The aims of this study were to evaluate the safety, feasibility and tumour response of irreversible electroporation, a non-thermal ablation technique, for the treatment of unresectable hepatocellular carcinoma. The endpoints were safety and local treatment efficacy. Patients with unresectable tumours and tumours not amenable for radiofrequency ablation because of their vicinity to organs vulnerable to thermal damage such as the bowel or because they were close to large blood vessels that would limit efficacy of ablation due to the heat sink effect were treated with irreversible electroporation using percutaneous ultrasound and/or computed tomography guided electrode placement between November 2008 and December 2009. Early, late, minor and major complications were recorded. Tumour response was determined on triphasic helical computed tomography follow-up at one month, then every three months post-procedure. Eleven patients received IRE therapy to 18 HCC lesions (Mean diameter 2.44 ± 0.99 cm; range 1.0–6.1 cm) with five patients having more than one treated HCC. Mean follow-up was 18 months (range 14–24 months). Six patients required repeat treatments for local residual or recurrent disease; two of these also had IRE for distant intrahepatic recurrence. No serious complications were observed despite seven lesions lying adjacent to important structures or organs. Four patients developed transient urinary retention and seven developed transient local post-procedure pain. After IRE therapy, 13 (72%) lesions were completely ablated with 93% success for lesions <3 cm (13/14). The local recurrence-free period was 18 ± 4 months and the distance recurrence free period was 14 ± 6 months. These preliminary results suggest that IRE is a safe and feasible technique for local ablation of HCC, particularly for lesions less than 3 cm. No major complications were encountered during this study even for tumours close to essential structures or organs.

Objectives: Trauma chest radiographs may contain subtle and time-critical pathology. Artificial intelligence (AI) may aid in accurate reporting, timely identification and worklist prioritisation. However, few AI programs have been externally validated. This study aimed to evaluate the performance of a commercially available deep convolutional neural network – Annalise CXR V1.2 (Annalise.ai) – for detection of traumatic injuries on supine chest radiographs. Methods: Chest radiographs with a CT performed within 24 h in the setting of trauma were retrospectively identified at a level one adult trauma centre between January 2009 and June 2019. Annalise.ai assessment of the chest radiograph was compared to the radiologist report of the chest radiograph. Contemporaneous CT report was taken as the ground truth. Agreement with CT was measured using Cohen’s κ and sensitivity/specificity for both AI and radiologists were calculated. Results: There were 1404 cases identified with a median age of 52 (IQR 33–69) years, 949 males. AI demonstrated superior performance compared to radiologists in identifying pneumothorax (p = 0.007) and segmental collapse (p = 0.012) on chest radiograph. Radiologists performed better than AI for clavicle fracture (p = 0.002), humerus fracture (p < 0.0015) and scapula fracture (p = 0.014). No statistical difference was found for identification of rib fractures and pneumomediastinum. Conclusion: The evaluated AI performed comparably to radiologists in interpreting chest radiographs. Further evaluation of this AI program has the potential to enable it to be safely incorporated in clinical processes. Advances in knowledge: Clinically useful AI programs represent promising decision support tools.

Background: Pancreatic neuroendocrine neoplasms (PNENs) are rare but clinically significant tumors with variable radiological presentations that complicate diagnosis. While typical PNENs are well characterized, atypical features, such as cystic or hypoenhancing patterns, are less understood and can lead to diagnostic delays or misdiagnosis. This study aimed to evaluate atypical radiological presentations of PNENs, focusing on their impact on diagnostic pathways and differentiation from other pancreatic pathologies. Methods: A retrospective review was conducted of all PNEN cases diagnosed at a single tertiary center between 2010 and 2020. Cases with histopathological confirmation and available cross-sectional imaging were included. Radiological features were categorized as typical (solid and hyperenhancing) or atypical (cystic and hypoenhancing). Demographic, radiological, and pathological data were analyzed. Comparisons between typical and atypical PNENs were performed using descriptive and inferential statistics. Results: Among 77 PNEN cases, 39 met the inclusion criteria. Atypical radiological presentations were identified in 46% of cases, including cystic (18%) and hypoenhancing (28%) lesions. Hypoenhancing PNENs were significantly more likely to present with advanced disease (54% vs. 14% in typical PNENs, p = 0.016). In contrast, none of the cystic PNENs exhibited advanced disease. Atypical PNENs posed greater diagnostic challenges, with alternative diagnoses initially considered in 64% of hypoenhancing and 43% of cystic cases compared to 10% of typical PNENs (p = 0.0042). Conclusions: Atypical PNENs, particularly hypoenhancing lesions, present significant diagnostic challenges and are more likely to be associated with advanced disease. These findings highlight the need for improved recognition of atypical imaging patterns and more precise diagnostic strategies. However, the retrospective design and small cohort size limit the generalizability of our findings. Further multicenter studies are warranted to refine the imaging criteria and optimize the differentiation from other pancreatic neoplasms.