Insights into Imaging最新文献

Introduction: Artificial intelligence (AI) in radiology is a rapidly evolving field. In breast imaging, AI has already been applied in a real-world setting and multiple studies have been conducted in the area. The aim of this analysis is to identify the most influential publications on the topic of artificial intelligence in breast imaging.

Methods: A retrospective bibliometric analysis was conducted on artificial intelligence in breast radiology using the Web of Science database. The search strategy involved searching for the keywords 'breast radiology' or 'breast imaging' and the various keywords associated with AI such as 'deep learning', 'machine learning,' and 'neural networks'.

Results: From the top 100 list, the number of citations per article ranged from 30 to 346 (average 85). The highest cited article titled 'Artificial Neural Networks In Mammography-Application To Decision-Making In The Diagnosis Of Breast-Cancer' was published in Radiology in 1993. Eighty-three of the articles were published in the last 10 years. The journal with the greatest number of articles was Radiology (n = 22). The most common country of origin was the United States (n = 51). Commonly occurring topics published were the use of deep learning models for breast cancer detection in mammography or ultrasound, radiomics in breast cancer, and the use of AI for breast cancer risk prediction.

Conclusion: This study provides a comprehensive analysis of the top 100 most-cited papers on the subject of artificial intelligence in breast radiology and discusses the current most influential papers in the field.

Clinical relevance statement: This article provides a concise summary of the top 100 most-cited articles in the field of artificial intelligence in breast radiology. It discusses the most impactful articles and explores the recent trends and topics of research in the field.

Key points: Multiple studies have been conducted on AI in breast radiology. The most-cited article was published in the journal Radiology in 1993. This study highlights influential articles and topics on AI in breast radiology.

The reported incidence of meniscal tears in the pediatric age group has increased because of increased sports participation and more widespread use of MRI. Meniscal injury is one of the most commonly reported internal derangements in skeletally immature knees and can be associated with early degenerative joint disease leading to disability. The pediatric meniscus has particularities, and knowledge of normal anatomy, anatomical variations, and the patterns of meniscal injury in the pediatric age group is essential to provide a correct diagnosis. CRITICAL RELEVANCE STATEMENT: Accurate MRI interpretation of pediatric meniscal injuries is crucial. Understanding age-specific anatomy, vascularity, and variations can improve diagnostic precision, guiding targeted treatments to prevent early joint degeneration and disability. KEY POINTS: Meniscal lesions are common injuries in skeletally immature knees. Awareness of anatomical meniscus variants, patterns of injury, and associated injuries is essential. Meniscal tears in pediatric patients should be repaired if possible.

Objectives: Radical prostatectomy (RP) is a common intervention in patients with localized prostate cancer (PCa), with nerve-sparing RP recommended to reduce adverse effects on patient quality of life. Accurate pre-operative detection of extraprostatic extension (EPE) remains challenging, often leading to the application of suboptimal treatment. The aim of this study was to enhance pre-operative EPE detection through multimodal data integration using explainable machine learning (ML).

Methods: Patients with newly diagnosed PCa who underwent [⁶⁸Ga]Ga-PSMA-11 PET/MRI and subsequent RP were recruited retrospectively from two time ranges for training, cross-validation, and independent validation. The presence of EPE was measured from post-surgical histopathology and predicted using ML and pre-operative parameters, including PET/MRI-derived features, blood-based markers, histology-derived parameters, and demographic parameters. ML models were subsequently compared with conventional PET/MRI-based image readings.

Results: The study involved 107 patients, 59 (55%) of whom were affected by EPE according to postoperative findings for the initial training and cross-validation. The ML models demonstrated superior diagnostic performance over conventional PET/MRI image readings, with the explainable boosting machine model achieving an AUC of 0.88 (95% CI 0.87-0.89) during cross-validation and an AUC of 0.88 (95% CI 0.75-0.97) during independent validation. The ML approach integrating invasive features demonstrated better predictive capabilities for EPE compared to visual clinical read-outs (Cross-validation AUC 0.88 versus 0.71, p = 0.02).

Conclusion: ML based on routinely acquired clinical data can significantly improve the pre-operative detection of EPE in PCa patients, potentially enabling more accurate clinical staging and decision-making, thereby improving patient outcomes.

