Background: ErisNet, a novel AI model to reduce noise in Computed Tomography images. Methods: We trained ErisNet on 23 post-mortem whole-body CT scans. We assessed the objective performance with mean square error (MSE), peak signal-to-noise ratio (PSNR), structural similarity index (SSIM) measure, visual information fidelity (VIF), edge preservation index (EPI) and noise variance (NV). We assessed the qualitative performance by six radiologists. To support the visual assessment, we placed circular regions of interest (ROI) in the vitreous body, brain, liver and spleen parenchyma and paravertebral muscle. Results: ErisNet achieved MSE 64.07 ± 46.81, PSNR 31.32 ± 3.69 dB, SSIM 0.93 ± 0.06, VIF 0.49 ± 0.09, EPI 0.97 ± 0.01 and NV 64.69 ± 46.80. The ROI analysis showed a reduction in noise: the SD of the HU decreased by 8% in the vitreous body (from 17.6 to 16.2 HU), by 18% in the brain parenchyma (from 18.85 to 15.40 HU) and by 15–19% in the liver, spleen and paravertebral muscle. The six radiologists confirmed these results by assigning high scores (scale from one to five): overall quality 4.5 ± 0.6, noise suppression/detail preservation 4.7 ± 0.5 and diagnostic confidence 4.8 ± 0.4 (p < 0.01). Conclusions: ErisNet improves the quality of CT images and shows strong potential for processing low-dose scans.
Background/Objectives: Reviewing the entire history of imaging exams of a single patient’s records is an essential step in clinical practice, but it is time and resource consuming, with potential negative effects on workflow and on the quality of medical decisions. The main objective of this study was to evaluate the applicability of three open-source large language models (LLMs) for the automatic generation of concise summaries of patient’s imaging records. Secondary objectives were to assess correlations among the LLMs and to evaluate the length reduction provided by each model. Methods: Three state-of-the-art open-source large language models were selected: Llama 3.2 11B, Mistral 7B, and Falcon 7B. Each model was given a set of radiology reports. The summaries produced by the models were evaluated by two experienced radiologists and one experienced clinical physician using standardized metrics. Results: A variable number of radiological reports (n = 12–56) from four patients were selected and evaluated. The summaries generated by the three LLM showed a good level of accuracy compared with the information contained in the original reports, with positive ratings on both clinical relevance and ease of reference. According to the experts’ evaluations, the use of the summaries generated by LLMs could help to reduce the time spent on reviewing the previous imaging examinations performed, preserving the quality of clinical data. Conclusions: Our results suggest that LLMs are able to generate summaries of the imaging history of patients, and these summaries could improve radiology workflow making it easier to manage large volumes of reports.
Background/Objectives: Type II endoleak (T2EL) remains the most frequent complication after endovascular aortic aneurysm repair (EVAR), with uncertain clinical relevance and management. While most resolve spontaneously, persistent T2ELs can lead to sac enlargement and rupture risk. This study proposes a deep learning framework for preoperative prediction of T2EL occurrence and severity using volumetric computed tomography angiography (CTA) data.Methods: A retrospective analysis of 287 patients undergoing standard EVAR (2010–2024) was performed. Preoperative CTA scans were processed for volumetric normalization and fed into a 3D convolutional neural network (CNN) trained to classify patients into three categories: no T2EL, benign T2EL, or malignant T2EL. The model was trained on 224 cases, validated on 33, and tested on an independent cohort of 30 patients. Performance metrics included accuracy, precision, recall, F1-score, and area under the receiver operating characteristic curve (AUC).Results: The CNN achieved an overall accuracy of 76.7% (95% CI: 0.63–0.90), macro-averaged F1-score of 0.77, and AUC of 0.93. Class-specific AUCs were 0.93 for no T2EL, 0.91 for benign, and 0.96 for malignant cases, confirming high discriminative capacity across outcomes. Most misclassifications occurred between adjacent categories.Conclusion: This study introduces the first end-to-end 3D CNN capable of predicting both presence and severity of T2EL directly from preoperative CTA, without manual segmentation or handcrafted features. These findings suggest that preoperative imaging encodes latent structural information predictive of endoleak-driven sac reperfusion, potentially enabling personalized pre-emptive embolization strategies and tailored surveillance after EVAR.
Background/Objectives: Type II endoleak (T2EL) remains the most frequent complication after endovascular aortic aneurysm repair (EVAR), with uncertain clinical relevance and management. While most resolve spontaneously, persistent T2ELs can lead to sac enlargement and rupture risk. This study proposes a deep learning framework for preoperative prediction of T2EL occurrence and severity using volumetric computed tomography angiography (CTA) data. Methods: A retrospective analysis of 277 patients undergoing standard EVAR (2010–2023) was performed. Preoperative CTA scans were processed for volumetric normalization and fed into a 3D convolutional neural network (CNN) trained to classify patients into three categories: no T2EL, benign T2EL, or malignant T2EL. The model was trained on 175 cases, validated on 72, and tested on an independent cohort of 30 patients. Performance metrics included accuracy, precision, recall, F1-score, and area under the receiver operating characteristic curve (AUC). Results: The CNN achieved an overall accuracy of 76.7% (95% CI: 0.63–0.90), a macro-averaged F1-score of 0.77, and an AUC of 0.93. Class-specific AUCs were 0.93 for no T2EL, 0.91 for benign, and 0.96 for malignant cases, confirming high discriminative capacity across outcomes. Most misclassifications occurred between adjacent categories. Conclusions: This study introduces the first end-to-end 3D CNN capable of predicting both the presence and severity of T2EL directly from preoperative CTA, without manual segmentation or handcrafted features. These findings suggest that preoperative imaging encodes latent structural information predictive of endoleak-driven sac reperfusion, potentially enabling personalized pre-emptive embolization strategies and tailored surveillance after EVAR.