To evaluate the feasibility of generative pre-trained transformer-4 (GPT-4) in generating structured reports (SRs) from mixed-language (English and Korean) narrative-style CT reports for pancreatic ductal adenocarcinoma (PDAC) and to assess its accuracy in categorizing PDCA resectability. This retrospective study included consecutive free-text reports of pancreas-protocol CT for staging PDAC, from two institutions, written in English or Korean from January 2021 to December 2023. Both the GPT-4 Turbo and GPT-4o models were provided prompts along with the free-text reports via an application programming interface and tasked with generating SRs and categorizing tumor resectability according to the National Comprehensive Cancer Network guidelines version 2.2024. Prompts were optimized using the GPT-4 Turbo model and 50 reports from Institution B. The performances of the GPT-4 Turbo and GPT-4o models in the two tasks were evaluated using 115 reports from Institution A. Results were compared with a reference standard that was manually derived by an abdominal radiologist. Each report was consecutively processed three times, with the most frequent response selected as the final output. Error analysis was guided by the decision rationale provided by the models. Of the 115 narrative reports tested, 96 (83.5%) contained both English and Korean. For SR generation, GPT-4 Turbo and GPT-4o demonstrated comparable accuracies (92.3% [1592/1725] and 92.2% [1590/1725], respectively; <i>P</i> = 0.923). In the resectability categorization, GPT-4 Turbo showed higher accuracy than GPT-4o (81.7% [94/115] vs. 67.0% [77/115], respectively; <i>P</i> = 0.002). In the error analysis of GPT-4 Turbo, the SR generation error rate was 7.7% (133/1725 items), which was primarily attributed to inaccurate data extraction (54.1% [72/133]). The resectability categorization error rate was 18.3% (21/115), with the main cause being violation of the resectability criteria (61.9% [13/21]). Both GPT-4 Turbo and GPT-4o demonstrated acceptable accuracy in generating NCCN-based SRs on PDACs from mixed-language narrative reports. However, oversight by human radiologists is essential for determining resectability based on CT findings.

To investigate the behavior of artificial intelligence (AI) CT-based body composition biomarkers at different virtual monoenergetic imaging (VMI) levels using dual-energy CT (DECT). This retrospective study included 88 contrast-enhanced abdominopelvic CTs acquired with rapid-kVp switching DECT. Images were reconstructed into five VMI levels (40, 55, 70, 85, 100 keV). Fully automated algorithms for quantifying CT number (HU) in abdominal fat (subcutaneous and visceral), skeletal muscle, bone, calcium (abdominal Agatston score), and organ size (area or volume) were applied. Biomarker median difference relative to 70 keV and interquartile range were reported by energy level to characterize variation. Linear regression was performed to calibrate non-70 keV data and to estimate their equivalent 70 keV biomarker attenuation values. Relative to 70 keV, absolute median differences in attenuation-based biomarkers (excluding Agatston score) ranged 39-358, 12-102, 5-48, 9-75 HU for 40, 55, 85, 100 keV, respectively. For area-based biomarkers, differences ranged 6-15, 3-4, 2-7, 0-5 cm² for 40, 55, 85, 100 keV. For volume-based biomarkers, differences ranged 12-34, 8-68, 12-52, 1-57 cm³ for 40, 55, 85, 100 keV. Agatston score behavior was more spurious with median differences ranging 70-204 HU. In general, VMI < 70 keV showed more variation in median biomarker measurement than VMI > 70 keV. This study characterized the behavior of a fully automated AI CT biomarker toolkit across varying VMI levels obtained with DECT. The data showed relatively little biomarker value change when measured at or greater than 70 keV. Lower VMI datasets should be avoided due to larger deviations in measured value as compared to 70 keV, a level considered equivalent to conventional 120 kVp exams.

