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.

The PRECISE framework, defined as Patient-Focused Radiology Reports with Enhanced Clarity and Informative Summaries for Effective Communication, leverages GPT-4 to create patient-friendly summaries of radiology reports at a sixth-grade reading level. The purpose of the study was to evaluate the effectiveness of the PRECISE framework in improving the readability, reliability, and understandability of radiology reports. We hypothesized that the PRECISE framework improves the readability and patient understanding of radiology reports compared to the original versions. The PRECISE framework was assessed using 500 chest X-ray reports. Readability was evaluated using the Flesch Reading Ease, Gunning Fog Index, and Automated Readability Index. Reliability was gauged by clinical volunteers, while understandability was assessed by non-medical volunteers. Statistical analyses including t-tests, regression analyses, and Mann-Whitney U tests were conducted to determine the significance of the differences in readability scores between the original and PRECISE-generated reports. Readability scores significantly improved, with the mean Flesch Reading Ease score increasing from 38.28 to 80.82 (p-value < 0.001), the Gunning Fog Index decreasing from 13.04 to 6.99 (p-value < 0.001), and the ARI score improving from 13.33 to 5.86 (p-value < 0.001). Clinical volunteer assessments found 95 % of the summaries reliable, and non-medical volunteers rated 97 % of the PRECISE-generated summaries as fully understandable. The application of the PRECISE approach demonstrates promise in enhancing patient understanding and communication without adding significant burden to radiologists. With improved reliability and patient-friendly summaries, this approach holds promise for fostering patient engagement and understanding in healthcare decision-making. The PRECISE framework represents a pivotal step towards more inclusive and patient-centric care delivery.

Neuromuscular diseases (NMDs) are rare disorders characterized by progressive muscle fibre loss, leading to replacement by fibrotic and fatty tissue, muscle weakness and disability. Early diagnosis is critical for therapeutic decisions, care planning and genetic counselling. Muscle magnetic resonance imaging (MRI) has emerged as a valuable diagnostic tool by identifying characteristic patterns of muscle involvement. However, the increasing complexity of these patterns complicates their interpretation, limiting their clinical utility. Additionally, multi-study data aggregation introduces heterogeneity challenges. This study presents a novel multi-study harmonization pipeline for muscle MRI and an AI-driven diagnostic tool to assist clinicians in identifying disease-specific muscle involvement patterns. We developed a preprocessing pipeline to standardize MRI fat content across datasets, minimizing source bias. An ensemble of XGBoost models was trained to classify patients based on intramuscular fat replacement, age at MRI and sex. The SHapley Additive exPlanations (SHAP) framework was adapted to analyse model predictions and identify disease-specific muscle involvement patterns. To address class imbalance, training and evaluation were conducted using class-balanced metrics. The model's performance was compared against four expert clinicians using 14 previously unseen MRI scans. Using our harmonization approach, we curated a dataset of 2961 MRI samples from genetically confirmed cases of 20 paediatric and adult NMDs. The model achieved a balanced accuracy of 64.8% ± 3.4%, with a weighted top-3 accuracy of 84.7% ± 1.8% and top-5 accuracy of 90.2% ± 2.4%. It also identified key features relevant for differential diagnosis, aiding clinical decision-making. Compared to four expert clinicians, the model obtained the highest top-3 accuracy (75.0% ± 4.8%). The diagnostic tool has been implemented as a free web platform, providing global access to the medical community. The application of AI in muscle MRI for NMD diagnosis remains underexplored due to data scarcity. This study introduces a framework for dataset harmonization, enabling advanced computational techniques. Our findings demonstrate the potential of AI-based approaches to enhance differential diagnosis by identifying disease-specific muscle involvement patterns. The developed tool surpasses expert performance in diagnostic ranking and is accessible to clinicians worldwide via the Myo-Guide online platform.

