We investigated pharmacokinetics, dosimetric patterns, and absorbed dose (AD)-effect correlations in [¹⁷⁷Lu]Lu-DOTATATE peptide receptor radionuclide therapy (PRRT) for metastatic neuroendocrine tumors (NETs) to develop strategies for future personalized dosimetry-guided treatments. <b>Methods:</b> Patients treated with standard [¹⁷⁷Lu]Lu-DOTATATE PRRT were recruited for serial SPECT/CT imaging. Kidneys were segmented on CT using a deep learning algorithm, and tumors were segmented at each cycle using a SPECT gradient-based tool, guided by radiologist-defined contours on baseline CT/MRI. Dosimetry was performed using an automated workflow that included contour intensity-based SPECT-SPECT registration, generation of Monte Carlo dose-rate maps, and dose-rate fitting. Lesion-level response at first follow-up was evaluated using both radiologic (RECIST and modified RECIST) and [⁶⁸Ga]Ga-DOTATATE PET-based criteria. Kidney toxicity was evaluated based on the estimated glomerular filtration rate (eGFR) at 9 mo after PRRT. <b>Results:</b> Dosimetry was performed after cycle 1 in 30 patients and after all cycles in 22 of 30 patients who completed SPECT/CT imaging after each cycle. Median cumulative tumor (<i>n</i> = 78) AD was 2.2 Gy/GBq (range, 0.1-20.8 Gy/GBq), whereas median kidney AD was 0.44 Gy/GBq (range, 0.25-0.96 Gy/GBq). The tumor-to-kidney AD ratio decreased with each cycle (median, 6.4, 5.7, 4.7, and 3.9 for cycles 1-4) because of a decrease in tumor AD, while kidney AD remained relatively constant. Higher-grade (grade 2) and pancreatic NETs showed a significantly larger drop in AD with each cycle, as well as significantly lower AD and effective half-life (T_eff), than did low-grade (grade 1) and small intestinal NETs, respectively. T_eff remained relatively constant with each cycle for both tumors and kidneys. Kidney T_eff and AD were significantly higher in patients with low eGFR than in those with high eGFR. Tumor AD was not significantly associated with response measures. There was no nephrotoxicity higher than grade 2; however, a significant negative association was found in univariate analyses between eGFR at 9 mo and AD to the kidney, which improved in a multivariable model that also adjusted for baseline eGFR (cycle 1 AD, <i>P</i> = 0.020, adjusted <i>R</i> ² = 0.57; cumulative AD, <i>P</i> = 0.049, adjusted <i>R</i> ² = 0.65). The association between percentage change in eGFR and AD to the kidney was also significant in univariate analysis and after adjusting for baseline eGFR (cycle 1 AD, <i>P</i> = 0.006, adjusted <i>R</i> ² = 0.21; cumulative AD, <i>P</i> = 0.019, adjusted <i>R</i> ² = 0.21). <b>Conclusion:</b> The dosimetric behavior we report over different cycles and for different NET subgroups can be considered when optimizing PRRT to individual patients. The models we present for the relationship between eGFR and AD have potential for clinical use in predicting renal function early in the treatment course. Furthermore, reported pharmacokinetics for patient subgroups allow more appropriate selection of population parameters to be used in protocols with fewer imaging time points that facilitate more widespread adoption of dosimetry.

BackgroundDifferent Natural Language Processing (NLP) techniques have demonstrated promising results for data extraction from radiological reports. Both traditional rule-based methods like regular expressions (Regex) and modern Large Language Models (LLMs) can extract structured information. However, comparison between these approaches for extraction of specific radiological data elements has not been widely conducted. MethodsWe compared accuracy and processing time between Regex and LLM-based approaches for extracting BI-RADS scores from 7,764 radiology reports (mammography, ultrasound, MRI, and biopsy). We developed a rule-based algorithm using Regex patterns and implemented an LLM-based extraction using the Rombos-LLM-V2.6-Qwen-14b model. A ground truth dataset of 199 manually classified reports was used for evaluation. ResultsThere was no statistically significant difference in the accuracy in extracting BI-RADS scores between Regex and an LLM-based method (accuracy of 89.20% for Regex versus 87.69% for the LLM-based method; p=0.56). Compared to the LLM-based method, Regex processing was more efficient, completing the task 28,120 times faster (0.06 seconds vs. 1687.20 seconds). Further analysis revealed LLMs favored common classifications (particularly BI-RADS value of 2) while Regex more frequently returned "unclear" values. We also could confirm in our sample an already known laterality bias for breast cancer (BI-RADS 6) and detected a slight laterality skew for suspected breast cancer (BI-RADS 5) as well. ConclusionFor structured, standardized data like BI-RADS, traditional NLP techniques seem to be superior, though future work should explore hybrid approaches combining Regex precision for standardized elements with LLM contextual understanding for more complex information extraction tasks.

