Mammographic density assessment: comparison of radiologists, automated volumetric measurement, and artificial intelligence-based computer-assisted diagnosis.

BACKGROUND: Artificial intelligence-based computer-assisted diagnosis (AI-CAD) is increasingly used for mammographic exams, and its role in mammographic density assessment should be evaluated. PURPOSE: To assess the inter-modality agreement between radiologists, automated volumetric density measurement program (Volpara), and AI-CAD system in breast density categorization using the Breast Imaging-Reporting and Data System (BI-RADS) density categories. MATERIAL AND METHODS: A retrospective review was conducted on 1015 screening digital mammograms that were performed in Asian female patients (mean age = 56 years ± 10 years) in our health examination center between December 2022 and January 2023. Four radiologists with two different levels of experience (expert and general radiologists) performed density assessments. Agreement between the radiologists, Volpara, and AI-CAD (Lunit INSIGHT MMG) was evaluated using weighted kappa statistics and matched rates. RESULTS: Inter-reader agreement between expert and general radiologists was substantial (k = 0.65) with a matched rate of 72.8%. The agreement was substantial between expert or general radiologists and Volpara (k = 0.64-0.67) with a matched rate of 72.0% but moderate between expert or general radiologists and AI-CAD (k = 0.45-0.58) with matched rates of 56.7%-67.0%. The agreement between Volpara and AI-CAD was moderate (k = 0.53) with a matched rate of 60.8%. CONCLUSION: The agreement in breast density categorization between radiologists and automated volumetric density measurement program (Volpara) was higher than the agreement between radiologists and AI-CAD (Lunit INSIGHT MMG).

Small (≤5 cm) soft tissue tumors of the extremity and superficial trunk: MRI features and demographic characteristics associated with malignancy.

BACKGROUND: Although a substantial proportion of small soft tissue tumors are malignant, magnetic resonance imaging (MRI) features and demographic characteristics associated with these tumors have not been well described. PURPOSE: To investigate the MRI features and demographic characteristics associated with small (≤5 cm) malignant soft tissue tumors, and to identify independent predictors that allow differentiation of small benign and malignant soft tissue tumors. MATERIAL AND METHODS: This retrospective study evaluated patients who underwent surgical excision of small soft tissue tumors of the extremities and superficial trunk, and preoperative contrast-enhanced MRI. Seven MRI findings (tumor depth, tumor-fascia relationship, heterogeneity of signal intensity, necrosis, peritumoral edema, peritumoral enhancement, and margin) and two demographic parameters (age and sex) were included in univariate and multivariate logistic regression analyses to identify independent predictors of small malignant soft tissue tumors. RESULTS: A total of 221 patients (102 men; mean age=45.6 ± 17.6 years) with 72 malignant and 149 benign tumors were included. In the univariate analysis, peritumoral edema (odds ratio [OR] = 3.854; P < 0.001) and peritumoral enhancement (OR = 3.966; P < 0.001) and patient age (≥46 years) (OR = 2.154; P = 0.009) were significantly associated with malignancy. Multivariate analysis showed that peritumoral enhancement on MRI (OR = 3.728; P < 0.001) and patient age (≥46 years) (OR = 1.907; P = 0.036) were independent predictors of malignancy. The combination of these two parameters showed accuracy of 75.1%, sensitivity of 55.6%, and specificity of 84.6% to predict malignancy. CONCLUSION: Among several MRI and demographic features, the presence of peritumoral enhancement on MRI and patient age (≥46 years) were independent predictors of malignancy in small soft tissue tumors.

Venous Sinus Thrombosis in the Hypoglossal Canal Mimics a Neurogenic Tumor in a Patient with Presumed Idiopathic Intracranial Hypertension: A Case Report.

Presumed idiopathic intracranial hypertension (IIH) is a disorder of elevated intracranial pressure with unknown etiology, and 10% of cases occur secondarily to cerebral venous sinus thrombosis (CVST). CVST may be underestimated when findings of IIH are missed in a normal-weight patient without risk factors of coagulopathy. Here, we present a case of CVST that mimics a neurogenic tumor in the hypoglossal canal in a normal-weight patient without risk factors of coagulopathy.

