 68%. Diabetes mellitus was significantly associated with stenosis (odds ratio = 5.16; p = 0.049). An innominate-to-internal jugular vein peak vein velocity ratio cutoff of ⩾1.65 identified innominate vein stenosis (>30% luminal narrowing) with 97.1% sensitivity, 56.2% specificity, and an area under the curve of 0.71. Conclusion: Peak vein velocity ratio shows high sensitivity and negative predictive value as a screening tool for catheter-related innominate vein stenosis. A cutoff of 1.65 makes significant stenosis unlikely when not exceeded, supporting a possible role as a non-invasive triage step before venography in resource-limited settings, pending external validation.
Introduction: Harlequin ichthyosis is a rare and severe congenital skin disorder of autosomal recessive origin. Ichthyosis represents a group of inherited disorders characterised by abnormal keratinisation and skin barrier dysfunction. Sonographers, fetal medicine specialists and radiologists need to be aware of this condition as it presents as an evolving fetal anomaly. Case presentation: We report a 27-year-old pregnant female who presented for a routine antenatal scan at 29 weeks. The ultrasound examination revealed a single live fetus with thick skin, abnormal facial features and amniotic fluid containing internal echoes. Based on these ultrasound features, the condition was diagnosed as harlequin ichthyosis. After genetic counselling, the couple opted to discontinue the pregnancy and underwent medical termination of pregnancy. The morphology of the phenotype confirmed the ultrasound imaging findings. Discussion: Harlequin ichthyosis is a rare and severe congenital ichthyosis of autosomal recessive origin. The condition is characterised by plate-like scales covering the entire body, typical facial features such as ectropion, eclabium, flat nose, fixed flexion deformities of the extremities and amniotic fluid with internal echoes. Ultrasound is the optimal first-line imaging modality for assessment of these features. The combined use of two-dimensional and three-dimensional ultrasound imaging aids in better depiction of the facial morphology. Detailed genetic evaluation is recommended to assess gene mutations and predict recurrence risk in subsequent pregnancies. Conclusion: While harlequin ichthyosis is rare, recognising the sonographic features early can help in genetic counselling and guide management, empowering obstetricians.