Critical relevance statement: This study demonstrates that integrating multimodal data with machine learning significantly improves the pre-operative detection of extraprostatic extension in prostate cancer patients, outperforming conventional imaging methods and potentially leading to more accurate clinical staging and better treatment decisions.

Key points: Extraprostatic extension is an important indicator guiding treatment approaches. Current assessment of extraprostatic extension is difficult and lacks accuracy. Machine learning improves detection of extraprostatic extension using PSMA-PET/MRI and histopathology.

Objective: Investigate the association between the relative tumor enhancement (RTE) of gadoxetic acid across various MRI phases and immunohistochemical (IHC) features in patients with liver metastases (LM) from colorectal cancer (CRC), breast cancer (BC), and pancreatic cancer (PC).

Methods: A retrospective analysis was conducted on 68 patients with LM who underwent 1.5-T MRI scans. Non-contrast and contrast-enhanced T1-weighted (T1-w) gradient echo (GRE) sequences were acquired before LM biopsy. RTE values among LM groups were compared by cancer type using analysis of variance. The relationships between RTE and IHC features tumor stroma ratio, cell count, Ki67 proliferation index, and CD45 expression were evaluated using Spearman's rank correlation coefficients.

Results: Significant differences in RTE were observed across different MRI phases among patients with BCLM, CRCLM, and PCLM: arterial phase (0.75 ± 0.42, 0.37 ± 0.36, and 0.44 ± 0.19), portal venous phase (1.09 ± 0.41, 0.59 ± 0.44, and 0.53 ± 0.24), and venous phase (1.11 ± 0.45, 0.65 ± 0.61, and 0.50 ± 0.20). In CRCLM, RTE inversely correlated with mean Ki67 (r = -0.50, p = 0.01) in the hepatobiliary phase. Negative correlations between RTE and CD45 expression were found in PCLM and CRCLM in the portal venous phase (r = -0.69, p = 0.01 and r = -0.41, p = 0.04) and the venous phase (r = -0.65, p = 0.01 and r = -0.44, p = 0.02).

Conclusion: Significant variations in RTE were identified among different types of LM, with correlations between RTE values and IHC markers such as CD45 and Ki67 suggesting that RTE may serve as a non-invasive biomarker for predicting IHC features in LM.

Critical relevance statement: RTE values serve as a predictive biomarker for IHC features in liver metastasis, potentially enhancing non-invasive patient assessment, disease monitoring, and treatment planning.

Key points: Few studies link gadoxetic acid-enhanced MRI with immunohistochemistry in LM. RTE varies by liver metastasis type and correlates with CD45 and Ki67. RTE reflects IHC features in LM, aiding non-invasive assessment.

Objectives: To explore the diagnostic efficacy of advanced intelligent clear-IQ engine (AiCE) and adaptive iterative dose reduction 3D (AIDR 3D), combination with and without the black blood CT technique (BBCT), for detecting vascular invasion in patients diagnosed with nonmetastatic pancreatic ductal adenocarcinoma (PDAC).

Methods: A total of 35 consecutive patients diagnosed with PDAC, proceeding with contrast-enhanced abdominal CT scans, were enrolled in this study. The arterial and portal venous phase images were reconstructed using AiCE and AIDR 3D. The corresponding BBCT images were established as AiCE-BBCT and AIDR 3D-BBCT, respectively. Two observers scored the image quality independently. Cohen's kappa (k) value or intraclass correlation coefficient (ICC) was used to analyze consistency. The diagnostic performance of four algorithms in detecting vascular invasion in PDAC patients was assessed using the area under the curve (AUC).

Results: The AiCE and AiCE-BBCT groups demonstrated superior image noise and diagnostic acceptability compared with AIDR 3D and AIDR 3D-BBCT groups (all p < 0.001), and the k value was 0.861-0.967 for both reviewers. In terms of diagnostic capability for vascular invasion in PDAC, the AiCE-BBCT group exhibited higher specificity (95.0%) and sensitivity (93.3%) compared to the AIDR 3D and AIDR 3D-BBCT groups, with an AUC of 0.942 (95% CI: 0.849-1.000, p < 0.05). Furthermore, all vascular evaluations conducted using AiCE-BBCT demonstrated better consistency (ICC: 0.847-0.935).

Conclusion: The BBCT technique in conjunction with AiCE could lead to notable enhancements in both the image quality of PDAC images and the diagnostic performance for tumor vascular invasion.

