Artificial intelligence and pelvic fracture diagnosis on X-rays: a preliminary study on performance, workflow integration and radiologists' feedback assessment in a spoke emergency hospital.

Purpose: The aim of our study is to evaluate artificial intelligence (AI) support in pelvic fracture diagnosis on X-rays, focusing on performance, workflow integration and radiologists' feedback in a spoke emergency hospital. Materials and methods: Between August and November 2021, a total of 235 sites of fracture or suspected fracture were evaluated and enrolled in the prospective study. Radiologist's specificity, sensibility accuracy, positive and negative predictive values were compared to AI. Cohen's kappa was used to calculate the agreement between AI and radiologist. We also reviewed the AI workflow integration process, focusing on potential issues and assessed radiologists' opinion on AI via a survey. Results: The radiologist performance in accuracy, sensitivity and specificity was better than AI but McNemar test demonstrated no statistically significant difference between AI and radiologist's performance (p = 0.32). Calculated Cohen's K of 0.64. Conclusion: Contrary to expectations, our preliminary results did not prove a real improvement of patient outcome nor in reporting time but demonstrated AI high NPV (94,62%) and non-inferiority to radiologist performance. Moreover, the commercially available AI algorithm used in our study automatically learn from data and so we expect a progressive performance improvement. AI could be considered as a promising tool to rule-out fractures (especially when used as a "second reader") and to prioritize positive cases, especially in increasing workload scenarios (ED, nightshifts) but further research is needed to evaluate the real impact on the clinical practice.

Imaging assessment of deep neck spaces infections: an anatomical approach.

Despite the widespread use of antibiotics, deep neck infections are still a relevant clinical problem and can cause severe and potentially life-threatening complications. Imaging plays a pivotal role into the clinical management of deep neck infections, allowing identification of the source of infections, definition of the precise extension of the disease and early diagnosis of local or distant complications. The complex anatomy of deep neck spaces may hide the actual extent of disease: the knowledge of neck anatomy enables radiologists to rapidly evaluate expected routes of spread of infections and to effectively communicate crucial information to surgeons. Computed Tomography is the most used imaging modality for assessing primary site of infection, extent of disease and local complications. Magnetic resonance imaging may be used as a second level imaging modality for individuating intracranial or spinal complications. The present work aims to review the imaging of deep neck spaces infections, focusing on relevant anatomy and clinical scenarios, underlining practical teaching points for each of them. Familiarity with deep neck spaces anatomy and knowledge of most common routes of spread of infections allow the radiologist to make a prompt diagnosis and to look for early signs of potential complications.

High-Frequency Intraoral Ultrasound for Preoperative Assessment of Depth of Invasion for Early Tongue Squamous Cell Carcinoma: Radiological-Pathological Correlations.

The eighth edition of the TNM classification officially introduced "depth of invasion" (DOI) as a criterion for determining the T stage in tongue squamous cell carcinoma. The DOI is a well-known independent risk factor for nodal metastases. In fact, several experts strongly suggest elective neck dissection for tongue cancer with a DOI > 4 mm due to the high risk of early and occult nodal metastases. Imaging plays a pivotal role in preoperative assessments of the DOI and, hence, in planning the surgical approach. Intraoral ultrasound (IOUS) has been proposed for early-stage SCC of the oral tongue as an alternative to magnetic resonance imaging (MRI) for local staging. The aim of this work is to investigate the accuracy of IOUS in the assessment of the DOI in early oral SCC (CIS, pT1, and pT2). A total of 41 patients with tongue SCCs (CIS-T2) underwent a preoperative high-frequency IOUS. An IOUS was performed using a small-size, high-frequency hockey-stick linear probe. The ultrasonographic DOI (usDOI) was retrospectively compared to the pathological DOI (pDOI) as the standard reference. In patients who underwent a preoperative MRI, their usDOI, magnetic resonance DOI (mriDOI), and pDOI were compared. Specificity and sensitivity for the IOUS to predict a pDOI > 4 mm and to differentiate invasive and noninvasive tumors were also evaluated. A high correlation was found between the pDOI and usDOI, pDOI and mriDOI, and usDOI and mriDOI (Spearman's ρ = 0.84, p < 0.0001, Spearman's ρ = 0.79, p < 0.0001, and Spearman's ρ = 0.91, p < 0.0001, respectively). A Bland-Altman plot showed a high agreement between the usDOI and pDOI, even though a mean systematic error was found between the usDOI and pDOI (0.7 mm), mriDOI and pDOI (1.6 mm), and usDOI and mriDOI (-0.7 mm). The IOUS was accurate at determining the T stage (p < 0.0001). The sensitivity and specificity for the IOUS to predict a pDOI ≥4 mm were 92.31% and 82.14%, respectively, with an AUC of 0.87 (p < 0.0001). The specificity, sensitivity, negative predictive value (NPV), and positive predictive value (PPV) for the IOUS to predict an invasive cancer were 100%, 94.7%, 60%, and 100%, respectively. The AUC was 0.8 (95% CI 0.646-0.908, p < 0.0001). The IOUS was accurate in a preoperative assessment of a pDOI and T stage, and can be proposed as an alternative to MRI in the preoperative staging of tongue SCC.