Bengbu Medical College

This study aimed to investigate the strength of the correlation between organ doses and four computed tomography (CT) dose metrics, and to explore a method for rapidly estimating organ doses in patients undergoing chest-abdominal-pelvic (CAP) CT examinations. We retrospectively collected DICOM images of 43 patients who underwent CAP CT examinations. These images were imported into Archer-CT for organ segmentation and dose calculation. Then, regarding the six radiosensitive organs (spinal cord, lung, oesophagus, stomach, liver, and bladder) that were included in the study, various radiation dose metrics were calculated based on size-specific dose estimate (SSDE) derived from equivalent water diameters ( Dw ), i.e. SSDE centre based on Dw in the centre of the scan range, the SSDE mean based on the mean SSDEs of all slices, and the organ-specific SSDE organ for various organs, as prescribed in American Association of Physicists in Medicine Report 220. Subsequently, correlation analyses were applied to evaluate the relationship between the organ doses and each radiation dose metric and to derive conversion factors for rapid estimation of organ doses. The ranked linear correlations of the four dose metrics with each organ dose were as follows, in descending order: SSDE organ , SSDE mean , CTDI vol , and SSDE centre . The strongest and weakest correlations were found between organ dose to liver and SSDE Liver ( R2 = 0.88),and between organ dose to the bladder and SSDE Bladder ( R2 = 0.62) respectively, while the R2 of the rest of the organ doses to SSDE organ was around 0.8. The conversion coefficients for estimating organ doses based on SSDE organ for the spinal cord, lung, oesophagus, liver, stomach, and bladder were 0.75, 1.24, 0.89, 1.17, 1.18, and 0.83, respectively. Higher correlations were observed between organ doses and SSDE organ for organs involved in this study during CAP CT examinations. Thus, SSDE organ can be used to simplify and estimate the individualised organ dose for CAP CT examinations.

Chronic kidney disease (CKD) is getting more common in elderly people with metabolic syndrome, but early detection and risk prediction are still hard. So we created and validated a CKD risk prediction model tailored for these patients to get past the limits of existing clinical screening tools.

We used data from the National Health and Nutrition Examination Survey and the China Health and Retirement Longitudinal Study to develop and validate the CKD prediction model. We defined CKD based on an estimated glomerular filtration rate < 60 mL/min/1.73 m 2 , calculated with the chronic kidney disease epidemiology collaboration (CKD-EPI) formula. We did feature selection with a mix of strict methods and evaluated model performance across both internal and external validation cohorts. We used 10 machine learning algorithms and four data balancing strategies. The final CKD prediction model had strong predictive power and clinical interpretability for early detection of prevalent CKD.

The final model, which uses four key variables (uric acid to high-density lipoprotein cholesterol ratio (UHR), age, uric acid, blood urea nitrogen (BUN)), had predictive accuracies of 0.864 (internal) and 0.831 (external) when using logistic regression. This shows it is robust and can be applied widely. Shapley Additive exPlanations analysis and restricted cubic spline analysis found significant links between the key features and CKD risk.

This study developed a reliable, interpretable CKD risk prediction model using routine clinical indicators. It gives an effective tool for large-scale early CKD screening and stratified management in primary care and has high clinical applicability.