Critical relevance statement: Better diagnostic accuracy of vascular invasion of PDAC based on BBCT in combination with an AiCE is a critical factor in determining treatment strategies and patient outcomes.

Key points: Identifying vascular invasion of PDAC is important for prognostication. Combined images provide improved image quality and higher diagnostic accuracy. Combined images can excellently display the vascular wall and invasion.

Objective: To develop and validate a model integrating dual-layer detector spectral computed tomography (DLCT) three-dimensional (3D) volume of interest (VOI)-based quantitative parameters and clinical features for predicting Ki-67 proliferation index (PI) in pancreatic ductal adenocarcinoma (PDAC).

Materials and methods: A total of 162 patients with histopathologically confirmed PDAC who underwent DLCT examination were included and allocated to the training (114) and validation (48) sets. 3D VOI-iodine concentration (IC), 3D VOI-slope of the spectral attenuation curves, and 3D VOI-effective atomic number were obtained from the portal venous phase. The significant clinical features and DLCT quantitative parameters were identified through univariate analysis and multivariate logistic regression. The discrimination capability and clinical applicability of the clinical, DLCT, and DLCT-clinical models were quantified by the Receiver Operating Characteristic curve (ROC) and Decision Curve Analysis (DCA), respectively. The optimal model was then used to develop a nomogram, with the goodness-of-fit evaluated through the calibration curve.

Results: The DLCT-clinical model demonstrated superior predictive capability and a satisfactory net benefit for Ki-67 PI in PDAC compared to the clinical and DLCT models. The DLCT-clinical model integrating 3D VOI-IC and CA125 showed area under the ROC curves of 0.939 (95% CI, 0.895-0.982) and 0.915 (95% CI, 0.834-0.996) in the training and validation sets, respectively. The nomogram derived from the DLCT-clinical model exhibited favorable calibration, as depicted by the calibration curve.

Conclusions: The proposed model based on DLCT 3D VOI-IC and CA125 is a non-invasive and effective preoperative prediction tool demonstrating favorable predictive performance for Ki-67 PI in PDAC.

Critical relevance statement: The dual-layer detector spectral computed tomography-clinical model could help predict high Ki-67 PI in pancreatic ductal adenocarcinoma patients, which may help clinicians provide appropriate and individualized treatments.

Key points: Dual-layer detector spectral CT (DLCT) could predict Ki-67 in pancreatic ductal adenocarcinoma (PDAC). The DLCT-clinical model improved the differential diagnosis of Ki-67. The nomogram showed satisfactory calibration and net benefit for discriminating Ki-67.

Objective: To develop a multi-scene model that can automatically segment acute vertebral compression fractures (VCFs) from spine radiographs.

Methods: In this multicenter study, we collected radiographs from five hospitals (Hospitals A-E) between November 2016 and October 2019. The study included participants with acute VCFs, as well as healthy controls. For the development of the Positioning and Focus Network (PFNet), we used a training dataset consisting of 1071 participants from Hospitals A and B. The validation dataset included 458 participants from Hospitals A and B, whereas external test datasets 1-3 included 301 participants from Hospital C, 223 from Hospital D, and 261 from Hospital E, respectively. We evaluated the segmentation performance of the PFNet model and compared it with previously described approaches. Additionally, we used qualitative comparison and gradient-weighted class activation mapping (Grad-CAM) to explain the feature learning and segmentation results of the PFNet model.

Results: The PFNet model achieved accuracies of 99.93%, 98.53%, 99.21%, and 100% for the segmentation of acute VCFs in the validation dataset and external test datasets 1-3, respectively. The receiver operating characteristic curves comparing the four models across the validation and external test datasets consistently showed that the PFNet model outperformed other approaches, achieving the highest values for all measures. The qualitative comparison and Grad-CAM provided an intuitive view of the interpretability and effectiveness of our PFNet model.

Conclusion: In this study, we successfully developed a multi-scene model based on spine radiographs for precise preoperative and intraoperative segmentation of acute VCFs.

Critical relevance statement: Our PFNet model demonstrated high accuracy in multi-scene segmentation in clinical settings, making it a significant advancement in this field.