To develop a semi-supervised domain adaptation technique for metal artifact reduction with a spatial-frequency transformer (SFTrans) model (Semi-SFTrans), and to quantitatively compare its performance with supervised models (Sup-SFTrans and ResUNet) and traditional linear interpolation MAR method (LI) in oral and maxillofacial CT. Supervised models, including Sup-SFTrans and a state-of-the-art model termed ResUNet, were trained with paired simulated CT images, while semi-supervised model, Semi-SFTrans, was trained with both paired simulated and unpaired clinical CT images. For evaluation on the simulated data, we calculated Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index Measure (SSIM) on the images corrected by four methods: LI, ResUNet, Sup-SFTrans, and Semi-SFTrans. For evaluation on the clinical data, we collected twenty-two clinical cases with real metal artifacts, and the corresponding intraoral scan data. Three radiologists visually assessed the severity of artifacts using Likert scales on the original, Sup-SFTrans-corrected, and Semi-SFTrans-corrected images. Quantitative MAR evaluation was conducted by measuring Mean Hounsfield Unit (HU) values, standard deviations, and Signal-to-Noise Ratios (SNRs) across Regions of Interest (ROIs) such as the tongue, bilateral buccal, lips, and bilateral masseter muscles, using paired t-tests and Wilcoxon signed-rank tests. Further, teeth integrity in the corrected images was assessed by comparing teeth segmentation results from the corrected images against the ground-truth segmentation derived from registered intraoral scan data, using Dice Score and Hausdorff Distance. Sup-SFTrans outperformed LI, ResUNet and Semi-SFTrans on the simulated dataset. Visual assessments from the radiologists showed that average scores were (2.02 ± 0.91) for original CT, (4.46 ± 0.51) for Semi-SFTrans CT, and (3.64 ± 0.90) for Sup-SFTrans CT, with intra correlation coefficients (ICCs)＞0.8 of all groups and p < 0.001 between groups. On soft tissue, both Semi-SFTrans and Sup-SFTrans significantly reduced metal artifacts in tongue (p < 0.001), lips, bilateral buccal regions, and masseter muscle areas (p < 0.05). Semi-SFTrans achieved superior metal artifact reduction than Sup-SFTrans in all ROIs (p < 0.001). SNR results indicated significant differences between Semi-SFTrans and Sup-SFTrans in tongue (p = 0.0391), bilateral buccal (p = 0.0067), lips (p = 0.0208), and bilateral masseter muscle areas (p = 0.0031). Notably, Semi-SFTrans demonstrated better teeth integrity preservation than Sup-SFTrans (Dice Score: p < 0.001; Hausdorff Distance: p = 0.0022). The semi-supervised MAR model, Semi-SFTrans, demonstrated superior metal artifact reduction performance over supervised counterparts in real dental CT images.

To compare prostate cancer lesion detection using conventional and artificial intelligence (AI)-assisted image interpretation at multiparametric MRI (mpMRI). A retrospective study of 53 consecutive patients who underwent prostate mpMRI and subsequent prostate tissue sampling was performed. Two board-certified radiologists (with 4 and 12 years of experience) blinded to the clinical information interpreted anonymized exams using the PI-RADS v2.1 framework without and with an AI-assistance tool. The AI software tool provided radiologists with gland segmentation and automated lesion detection assigning a probability score for the likelihood of the presence of clinically significant prostate cancer (csPCa). The reference standard for all cases was the prostate pathology from systematic and targeted biopsies. Statistical analyses assessed interrater agreement and compared diagnostic performances with and without AI assistance. Within the entire cohort, 42 patients (79 %) harbored Gleason-positive disease, with 25 patients (47 %) having csPCa. Radiologists' diagnostic performance for csPCa was significantly improved over conventional interpretation with AI assistance (reader A: AUC 0.82 vs. 0.72, p = 0.03; reader B: AUC 0.78 vs. 0.69, p = 0.03). Without AI assistance, 81 % (n = 36; 95 % CI: 0.89-0.91) of the lesions were scored similarly by radiologists for lesion-level characteristics, and with AI assistance, 59 % (26, 0.82-0.89) of the lesions were scored similarly. For reader A, there was a significant difference in PI-RADS scores (p = 0.02) between AI-assisted and non-assisted assessments. Signficant differences were not detected for reader B. AI-assisted prostate mMRI interpretation improved radiologist diagnostic performance over conventional interpretation independent of reader experience.