This study aimed to develop an automated early warning system using a large language model (LLM) to identify acute to subacute brain infarction from free-text computed tomography (CT) or magnetic resonance imaging (MRI) radiology reports. In this retrospective study, 5,573, 1,883, and 834 patients were included in the training (mean age, 67.5 ± 17.2 years; 2,831 males), validation (mean age, 61.5 ± 18.3 years; 994 males), and test (mean age, 66.5 ± 16.1 years; 488 males) datasets. An LLM (Japanese Bidirectional Encoder Representations from Transformers model) was fine-tuned to classify the CT and MRI reports into three groups (group 0, newly identified acute to subacute infarction; group 1, known acute to subacute infarction or old infarction; group 2, without infarction). The training and validation processes were repeated 15 times, and the best-performing model on the validation dataset was selected to further evaluate its performance on the test dataset. The best fine-tuned model exhibited sensitivities of 0.891, 0.905, and 0.959 for groups 0, 1, and 2, respectively, in the test dataset. The macrosensitivity (the average of sensitivity for all groups) and accuracy were 0.918 and 0.923, respectively. The model's performance in extracting newly identified acute brain infarcts was high, with an area under the receiver operating characteristic curve of 0.979 (95% confidence interval, 0.956-1.000). The average prediction time was 0.115 ± 0.037 s per patient. A fine-tuned LLM could extract newly identified acute to subacute brain infarcts based on CT or MRI findings with high performance.

The existing assessment of adjacent segment degeneration (ASD) risk after lumbar fusion surgery focuses on a single type of clinical information or imaging manifestations. In the early stages, it is difficult to show obvious degeneration characteristics, and the patients' true risks cannot be fully revealed. The evaluation results based on imaging ignore the clinical symptoms and changes in quality of life of patients, limiting the understanding of the natural process of ASD and the comprehensive assessment of its risk factors, and hindering the development of effective prevention strategies. To improve the quality of postoperative management and effectively identify the characteristics of ASD, this paper studies the risk assessment of ASD after lumbar fusion surgery by combining the artificial intelligence (AI) medical image-aided diagnosis system. First, the collaborative attention mechanism is adopted to start with the extraction of single-modal features and fuse the multi-modal features of computed tomography (CT) and magnetic resonance imaging (MRI) images. Then, the similarity matrix is weighted to achieve the complementarity of multi-modal information, and the stability of feature extraction is improved through the residual network structure. Finally, the fully connected network (FCN) is combined with the multi-task learning framework to provide a more comprehensive assessment of the risk of ASD. The experimental analysis results show that compared with three advanced models, three dimensional-convolutional neural networks (3D-CNN), U-Net++, and deep residual networks (DRN), the accuracy of the model in this paper is 3.82 %, 6.17 %, and 6.68 % higher respectively; the precision is 0.56 %, 1.09 %, and 4.01 % higher respectively; the recall is 3.41 %, 4.85 %, and 5.79 % higher respectively. The conclusion shows that the AI medical image-aided diagnosis system can help to accurately identify the characteristics of ASD and effectively assess the risks after lumbar fusion surgery.

