Marc Zins

Purpose To develop and externally validate a multimodal AI model for detecting ischaemia complicating small-bowel obstruction (SBO). Methods We combined 3D CT data with routine laboratory markers (C-reactive protein, neutrophil count) and, optionally, radiology report text. From two centers, 1,350 CT examinations were curated; 771 confirmed SBO scans were used for model development with patient-level splits. Ischemia labels were defined by surgical confirmation within 24 hours of imaging. Models (MViT, ResNet-101, DaViT) were trained as unimodal and multimodal variants. External testing was used for 66 independent cases from a third center. Two radiologists (attending, resident) read the test set with and without AI assistance. Performance was assessed using AUC, sensitivity, specificity, and 95% bootstrap confidence intervals; predictions included a confidence score. Results The image-plus-laboratory model performed best on external testing (AUC 0.69 [0.59–0.79], sensitivity 0.89 [0.76–1.00], and specificity 0.44 [0.35–0.54]). Adding report text improved internal validation but did not generalize externally; image+text and full multimodal variants did not exceed image+laboratory performance. Without AI, the attending outperformed the resident (AUC 0.745 [0.617–0.845] vs 0.706 [0.581–0.818]); with AI, both improved, attending 0.752 [0.637–0.853] and resident 0.752 [0.629–0.867], rising to 0.750 [0.631–0.839] and 0.773 [0.657–0.867] with confidence display; differences were not statistically significant. Conclusion A multimodal AI that combines CT images with routine laboratory markers outperforms single-modality approaches and boosts radiologist readers’ performance notably junior, supporting earlier, more consistent decisions within the first 24 hours. Key Points A multimodal artificial intelligence (AI) model that combines CT images with laboratory markers detected ischemia in small-bowel obstruction with AUC 0.69 (95% CI 0.59–0.79) and sensitivity 0.89 (0.76–1.00) on external testing, outperforming single-modality models. Adding report text did not generalize across sites: the image+text model fell from AUC 0.82 (internal) to 0.53 (external), and adding text to image+biology left external AUC unchanged (0.69) with similar specificity (0.43–0.44). With AI assistance both junior and senior readers improved; the junior’s AUC rose from 0.71 to 0.77, reaching senior-level performance. Summary Statement A multicentric AI model combining CT and routine laboratory data (CRP and neutrophilia) improved radiologists’ detection of ischemia in small-bowel obstruction. This tool supports earlier decision-making within the first 24 hours.

Perineal suppuration encompasses a broad spectrum of pathologies involving the anatomically complex perineal region. While anal fistula, with or without abscess, is the most frequent cause, infections may also originate from the skin, accessory urinary or genital glands, pelvic and perineal implants, or supralevator digestive structures. Clinical presentations range from benign acute episodes to severe life-threatening conditions or chronic relapsing infections. Diagnosis is often challenging due to overlapping symptoms and variable anatomical involvement. Imaging plays a pivotal role in delineating the extent and origin of disease, particularly in distinguishing anal from extra-anal sources, an essential step for appropriate management. Ultrasonography (US) and Magnetic Resonance Imaging (MRI) are well suited for evaluating anal fistulas; MRI is preferred for chronic, recurrent, or deep-seated infections, particularly those of cutaneous origin or associated with implant infection, while Computed tomography (CT) is favored for supralevator digestive involvement and Fournier’s gangrene. A detailed understanding of perineal anatomy and its imaging correlates is crucial for accurate diagnosis and treatment planning. This pictorial essay illustrates the imaging features of perineal suppurations from diverse etiologies, emphasizing the diagnostic value of cross-sectional imaging and the central role of radiology in clinical decision-making.

Hepatocellular carcinoma (HCC) is the most common primary hepatic malignancy and a leading cause of cancer related death worldwide. Current guidelines for the noninvasive diagnosis of HCC are provided by the European Association for the Study of the Liver (EASL), the American Association for the Study of Liver Diseases (AASLD) which endorsed the Liver Imaging Reporting and Data System (LI-RADS) algorithm, the Korean Liver Cancer Association-National Cancer Center (KLCA-NCC), and the Asian-Pacific Association for the Study of the Liver (APASL). These allow the diagnosis of HCC in high-risk patients in the presence of typical imaging features on contrast-enhanced CT, MRI, or contrast-enhanced ultrasound. Size, non-rim arterial phase hyperenhancement, non-peripheral washout, enhancing capsule, and growth are major imaging features and they should be combined for the diagnosis of HCC. This article provides concise and relevant practice recommendations aimed at general radiologist audience, summarizing the best practice and informing on the essential imaging criteria for the diagnosis of HCC, while also discussing the high-risk population criteria, imaging modalities, and imaging features according to the current guidelines. Key Points: • Noninvasive diagnosis of hepatocellular carcinoma (HCC) can be provided only in patients at high risk. • Contrast-enhanced CT or MRI are the first-line imaging exams for the diagnosis of HCC. • Major imaging features should be combined to provide the diagnosis of definitive HCC.

Sacrocolpopexy and rectopexy are commonly used surgical options for treatment of patients with pelvic organ and rectal prolapse, respectively. These procedures involve surgical fixation of the vaginal vault or the rectum to the sacral promontory with mesh material and can be performed independently of each other or in a combined fashion and by using an open abdominal approach or laparoscopy with or without robotic assistance. Radiologists can be particularly helpful in cases where patients' surgical histories are unclear by identifying normal sacrocolpopexy or rectopexy mesh material and any associated complications. Acute complications such as bleeding or urinary tract injury or stricture are generally evaluated with CT. More chronic complications such as mesh extrusion or exposure with or without fistulization to surrounding structures are generally evaluated with MRI. Other complications can have a variable time of onset after surgery. Patients with suspected bowel obstruction are generally evaluated with CT. Those with suspected infection, abscess formation, and discitis or osteomyelitis may be evaluated with MRI, although CT evaluation may be appropriate in certain scenarios. The authors review the sacrocolpopexy and rectopexy surgical techniques, discuss appropriate imaging protocols for evaluation of patients with suspected complications, and illustrate the normal appearance and common complications of these procedures. ^© RSNA, 2023 Quiz questions for this article are available in the supplemental material.