Obesity is a risk factor for diverticulitis. However, it remains unclear whether visceral fat area, a more precise measurement of abdominal fat, is associated with the risk of diverticulitis. To estimate the risk of incident and recurrent diverticulitis according to visceral fat area. A retrospective cohort study. The Mass General Brigham Biobank. A total of 6654 patients who underwent abdominal CT for clinical indications and had no diagnosis of diverticulitis, IBD, or cancer before the scan were included. Visceral fat area, subcutaneous fat area, and skeletal muscle area were quantified using a deep-learning model applied to abdominal CT. The main exposures were z -scores of body composition metrics normalized by age, sex, and race. Diverticulitis cases were identified using the International Classification of Diseases codes for the primary or admitting diagnosis from the electronic health records. The risks of incident diverticulitis, complicated diverticulitis, and recurrent diverticulitis requiring hospitalization according to quartiles of body composition metrics z -scores were estimated. A higher visceral fat area z -score was associated with an increased risk of incident diverticulitis (multivariable HR comparing the highest vs lowest quartile, 2.09; 95% CI, 1.48-2.95; p for trend <0.0001), complicated diverticulitis (HR, 2.56; 95% CI, 1.10-5.99; p for trend = 0.02), and recurrence requiring hospitalization (HR, 2.76; 95% CI, 1.15-6.62; p for trend = 0.03). The association between visceral fat area and diverticulitis was not materially different among different strata of BMI. Subcutaneous fat area and skeletal muscle area were not significantly associated with diverticulitis. The study population was limited to individuals who underwent CT scans for medical indication. Higher visceral fat area derived from CT was associated with incident and recurrent diverticulitis. Our findings provide insight into the underlying pathophysiology of diverticulitis and may have implications for preventive strategies. See Video Abstract . ANTECEDENTES:La obesidad es un factor de riesgo de la diverticulitis. Sin embargo, sigue sin estar claro si el área de grasa visceral, con medida más precisa de la grasa abdominal esté asociada con el riesgo de diverticulitis.OBJETIVO:Estimar el riesgo de diverticulitis incidente y recurrente de acuerdo con el área de grasa visceral.DISEÑO:Un estudio de cohorte retrospectivo.AJUSTE:El Biobanco Mass General Brigham.PACIENTES:6.654 pacientes sometidos a una TC abdominal por indicaciones clínicas y sin diagnóstico de diverticulitis, enfermedad inflamatoria intestinal o cáncer antes de la exploración.PRINCIPALES MEDIDAS DE RESULTADOS:Se cuantificaron, área de grasa visceral, área de grasa subcutánea y área de músculo esquelético, utilizando un modelo de aprendizaje profundo aplicado a la TC abdominal. Las principales exposiciones fueron puntuaciones z de métricas de composición corporal, normalizadas por edad, sexo y raza. Los casos de diverticulitis se definieron con los códigos ICD para el diagnóstico primario o de admisión de los registros de salud electrónicos. Se estimaron los riesgos de diverticulitis incidente, diverticulitis complicada y diverticulitis recurrente que requiriera hospitalización según los cuartiles de las puntuaciones z de las métricas de composición corporal.RESULTADOS:Una puntuación z más alta del área de grasa visceral se asoció con un mayor riesgo de diverticulitis incidente (HR multivariable que compara el cuartil más alto con el más bajo, 2,09; IC del 95 %, 1,48-2,95; P para la tendencia < 0,0001), diverticulitis complicada (HR, 2,56; IC del 95 %, 1,10-5,99; P para la tendencia = 0,02) y recurrencia que requiriera hospitalización (HR, 2,76; IC del 95 %, 1,15-6,62; P para la tendencia = 0,03). La asociación entre el área de grasa visceral y la diverticulitis no fue materialmente diferente entre los diferentes estratos del índice de masa corporal. El área de grasa subcutánea y el área del músculo esquelético no se asociaron significativamente con la diverticulitis.LIMITACIONES:La población del estudio se limitó a individuos sometidos a tomografías computarizadas por indicación médica.CONCLUSIÓN:Una mayor área de grasa visceral derivada de la tomografía computarizada se asoció con diverticulitis incidente y recurrente. Nuestros hallazgos brindan información sobre la fisiopatología subyacente de la diverticulitis y pueden tener implicaciones para las estrategias preventivas. (Traducción: Dr. Fidel Ruiz Healy ).