Key points: This study developed the first multi-scene deep learning model capable of segmenting acute VCFs from spine radiographs. The model's architecture consists of two crucial modules: an attention-guided module and a supervised decoding module. The exceptional generalization and consistently superior performance of our model were validated using multicenter external test datasets.

Background: The contrast-enhanced US (CEUS) Bosniak classification, proposed by the European Federation for Ultrasound in Medicine and Biology (EFSUMB) in 2020, predicts malignancy in cystic renal masses (CRMs). However, intra- and inter-rater reproducibility for CEUS features has not been well investigated.

Purpose: To explore intra- and inter-rater agreement for US features, identify confounding features, and assess the diagnostic performance of CEUS Bosniak classification.

Materials and methods: This retrospective study included patients with complex CRMs who underwent CEUS examination from January 2013 to August 2023. Radiologists (3 experts and 3 novices) evaluated calcification, echogenic content, wall, septa, and internal nodules of CRMs using CEUS Bosniak classification. Intra- and inter-rater agreements were assessed using the Gwet agreement coefficient (Gwet's AC). Linear regression identified features associated with discrepancies in Bosniak category assignment. Diagnostic performance was evaluated using the area under the receiver operating characteristic curve (AUC).

Results: A total of 103 complex CRMs were analyzed in 103 patients (mean age, 50 ± 15 years; 66 males). Intra-rater agreement for the Bosniak category was substantial to almost perfect (Gwet's AC 0.73-0.87). Inter-rater agreement was substantial for the Bosniak category (Gwet's AC 0.75) and moderate to almost perfect for US features (Gwet's AC 0.44-0.94). Nodule variation (i.e., absence vs. obtuse margin vs. acute margin) explained 84% of the variability in the Bosniak category assignment. CEUS Bosniak classification showed good diagnostic performance, with AUCs ranging from 0.78 to 0.90 for each rater.

Conclusions: CEUS Bosniak classification demonstrated substantial intra- and inter-rater reproducibility and good diagnostic performance in predicting the malignancy potential of CRMs. Nodule variations significantly predicted differences in Bosniak category assignments.

Critical relevance statement: Contrast-enhanced US Bosniak classification reliably predicts malignancy in cystic renal masses, demonstrating substantial reproducibility and diagnostic accuracy. This improves clinical decision-making and patient management.

Key points: Intra- and inter-rater reproducibility for contrast-enhance US features for Bosniak classification have not been well investigated. Substantial inter-rater agreements for the Bosniak category and variable agreements for determining imaging features were found. Contrast-enhanced US Bosniak classification is reproducible and has good diagnostic performance for predicting malignancy in cystic renal masses.

Various processes, including benign or malignant (mostly metastasis) processes, contribute to the occurrence of multiple pulmonary nodules. For differential diagnosis, metastasis must be excluded as an etiological factor in patients who have multiple pulmonary nodules with a known primary malignancy. However, differential diagnosis of multiple pulmonary nodules caused by benign diseases and malignant processes is challenging. Multiple pulmonary nodules resulting from metastasis may mimic those resulting from infections, inflammatory processes, and rare benign diseases. Some rare diseases, such as pulmonary sclerosing pneumocytoma and pulmonary epithelioid hemangioendothelioma, or common diseases with a rare presentation of multiple nodules must be considered in the differential diagnosis of metastasis. In addition to the clinical and laboratory findings, radiological features are crucial for differential diagnosis. The size, density, location, and border characteristics (well-defined or poorly defined) of pulmonary nodules, as well as their internal structure (solid, subsolid, or ground glass nodule), growth rate during follow-up, and associated pulmonary and extrapulmonary findings are important for differential diagnosis along with clinical and laboratory data. This article summarizes the general features and imaging findings of these diseases, which less frequently present with multiple pulmonary nodules, and the clues that can be used to distinguish these diseases from metastasis. CRITICAL RELEVANCE STATEMENT: The radiological features, clinical findings, and temporal changes during follow-up are important in distinguishing non-metastatic causes of multiple pulmonary nodules from metastatic causes and guiding diagnosis and early treatment, especially in patients with primary malignancy. KEY POINTS: Multiple pulmonary nodules have a wide range of etiologies, including metastatic disease. Metastasis as an etiology must be excluded in patients with multiple pulmonary nodules. Correlation of radiological findings (nodule size, position, and associated findings) with clinical history is crucial for differential diagnosis.