Academic Radiology最新文献

Rationale and objectives: To investigate a computed tomography (CT)-based multiparameter deep learning-radiomic model (DLRM) for predicting the preoperative tumor budding (TB) grade in patients with rectal cancer.

Methods: Data from 135 patients with histologically confirmed rectal cancer (85 in the Bd1+2 group and 50 in the Bd3 group) were retrospectively included. Deep learning (DL) features and hand-crafted radiomic (HCR) features were separately extracted and selected from preoperative CT-based extracellular volume (ECV) parameter images and venous-phase images. Six predictive signatures were subsequently constructed from machine learning classification algorithms. Finally, a combined DL and HCR model, the DLRM, was established to predict the TB grade of rectal cancer patients by merging the DL and HCR features from the two image sets.

Results: In the training and test cohorts, the AUC values of the DLRM were 0.976 [95% CI: 0.942-0.997] and 0.976 [95% CI: 0.942-1.00], respectively. The DLRM had good output agreement and clinical applicability according to calibration curve analysis and DCA, respectively. The DLRM outperformed the individual DL and HCR signatures in terms of predicting the TB grade of rectal cancer patients (p < 0.05).

Conclusion: The DLRM can be used to evaluate the TB grade of rectal cancer patients in a noninvasive manner before surgery, thereby providing support for clinical treatment decision-making for these patients.

Rationale and objectives: This study aims to quantitatively evaluate the clinical significance of different grades of cervical lymph node necrosis (CNN). Furthermore, a nomogram was developed and validated to predict overall survival (OS) in patients with lymph node-positive (LN-positive) locally advanced nasopharyngeal carcinoma (LA-NPC), incorporating the different grades of CNN.

Patients and methods: We retrospectively analyzed patients with newly diagnosed, LN-positive LA-NPC at our center from April 2014 to December 2018. Independent predictors were identified through Cox regression analyses, which examined the grade of CNN and other clinical variables associated with OS. Based on the results and a key clinical variable, a nomogram was developed to predict OS. Model performance was evaluated through discrimination, calibration, and clinical utility. Risk stratification was performed using the risk score derived from the nomogram, and the prognoses of two distinct risk groups were compared using the Kaplan-Meier method.

Results: A total of 984 patients were enrolled. Independent predictors for OS, confirmed by multivariate Cox analysis, included age (hazard ratio [HR]: 1.57, 95% CI: 1.09-2.26, P=0.016), Epstein-Barr virus (EBV) DNA (HR: 2.02, 95% CI: 1.40-2.92, P<0.001), N3 (HR: 2.30, 95% CI: 1.42-3.72, P=0.001), Grade of CNN (HR: 1.53, 95% CI: 1.02-2.30, P=0.042), and LDH (HR: 1.48, 95% CI: 1.01-2.15, P=0.043). The nomogram developed by combining these five variables and T stage demonstrated a higher C-index in both the training cohort (0.715 versus 0.624, P<0.001) and validation cohort (0.744 versus 0.629, P<0.001), as well as a higher net clinical benefit compared to the 8th edition TNM staging system (TNM-8).

Conclusion: The grade of CNN is a promising adverse predictor for patients with LA-NPC. Compared to the TNM-8, the nomogram incorporating the Grade of CNN demonstrates superior predictive efficacy and enhanced risk stratification.

Availability of data and material: The data are available from the corresponding author upon request.

Rationale and objectives: The Diagnostic Radiology Resident Physics Curriculum (DRRPC), initiated in 2007 by the American Association of Physicists in Medicine (AAPM) and last updated in 2018, is an essential educational resource for those teaching physics to radiology residents. Regular updates are crucial to ensure the curriculum aligns with evolving technologies and clinical practices, maintaining its relevance and effectiveness in educating the next generation of radiologists. The paper aims to describe the update strategies of the DRRPC, focusing on the current iteration, its structure, and the newest updates.