The generation of Synthetic Computed Tomography (sCT) images has become a pivotal methodology in modern clinical practice, particularly in the context of Radiotherapy (RT) treatment planning. The use of sCT enables the calculation of doses, pushing towards Magnetic Resonance Imaging (MRI) guided radiotherapy treatments. Moreover, with the introduction of MRI-Positron Emission Tomography (PET) hybrid scanners, the derivation of sCT from MRI can improve the attenuation correction of PET images. Deep learning methods for MRI-to-sCT have shown promising results, but their reliance on single-centre training dataset limits generalisation capabilities to diverse clinical settings. Moreover, creating centralised multi-centre datasets may pose privacy concerns. To address the aforementioned issues, we introduced FedSynthCT-Brain, an approach based on the Federated Learning (FL) paradigm for MRI-to-sCT in brain imaging. This is among the first applications of FL for MRI-to-sCT, employing a cross-silo horizontal FL approach that allows multiple centres to collaboratively train a U-Net-based deep learning model. We validated our method using real multicentre data from four European and American centres, simulating heterogeneous scanner types and acquisition modalities, and tested its performance on an independent dataset from a centre outside the federation. In the case of the unseen centre, the federated model achieved a median Mean Absolute Error (MAE) of 102.0 HU across 23 patients, with an interquartile range of 96.7-110.5 HU. The median (interquartile range) for the Structural Similarity Index (SSIM) and the Peak Signal to Noise Ratio (PNSR) were 0.89 (0.86-0.89) and 26.58 (25.52-27.42), respectively. The analysis of the results showed acceptable performances of the federated approach, thus highlighting the potential of FL to enhance MRI-to-sCT to improve generalisability and advancing safe and equitable clinical applications while fostering collaboration and preserving data privacy.

Identifying patients at risk of low bone mineral density (BMD) from X-rays presents an attractive approach to increase case finding. This paper showed the diagnostic accuracy, reproducibility, and robustness of a new technology: OsteoSight™. OsteoSight could increase diagnosis and preventive treatment rates for patients with low BMD. This study aimed to evaluate the diagnostic accuracy, reproducibility, and robustness of OsteoSight™, an automated image analysis tool designed to identify low bone mineral density (BMD) from routine hip and pelvic X-rays. Given the global rise in osteoporosis-related fractures and the limitations of current diagnostic paradigms, OsteoSight offers a scalable solution that integrates into existing clinical workflows. Performance of the technology was tested across three key areas: (1) diagnostic accuracy in identifying low BMD as compared to dual-energy X-ray absorptiometry (DXA), the clinical gold standard; (2) reproducibility, through analysis of two images from the same patient; and (3) robustness, by evaluating the tool's performance across different patient demographics and X-ray scanner hardware. The diagnostic accuracy of OsteoSight for identifying patients at risk of low BMD was area under the receiver operating characteristic curve (AUROC) 0.834 [0.789-0.880], with consistent results across subgroups of clinical confounders and X-ray scanner hardware. Specificity 0.852 [0.783-0.930] and sensitivity 0.628 [0.538-0.743] met pre-specified acceptance criteria. The pre-processing pipeline successfully excluded unsuitable cases including incorrect body parts, metalwork, and unacceptable femur positioning. The results demonstrate that OsteoSight is accurate in identifying patients with low BMD. This suggests its utility as an opportunistic assessment tool, especially in settings where DXA accessibility is limited or not recently performed. The tool's reproducibility and robust performance across various clinical confounders further supports its integration into routine orthopedic and medical practices, potentially broadening the reach of osteoporosis assessment and enabling earlier intervention for at-risk patients.