To identify predictors of a true scaphoid fracture among patients with radial wrist pain following acute trauma, train 5 machine learning (ML) algorithms in predicting scaphoid fracture probability, and design a decision rule to initiate advanced imaging in high-risk patients. Two prospective cohorts including 422 patients with radial wrist pain following wrist trauma were combined. There were 117 scaphoid fractures (28%) confirmed on computed tomography, magnetic resonance imaging, or radiographs. Eighteen fractures (15%) were occult. Predictors of a scaphoid fracture were identified among demographics, mechanism of injury and examination maneuvers. Five ML-algorithms were trained in calculating scaphoid fracture probability. ML-algorithms were assessed on ability to discriminate between patients with and without a fracture (area under the receiver operating characteristic curve), agreement between observed and predicted probabilities (calibration), and overall performance (Brier score). The best performing ML-algorithm was incorporated into a probability calculator. A decision rule was proposed to initiate advanced imaging among patients with negative radiographs. Pain over the scaphoid on ulnar deviation, sex, age, and mechanism of injury were most strongly associated with a true scaphoid fracture. The best performing ML-algorithm yielded an area under the receiver operating characteristic curve, calibration slope, intercept, and Brier score of 0.77, 0.84, -0.01 and 0.159, respectively. The ML-derived decision rule proposes to initiate advanced imaging in patients with radial-sided wrist pain, negative radiographs, and a fracture probability of ≥10%. When applied to our cohort, this would yield 100% sensitivity, 38% specificity, and would have reduced the number of patients undergoing advanced imaging by 36% without missing a fracture. The ML-algorithm accurately calculated scaphoid fracture probability based on scaphoid pain on ulnar deviation, sex, age, and mechanism of injury. The ML-decision rule may reduce the number of patients undergoing advanced imaging by a third with a small risk of missing a fracture. External validation is required before implementation. Diagnostic II.

The test-time adaptation (TTA) of deep-learning-based semantic segmentation models, specific to individual patient data, was addressed in this study. The existing TTA methods in medical imaging are often unconstrained, require anatomical prior information or additional neural networks built during training phase, making them less practical, and prone to performance deterioration. In this study, a novel framework based on information geometric principles was proposed to achieve generic, off-the-shelf, regularized patient-specific adaptation of models during test-time. By considering the pre-trained model and the adapted models as part of statistical neuromanifolds, test-time adaptation was treated as constrained functional regularization using information geometric measures, leading to improved generalization and patient optimality. The efficacy of the proposed approach was shown on three challenging problems: 1) improving generalization of state-of-the-art models for segmenting COVID-19 anomalies in Computed Tomography (CT) images 2) cross-institutional brain tumor segmentation from magnetic resonance (MR) images, 3) segmentation of retinal layers in Optical Coherence Tomography (OCT) images. Further, it was demonstrated that robust patient-specific adaptation can be achieved without adding significant computational burden, making it first of its kind based on information geometric principles.

Thyroid nodules classified within the Thyroid Imaging Reporting and Data Systems (TI-RADS) category 3-5 are typically regarded as having varying degrees of malignancy risk, with the risk increasing from TI-RADS 3 to TI-RADS 5. While some of these nodules may undergo fine-needle aspiration (FNA) biopsy to assess their nature, this procedure carries a risk of false negatives and inherent complications. To avoid the need for unnecessary biopsy examination, we explored a method for distinguishing the benign and malignant characteristics of thyroid TI-RADS 3-5 nodules based on deep-learning ultrasound images combined with computed tomography (CT). Thyroid nodules, assessed as American College of Radiology (ACR) TI-RADS category 3-5 through conventional ultrasound, all of which had postoperative pathology results, were examined using both conventional ultrasound and CT before operation. We investigated the effectiveness of deep-learning models based on ultrasound alone, CT alone, and a combination of both imaging modalities using the following metrics: Area Under Curve (AUC), sensitivity, accuracy, and positive predictive value (PPV). Additionally, we compared the diagnostic efficacy of the combined methods with manual readings of ultrasound and CT. A total of 768 thyroid nodules falling within TI-RADS categories 3-5 were identified across 768 patients. The dataset comprised 499 malignant and 269 benign cases. For the automatic identification of thyroid TI-RADS category 3-5 nodules, deep learning combined with ultrasound and CT demonstrated a significantly higher AUC (0.930; 95% CI: 0.892, 0.969) compared to the application of ultrasound alone AUC (0.901; 95% CI: 0.856, 0.947) or CT alone AUC (0.776; 95% CI: 0.713, 0.840). Additionally, the AUC of combined modalities surpassed that of radiologists'assessments using ultrasound alone AUCmean (0.725；95% CI:0.677, 0.773), CT alone AUCmean (0.617; 95% CI:0.564, 0.669). Deep learning method combined with ultrasound and CT imaging of thyroid can allow more accurate and precise classification of nodules within TI-RADS categories 3-5.