Materials and methods: The update process, led by the Diagnostic Radiology Resident Physics Curriculum Working Group, commenced with a comprehensive survey targeting AAPM members who contribute to radiology physics teaching. The survey was conducted to assess the curriculum's current applicability and gather feedback for improvements. Subsequent updates were based on extensive stakeholder consultations and detailed analysis of survey data.

Results: The revision process has led to significant enhancements in the curriculum, emphasizing practical clinical applications and the integration of cutting-edge technology. New modules on advanced image processing, artificial intelligence, informatics, and radiopharmaceutical therapy were developed, responding to the evolving needs of radiological education and practice.

Conclusion: The updated DRRPC supports educators in providing a dynamic and relevant training experience.

Rationale and objectives: Accurately and noninvasively predicting lymphovascular invasion (LVI) is critical for the prognosis of patients with rectal cancer (RC). The objective of this study was to create a nomogram model that combines clinical features with MRI-based radiomic characteristics of both intratumoral and peritumoral regions to predict LVI in patients with resectable rectal cancer.

Method: This study retrospectively included 149 RC patients diagnosed with LVI, who were randomly assigned to a training cohort (n=104) and a testing cohort (n=45). Radiomics features were derived from intratumoral and peritumoral areas using different expansion dimensions (3 and 5 mm) in T2-weighted imaging (T2WI) and Diffusion-Weighted Imaging (DWI). A nomogram was created by combining the optimal radiomics model with the most predictive clinical factors to enhance the LVI prediction.

Results: In the validation cohort, the radiomics models using 3 mm and 5 mm peritumoral regions in T2WI achieved AUC values of 0.786 and 0.675, respectively, surpassing the performance of models based on DWI. In both T2WI and DWI, the 3 mm peritumoral model outperformed the 5 mm model in predictive accuracy. Therefore, the combined radiomics model integrating intratumoral and the 3 mm peritumoral regions in T2WI was identified as the optimal radiomics model, achieving an AUC of 0.913. The decision and calibration curves showed that radiomics models incorporating both intratumoral and peritumoral regions outperformed traditional approaches. A nomogram was created by combining a clinical model that incorporates gender and mrN stage with the optional radiomics model, aiming to predict LVI in patients with RC.

Conclusion: The radiomics model based on the 3 mm peritumoral region in T2WI demonstrated greater precision and sensitivity in identifying LVI. The radiomics model, which combined features from both intratumoral and peritumoral regions, exhibited superior performance compared to models based solely on either intratumoral or peritumoral attributes. The optimal combination was the integration of intratumoral features with the 3 mm peritumoral region in T2WI. A nomogram integrating radiomics features from intratumoral and peritumoral regions with clinical parameters offers valuable support for the preoperative diagnosis of LVI in RC, demonstrating significant clinical utility.

Objectives: Systemic sclerosis (SSc) is a chronic autoimmune disease characterized by fibrosis, vascular damage, and immune dysregulation, often leading to muscle abnormalities. This study aimed to evaluate the prevalence of myopenia and myosteatosis in SSc patients using computed tomography (CT) and their associations with clinical features, including lung disease and esophageal dilatation.

Materials and methods: SSc patients followed at Gazi University Rheumatology Clinic (2000-2024) who had thoracic CT imaging were included. Muscle mass and density were assessed at the L1 vertebral level. Skeletal muscle area (SMA) and skeletal muscle radiation attenuation (SMRA) were measured to identify myopenia and myosteatosis. Lung disease involvement and widest esophageal diameter (WED) were assessed via CT. Statistical analyses explored correlations between muscle metrics and clinical variables, with multiple linear regression identifying predictors.

Results: Among 95 patients (54.7% diffuse SSc, 45.3% limited SSc; mean age 57.04 ± 13.65 years; female-to-male ratio 8.5:1), myopenia and myosteatosis prevalence were 27.3% and 41.1%, respectively. Myosteatosis was associated with female sex (p = 0.001), older age (p = 0.001), higher BMI (p = 0.043), and inflammation markers (CRP, ESR). Myopenia correlated with BMI (p = 0.001) but not clinical outcomes. Higher WED correlated with lower SMRA (p = 0.001). BMI predicted muscle mass (R² = 0.42), while age, gender, and BMI determined SMRA (R² = 0.67, p < 0.001).