To assess the impact of artificial intelligence (AI) on the diagnostic performance of radiologists with varying experience levels in mammography reading, considering single and simulated double reading approaches. In this retrospective study, 150 mammography examinations (30 with pathology-confirmed malignancies, 120 without malignancies [confirmed by 2-year follow-up]) were reviewed according to five approaches: A) human single reading by 26 radiologists of varying experience; B) AI single reading (Lunit INSIGHT MMG; C) human single reading with simultaneous AI support; D) simulated human-human double reading; E) simulated human-AI double reading, with AI as second independent reader flagging cases with a cancer probability ≥10 %. Sensitivity and specificity were calculated and compared using McNemar's test, univariate and multivariable logistic regression. Compared to single reading without AI support, single reading with simultaneous AI support improved mean sensitivity from 69.2 % (standard deviation [SD] 15.6) to 84.5 % (SD 8.1, p < 0.001), providing comparable mean specificity (91.8 % versus 90.8 %, p = 0.06). The sensitivity increase provided by the AI-supported single reading was largest in the group of radiologists with a sensitivity below the median in the non-supported single reading, from 56.7 % (SD 12.1) to 79.7 % (SD 10.2, p < 0.001). In the simulated human-AI double reading approach, sensitivity further increased to 91.8 % (SD 3.4), surpassing that of the human-human simulated double reading (87.4 %, SD 8.8, p = 0.016), with comparable mean specificity (from 84.0 % to 83.0 %, p = 0.17). AI support significantly enhanced sensitivity across all reading approaches, particularly benefiting worse performing radiologists. In the simulated double reading approaches, AI incorporation as independent second reader significantly increased sensitivity without compromising specificity.

Silicosis, a debilitating occupational lung disease caused by inhaling crystalline silica, continues to be a significant global health issue, especially with the increasing use of engineered stone (ES) surfaces containing high silica content. Traditional diagnostic methods, dependent on radiological interpretation, have low sensitivity, especially, in the early stages of the disease, and present variability between evaluators. This study explores the efficacy of deep learning techniques in automating the screening and staging of silicosis using chest X-ray images. Utilizing a comprehensive dataset, obtained from the medical records of a cohort of workers exposed to artificial quartz conglomerates, we implemented a preprocessing stage for rib-cage segmentation, followed by classification using state-of-the-art deep learning models. The segmentation model exhibited high precision, ensuring accurate identification of thoracic structures. In the screening phase, our models achieved near-perfect accuracy, with ROC AUC values reaching 1.0, effectively distinguishing between healthy individuals and those with silicosis. The models demonstrated remarkable precision in the staging of the disease. Nevertheless, differentiating between simple silicosis and progressive massive fibrosis, the evolved and complicated form of the disease, presented certain difficulties, especially during the transitional period, when assessment can be significantly subjective. Notwithstanding these difficulties, the models achieved an accuracy of around 81% and ROC AUC scores nearing 0.93. This study highlights the potential of deep learning to generate clinical decision support tools to increase the accuracy and effectiveness in the diagnosis and staging of silicosis, whose early detection would allow the patient to be moved away from all sources of occupational exposure, therefore constituting a substantial advancement in occupational health diagnostics.