Fully-supervised deep neural networks have achieved remarkable progress in medical image segmentation, yet they heavily rely on extensive manually labeled data and exhibit inflexibility for unseen tasks. Few-shot segmentation (FSS) addresses these issues by predicting unseen classes from a few labeled support examples. However, most existing FSS models struggle to generalize to diverse target tasks distinct from training domains. Furthermore, designing promising network architectures for such tasks is expertise-intensive and laborious. In this paper, we introduce MOE-FewSeg, a novel automatic design method for FSS architectures. Specifically, we construct a U-shaped encoder-decoder search space that incorporates capabilities for information interaction and feature selection, thereby enabling architectures to leverage prior knowledge from publicly available datasets across diverse domains for improved prediction of various target tasks. Given the potential conflicts among disparate target tasks, we formulate the multi-task problem as a multi-objective optimization problem. We employ a multi-objective genetic algorithm to identify the Pareto-optimal architectures for these target tasks within this search space. Furthermore, to mitigate the impact of noisy labels due to dataset quality variations, we propose a noise-robust loss function named NRL, which encourages the model to de-emphasize larger loss values. Empirical results demonstrate that MOE-FewSeg outperforms manually designed architectures and other related approaches.

The zero-watermarking methods provide a means of lossless, which was adopted to protect medical image copyright requiring high integrity. However, most existing studies have only focused on robustness and there has been little discussion about the analysis and experiment on discriminability. Therefore, this paper proposes a trusted robust zero-watermarking scheme for medical images based on Deep convolution neural network (DCNN) and the hyperchaotic encryption system. Firstly, the medical image is converted into several feature map matrices by the specific convolution layer of DCNN. Then, a stable Gram matrix is obtained by calculating the colinear correlation between different channels in feature map matrices. Finally, the Gram matrixes of the medical image and the feature map matrixes of the watermark image are fused by the trained DCNN to generate the zero-watermark. Meanwhile, we propose two feature evaluation criteria for finding differentiated eigenvalues. The eigenvalue is used as the explicit key to encrypt the generated zero-watermark by Lorenz hyperchaotic encryption, which enhances security and discriminability. The experimental results show that the proposed scheme can resist common image attacks and geometric attacks, and is distinguishable in experiments, being applicable for the copyright protection of medical images.

<i>Radiology: Cardiothoracic Imaging</i> publishes research, technical developments, and reviews related to cardiac, vascular, and thoracic imaging. The current review article, led by the <i>Radiology: Cardiothoracic Imaging</i> trainee editorial board, highlights the most impactful articles published in the journal between November 2023 and October 2024. The review encompasses various aspects of cardiac, vascular, and thoracic imaging related to coronary artery disease, cardiac MRI, valvular imaging, congenital and inherited heart diseases, thoracic imaging, lung cancer, artificial intelligence, and health services research. Key highlights include the role of CT fractional flow reserve analysis to guide patient management, the role of MRI elastography in identifying age-related myocardial stiffness associated with increased risk of heart failure, review of MRI in patients with cardiovascular implantable electronic devices and fractured or abandoned leads, imaging of mitral annular disjunction, specificity of the Lung Imaging Reporting and Data System version 2022 for detecting malignant airway nodules, and a radiomics-based reinforcement learning model to analyze serial low-dose CT scans in lung cancer screening. Ongoing research and future directions include artificial intelligence tools for applications such as plaque quantification using coronary CT angiography and growing understanding of the interconnectedness of environmental sustainability and cardiovascular imaging. <b>Keywords:</b> CT, MRI, CT-Coronary Angiography, Cardiac, Pulmonary, Coronary Arteries, Heart, Lung, Mediastinum, Mitral Valve, Aortic Valve, Artificial Intelligence © RSNA, 2025.