Coronary artery disease (CAD) is a major worldwide health concern, contributing significantly to the global burden of cardiovascular diseases (CVDs). According to the 2023 World Health Organization (WHO) report, CVDs account for approximately 17.9 million deaths annually. This emphasizies the need for advanced diagnostic tools such as coronary computed tomography angiography (CCTA). The incorporation of deep learning (DL) technologies could significantly improve CCTA analysis by automating the quantification of plaque and stenosis, thus enhancing the precision of cardiac risk assessments. A recent meta-analysis highlights the evolving role of CCTA in patient management, showing that CCTA-guided diagnosis and management reduced adverse cardiac events and improved event-free survival in patients with stable and acute coronary syndromes. An extensive literature search was carried out across various electronic databases, such as MEDLINE, Embase, and the Cochrane Library. This search utilized a specific strategy that included both Medical Subject Headings (MeSH) terms and pertinent keywords. The review adhered to PRISMA guidelines and focused on studies published between 2019 and 2024 that employed deep learning (DL) for coronary computed tomography angiography (CCTA) in patients aged 18 years or older. After implementing specific inclusion and exclusion criteria, a total of 10 articles were selected for systematic evaluation regarding quality and bias. This systematic review included a total of 10 studies, demonstrating the high diagnostic performance and predictive capabilities of various deep learning models compared to different imaging modalities. This analysis highlights the effectiveness of these models in enhancing diagnostic accuracy in imaging techniques. Notably, strong correlations were observed between DL-derived measurements and intravascular ultrasound findings, enhancing clinical decision-making and risk stratification for CAD. Deep learning-enabled CCTA represents a promising advancement in the quantification of coronary plaques and stenosis, facilitating improved cardiac risk prediction and enhancing clinical workflow efficiency. Despite variability in study designs and potential biases, the findings support the integration of DL technologies into routine clinical practice for better patient outcomes in CAD management.

To assess the accuracy of three commercially available and one open-source deep learning (DL) solutions for automatic tooth segmentation in cone beam computed tomography (CBCT) images of patients with multiple dental impactions. Twenty patients (20 CBCT scans) were selected from a retrospective cohort of individuals with multiple dental impactions. For each CBCT scan, one reference segmentation and four DL segmentations of the maxillary and mandibular teeth were obtained. Reference segmentations were generated by experts using a semi-automatic process. DL segmentations were automatically generated according to the manufacturer's instructions. Quantitative and qualitative evaluations of each DL segmentation were performed by comparing it with expert-generated segmentation. The quantitative metrics used were Dice similarity coefficient (DSC) and the normalized surface distance (NSD). The patients had an average of 12 retained teeth, with 12 of them diagnosed with a rare disease. DSC values ranged from 88.5% ± 3.2% to 95.6% ± 1.2%, and NSD values ranged from 95.3% ± 2.7% to 97.4% ± 6.5%. The number of completely unsegmented teeth ranged from 1 (0.1%) to 41 (6.0%). Two solutions (Diagnocat and DentalSegmentator) outperformed the others across all tested parameters. All the tested methods showed a mean NSD of approximately 95%, proving their overall efficiency for tooth segmentation. The accuracy of the methods varied among the four tested solutions owing to the presence of impacted teeth in our CBCT scans. DL solutions are evolving rapidly, and their future performance cannot be predicted based on our results.

Cardiovascular disease (CVD) is the leading cause of death worldwide. Coronary artery disease (CAD), a prevalent form of CVD, is typically assessed using catheter coronary angiography (CCA), an invasive, costly procedure with associated risks. While cardiac computed tomography angiography (CTA) presents a less invasive alternative, it suffers from limited temporal resolution, often resulting in motion artifacts that degrade diagnostic quality. Traditional ECG-based gating methods for CTA inadequately capture cardiac mechanical motion. To address this, we propose a novel multimodal approach that enhances CTA imaging by predicting cardiac quiescent periods using seismocardiogram (SCG) and ECG data, integrated through a weighted fusion (WF) approach and artificial neural networks (ANNs). We developed a regression-based ANN framework (r-ANN WF) designed to improve prediction accuracy and reduce computational complexity, which was compared with a classification-based framework (c-ANN WF), ECG gating, and US data. Our results demonstrate that the r-ANN WF approach improved overall diastolic and systolic cardiac quiescence prediction accuracy by 52.6% compared to ECG-based predictions, using ultrasound (US) as the ground truth, with an average prediction time of 4.83 ms. Comparative evaluations based on reconstructed CTA images show that both r-ANN WF and c-ANN WF offer diagnostic quality comparable to US-based gating, underscoring their clinical potential. Additionally, the lower computational complexity of r-ANN WF makes it suitable for real-time applications. This approach could enhance CTA's diagnostic quality, offering a more accurate and efficient method for CVD diagnosis and management.