Content-based Image Retrieval by Using Deep Learning for Interstitial Lung Disease Diagnosis with Chest CT.

Background Evaluation of interstitial lung disease (ILD) at CT is a challenging task that requires experience and is subject to substantial interreader variability. Purpose To investigate whether a proposed content-based image retrieval (CBIR) of similar chest CT images by using deep learning can aid in the diagnosis of ILD by readers with different levels of experience. Materials and Methods This retrospective study included patients with confirmed ILD after multidisciplinary discussion and available CT images identified between January 2000 and December 2015. Database was composed of four disease classes: usual interstitial pneumonia (UIP), nonspecific interstitial pneumonia (NSIP), cryptogenic organizing pneumonia, and chronic hypersensitivity pneumonitis. Eighty patients were selected as queries from the database. The proposed CBIR retrieved the top three similar CT images with diagnosis from the database by comparing the extent and distribution of different regional disease patterns quantified by a deep learning algorithm. Eight readers with varying experience interpreted the query CT images and provided their most probable diagnosis in two reading sessions 2 weeks apart, before and after applying CBIR. Diagnostic accuracy was analyzed by using McNemar test and generalized estimating equation, and interreader agreement was analyzed by using Fleiss κ. Results A total of 288 patients were included (mean age, 58 years ± 11 [standard deviation]; 145 women). After applying CBIR, the overall diagnostic accuracy improved in all readers (before CBIR, 46.1% [95% CI: 37.1, 55.3]; after CBIR, 60.9% [95% CI: 51.8, 69.3]; P < .001). In terms of disease category, the diagnostic accuracy improved after applying CBIR in UIP (before vs after CBIR, 52.4% vs 72.8%, respectively; P < .001) and NSIP cases (before vs after CBIR, 42.9% vs 61.6%, respectively; P < .001). Interreader agreement improved after CBIR (before vs after CBIR Fleiss κ, 0.32 vs 0.47, respectively; P = .005). Conclusion The proposed content-based image retrieval system for chest CT images with deep learning improved the diagnostic accuracy of interstitial lung disease and interreader agreement in readers with different levels of experience. © RSNA, 2021 Online supplemental material is available for this article. See also the editorial by Wielpütz in this issue.

Supplementary Anal Imaging by Magnetic Resonance Enterography in Patients with Crohn's Disease Not Suspected of Having Perianal Fistulas.

BACKGROUND & AIMS: Few data are available to guide the use of anal imaging for patients with Crohn's disease (CD) who are not suspected of having perianal fistulas. We aimed to evaluate the role of anal imaging supplementary to magnetic resonance enterography (MRE) in these patients. METHODS: In a prospective study, we added a round of anal MR imaging (MRI), collecting axial images alone, to MRE evaluation of 451 consecutive adults who were diagnosed with or suspected of having CD but not believed to have perianal fistulas. Images were examined for perianal tracts; if present, colorectal surgeons reexamined patients to identify external openings or perianal inflammation or abscess. Patients were followed and data were collected on dedicated treatment for perianal fistulas or abscess. We calculated the diagnostic yield for anal MRI, associated factors, and outcomes of MRI-detected asymptomatic perianal tracts. RESULTS: A total of 440 patients (mean age, 29.6±8.9 years) met the inclusion criteria. Anal MRI revealed perianal tracts in 53 patients (12%; 95% CI, 9.3%-15.4%). Surgeons however did not identify any lesions that required treatment. The asymptomatic tracts were mostly single unbranched (83%), inter-sphincteric (72%), or had a linear dark signal at the tract margin (79%). Younger age at MRE, female sex, and CD activity index scores of 220-450 were independently associated with detection of perianal tracts. MRI detection of asymptomatic tracts was independently associated with later development of perianal fistulas or abscess that required treatment: 17.8% cumulative incidence at 37 months and an adjusted hazard ratio of 3.06 (95% CI, 1.01-9.27; P = .048). CONCLUSIONS: In a prospective study of patients with CD, we found that adding anal MRI evaluation to MRE resulted in early identification of patients at risk for perianal complications.