Conclusions: Myosteatosis was more prevalent and strongly associated with clinical features, including lung disease and esophageal dilatation, than myopenia, underscoring the importance of muscle quality.

Rationale and objectives: Parry-Romberg Syndrome (PRS) and linear morphea en-coup-de-sabre (ECDS) are rare neurocutaneous disorders characterized by unilateral progressive hemifacial atrophy and linear scleroderma, respectively^1,2. Imaging is important for assessing soft tissue and intracranial involvement³, though literature is limited to case reports and series ^2,4-8. We aim to describe radiologic and clinical epidemiologic features of PRS and ECDS.

Materials and methods: A retrospective review of our institutional radiology database identified patients with PRS and ECDS who underwent MRI of the brain and/or face. Clinical data, including neurological symptoms and genetic testing, were collected from electronic medical records. Imaging data included distribution of soft tissue atrophy and signal changes emphasizing orbital, maxillofacial, vascular and intracranial findings.

Results: Among 51 included patients, 24 were diagnosed with PRS, 16 with ECDS, and 11 with both (PRS+ECDS). Females predominated (73%), with mean ages of 30.9 years for PRS, 17.9 for ECDS and 32.9 for PRS+ECDS. The interval between diagnosis and MRI was shorter for ECDS (0.8 years) than PRS (2.9 years) or PRS+ECDS (3.5 years). Seizures occurred in 25% of PRS cases. Intracranial abnormalities were observed in 37% of the cohort. PRS patients showed higher prevalence of masticator space (54%) and salivary gland atrophy, while calvarium thinning (36%) was more frequent in PRS+ECDS. Unlike PRS or PRS+ECDS, the orbits were unaffected in patients with ECDS alone.

Conclusion: We report a higher prevalence of exocrine gland involvement and seizures in PRS, with a prolonged duration between diagnosis and imaging. Comprehensive neuroimaging at the time of diagnosis is essential to determine disease extent, as craniofacial and intracranial findings are prevalent in these patients. Our findings may facilitate early radiologic diagnosis and expedite referral and treatment in patients with PRS and ECDS.

Rationale and objectives: This study aimed to compare CT features of COVID-19 and Influenza A pneumonia, develop a diagnostic differential model, and explore a prognostic model for lesion resolution.

Materials and methods: A total of 446 patients diagnosed with COVID-19 and 80 with Influenza A pneumonitis underwent baseline chest CT evaluation. Logistic regression analysis was conducted after multivariate analysis and the results were presented as nomograms. Machine learning models were also evaluated for their diagnostic performance. Prognostic factors for lesion resolution were analyzed using Cox regression after excluding patients who were lost to follow-up, with a nomogram being created.

Results: COVID-19 patients showed more features such as thickening of bronchovascular bundles, crazy paving sign and traction bronchiectasis. Influenza A patients exhibited more features such as consolidation, coarse banding and pleural effusion (P < 0.05). The logistic regression model achieved AUC values of 0.937 (training) and 0.931 (validation). Machine learning models exhibited area under the curve values ranging from 0.8486 to 0.9017. COVID-19 patients showed better lesion resolution. Independent prognostic factors for resolution at baseline included age, sex, lesion distribution, morphology, coarse banding, and widening of the main pulmonary artery.

Conclusion: Distinct imaging features can differentiate COVID-19 from Influenza A pneumonia. The logistic discriminative model and each machine - learning network model constructed in this study demonstrated efficacy. The nomogram for the logistic discriminative model exhibited high utility. Patients with COVID-19 may exhibit a better resolution of lesions. Certain baseline characteristics may act as independent prognostic factors for complete resolution of lesions.