Various hepatocellular carcinoma (HCC) prediction models have been proposed for patients with chronic hepatitis B (CHB) using clinical variables. We aimed to develop an artificial intelligence (AI)-based HCC prediction model by incorporating imaging biomarkers derived from abdominal computed tomography (CT) images along with clinical variables. An AI prediction model employing a gradient-boosting machine algorithm was developed utilizing imaging biomarkers extracted by DeepFore, a deep learning-based CT auto-segmentation software. The derivation cohort (n = 5,585) was randomly divided into the training and internal validation sets at a 3:1 ratio. The external validation cohort included 2,883 patients. Six imaging biomarkers (i.e. abdominal visceral fat-total fat volume ratio, total fat-trunk volume ratio, spleen volume, liver volume, liver-spleen Hounsfield unit ratio, and muscle Hounsfield unit) and eight clinical variables were selected as the main variables of our model, PLAN-B-DF. In the internal validation set (median follow-up duration = 7.4 years), PLAN-B-DF demonstrated an excellent predictive performance with a c-index of 0.91 and good calibration function (p = 0.78 by the Hosmer-Lemeshow test). In the external validation cohort (median follow-up duration = 4.6 years), PLAN-B-DF showed a significantly better discrimination function compared to previous models, including PLAN-B, PAGE-B, modified PAGE-B, and CU-HCC (c-index, 0.89 vs. 0.65-0.78; all p <0.001), and maintained a good calibration function (p = 0.42 by the Hosmer-Lemeshow test). When patients were classified into four groups according to the risk probability calculated by PLAN-B-DF, the 10-year cumulative HCC incidence was 0.0%, 0.4%, 16.0%, and 46.2% in the minimal-, low-, intermediate-, and high-risk groups, respectively. This AI prediction model, integrating deep learning-based auto-segmentation of CT images, offers improved performance in predicting HCC risk among patients with CHB compared to previous models. The novel predictive model PLAN-B-DF, employing an automated computed tomography segmentation algorithm, significantly improves predictive accuracy and risk stratification for hepatocellular carcinoma in patients with chronic hepatitis B (CHB). Using a gradient-boosting algorithm and computed tomography metrics, such as visceral fat volume and myosteatosis, PLAN-B-DF outperforms previous models based solely on clinical and demographic data. This model not only shows a higher c-index compared to previous models, but also effectively classifies patients with CHB into different risk groups. This model uses machine learning to analyze the complex relationships among various risk factors contributing to hepatocellular carcinoma occurrence, thereby enabling more personalized surveillance for patients with CHB.

Urolithiasis, a common and painful urological condition, is influenced by factors such as lifestyle, genetics, and medication. Differentiating between different types of kidney stones is crucial for personalized therapy. The purpose of this study is to investigate the use of photon-counting computed tomography (PCCT) in combination with radiomics and machine learning to develop a method for automated and detailed characterization of kidney stones. This approach aims to enhance the accuracy and detail of stone classification beyond what is achievable with conventional computed tomography (CT) and dual-energy CT (DECT). In this ex vivo study, 135 kidney stones were first classified using infrared spectroscopy. All stones were then scanned in a PCCT embedded in a phantom. Various monoenergetic reconstructions were generated, and radiomics features were extracted. Statistical analysis was performed using Random Forest (RF) classifiers for both individual reconstructions and a combined model. The combined model, using radiomics features from all monoenergetic reconstructions, significantly outperformed individual reconstructions and SPP parameters, with an AUC of 0.95 and test accuracy of 0.81 for differentiating all six stone types. Feature importance analysis identified key parameters, including NGTDM_Strength and wavelet-LLH_firstorder_Variance. This ex vivo study demonstrates that radiomics-driven PCCT analysis can improve differentiation between kidney stone subtypes. The combined model outperformed individual monoenergetic levels, highlighting the potential of spectral profiling in PCCT to optimize treatment through image-based strategies. Question How can photon-counting computed tomography (PCCT) combined with radiomics improve the differentiation of kidney stone types beyond conventional CT and dual-energy CT, enhancing personalized therapy? Findings Our ex vivo study demonstrates that a combined spectral-driven radiomics model achieved 95% AUC and 81% test accuracy in differentiating six kidney stone types. Clinical relevance Implementing PCCT-based spectral-driven radiomics allows for precise non-invasive differentiation of kidney stone types, leading to improved diagnostic accuracy and more personalized, effective treatment strategies, potentially reducing the need for invasive procedures and recurrence.

Ischemic changes are not visible on non-contrast head CT until several hours after infarction, though deep convolutional neural networks have shown promise in the detection of subtle imaging findings. This study aims to assess if dual-energy CT (DECT) acquisition can improve early infarct visibility for machine learning. The retrospective dataset consisted of 330 DECTs acquired up to 48 h prior to confirmation of a DWI positive infarct on MRI between 2016 and 2022. Infarct segmentation maps were generated from the MRI and co-registered to the CT to serve as ground truth for segmentation. A self-configuring 3D nnU-Net was trained for segmentation on (1) standard 120 kV mixed-images (2) 190 keV virtual monochromatic images and (3) 120 kV + 190 keV images as dual channel inputs. Algorithm performance was assessed with Dice scores with paired t-tests on a test set. Global aggregate Dice scores were 0.616, 0.645, and 0.665 for standard 120 kV images, 190 keV, and combined channel inputs respectively. Differences in overall Dice scores were statistically significant with highest performance for combined channel inputs (p < 0.01). Small but statistically significant differences were observed for infarcts between 6 and 12 h from last-known-well with higher performance for larger infarcts. Volumetric accuracy trended higher with combined inputs but differences were not statistically significant (p = 0.07). Supplementation of standard head CT images with dual-energy data provides earlier and more accurate segmentation of infarcts for machine learning particularly between 6 and 12 h after last-known-well.

In computer-aided diagnosis systems, precise feature extraction from CT scans of colorectal cancer using deep learning is essential for effective prognosis. However, existing convolutional neural networks struggle to capture long-range dependencies and contextual information, resulting in incomplete CT feature extraction. To address this, the PEDRA-EFB0 architecture integrates patch embeddings and a dual residual attention mechanism for enhanced feature extraction and survival prediction in colorectal cancer CT scans. A patch embedding method processes CT scans into patches, creating positional features for global representation and guiding spatial attention computation. Additionally, a dual residual attention mechanism during the upsampling stage selectively combines local and global features, enhancing CT data utilization. Furthermore, this paper proposes a feature selection algorithm that combines autoencoders and entropy technology, encoding and compressing high-dimensional data to reduce redundant information and using entropy to assess the importance of features, thereby achieving precise feature selection. Experimental results indicate the PEDRA-EFB0 model outperforms traditional methods on colorectal cancer CT metrics, notably in C-index, BS, MCC, and AUC, enhancing survival prediction accuracy. Our code is freely available at https://github.com/smile0208z/PEDRA .

Classical biomedical data science models are trained on a single modality and aimed at one specific task. However, the exponential increase in the size and capabilities of the foundation models inside and outside medicine shows a shift toward task-agnostic models using large-scale, often internet-based, data. Recent research into smaller foundation models trained on specific literature, such as programming textbooks, demonstrated that they can display capabilities similar to or superior to large generalist models, suggesting a potential middle ground between small task-specific and large foundation models. This study attempts to introduce a domain-specific multimodal model, Congress of Neurological Surgeons (CNS)-Contrastive Language-Image Pretraining (CLIP), developed for neurosurgical applications, leveraging data exclusively from Neurosurgery Publications. We constructed a multimodal data set of articles from Neurosurgery Publications through PDF data collection and figure-caption extraction using an artificial intelligence pipeline for quality control. Our final data set included 24 021 figure-caption pairs. We then developed a fine-tuning protocol for the OpenAI CLIP model. The model was evaluated on tasks including neurosurgical information retrieval, computed tomography imaging classification, and zero-shot ImageNet classification. CNS-CLIP demonstrated superior performance in neurosurgical information retrieval with a Top-1 accuracy of 24.56%, compared with 8.61% for the baseline. The average area under receiver operating characteristic across 6 neuroradiology tasks achieved by CNS-CLIP was 0.95, slightly superior to OpenAI's Contrastive Language-Image Pretraining at 0.94 and significantly outperforming a vanilla vision transformer at 0.62. In generalist classification, CNS-CLIP reached a Top-1 accuracy of 47.55%, a decrease from the baseline of 52.37%, demonstrating a catastrophic forgetting phenomenon. This study presents a pioneering effort in building a domain-specific multimodal model using data from a medical society publication. The results indicate that domain-specific models, while less globally versatile, can offer advantages in specialized contexts. This emphasizes the importance of using tailored data and domain-focused development in training foundation models in neurosurgery and general medicine.