Deep learning body-composition analysis of clinically acquired CT-scans estimates creatinine excretion with high accuracy in patients and healthy individuals

Radiomic-based biomarkers: Transforming age and body composition metrics into personalized age-informed indices

Chronological age has been the standard for quantifying the aging process. While it is simple to quantify it cannot fully discern the biological variability of aging between individuals. The growing body of interest in this variability of human aging has led to the introduction of new biomarkers to operationalize biological age. The inclusion of body composition may provide additional value to biological aging as a prediction and estimation factor of individual health outcomes. Diagnostic images based on radiomic techniques such as Computed Tomography contain an untapped wealth of patient-specific data that remain inaccessible to healthcare providers. These images are beneficial for collecting information from body composition that adds precision and granularity when compared to traditional measures. This information can subsequently be aggregated to construct models for changes in the human body associated with aging. In addition, aging leads to a natural decline in the best parameter of drug dosing in older adults, glomerular filtration rate. Since the conventional models of kidney function are correlated with age and body composition, the radiomic biomarkers representing age-related changes in body composition may also serve as potential new imaging biomarkers of kidney function for personalized dosing. Our review introduces potential radiomic biomarkers as measures of body composition change targeting the aging processes. As a functional example, we have hypothesized an age-related model of radiomics as a covariate of kidney function to improve personalized dosing. Future research focusing on evaluating this hypothesis in human subject studies is acknowledged.

The correlation of CT-derived muscle density, skeletal muscle index, and visceral adipose tissue with nutritional status in severely injured patients

BACKGROUND

This study explored if computerized tomography-derived body composition parameters (CT-BCPs) are related to malnutrition in severely injured patients admitted to the Intensive Care Unit (ICU).

METHODS

This prospective cohort study included severely injured (Injury Severity Score ≥ 16) patients, admitted to the ICU of three level-1 trauma centers between 2018 and 2022. Abdominal CT scans were retrospectively analyzed to assess the CT-BCPs: muscle density (MD), skeletal muscle index (SMI), and visceral adipose tissue (VAT). The Subjective Global Assessment was used to diagnose malnutrition at ICU admission and on day 5 of admission, and the modified Nutrition Risk in Critically ill at admission was used to assess the nutritional risk.

RESULTS

Seven (11%) of the 65 analyzed patients had malnutrition at ICU admission, increasing to 23 patients (35%) on day 5. Thirteen (20%) patients had high nutritional risk. CT-BCPs were not related to malnutrition at ICU admission and on day 5. Patients with high nutritional risk at admission had lower MD (median (IQR) 32.1 HU (25.8-43.3) vs. 46.9 HU (37.7-53.3); p < 0.01) and higher VAT (median 166.5 cm2 (80.6-342.6) vs. 92.0 cm2 (40.6-148.2); p = 0.01) than patients with low nutritional risk.

CONCLUSION

CT-BCPs do not seem related to malnutrition, but low MD and high VAT may be associated with high nutritional risk. These findings may prove beneficial for clinical practice, as they suggest that CT-derived parameters may provide valuable information on nutritional risk in severely injured patients, in addition to conventional nutritional assessment and screening tools.

LEVEL OF EVIDENCE

Level III, Prognostic/Epidemiological.

Estimation of renal function immediately after cessation of continuous renal replacement therapy at the ICU

Estimating glomerular filtration (eGFR) after Continuous Renal Replacement Therapy (CRRT) is important to guide drug dosing and to assess the need to re-initiate CRRT. Standard eGFR equations cannot be applied as these patients neither have steady-state serum creatinine concentration nor average muscle mass. In this study we evaluate the combination of dynamic renal function with CT-scan based correction for aberrant muscle mass to estimate renal function immediately after CRRT cessation. We prospectively included 31 patients admitted to an academic intensive care unit (ICU) with a total of 37 CRRT cessations and measured serum creatinine before cessation (T1), directly (T2) and 5 h (T3) after cessation and the following two days when eGFR stabilized (T4, T5). We used the dynamic creatinine clearance calculation (D3C) equation to calculate eGFR (D3CGFR) and creatinine clearance (D3Ccreat) between T2-T3. D3Ccreat was corrected for aberrant muscle mass when a CT-scan was available using the CRAFT equation. We compared D3CGFR to stabilized CKD-EPI at T5 and D3CCreat to 4-h urinary creatinine clearance (4-h uCrCl) between T2-T3. We retrospectively validated these results in a larger retrospective cohort (NICE database; 1856 patients, 2064 cessations). The D3CGFR was comparable to observed stabilized CKD-EPI at T5 in the prospective cohort (MPE = - 1.6 ml/min/1.73 m2, p30 = 76%) and in the retrospective NICE-database (MPE = 3.2 ml/min/1.73 m2, p30 = 80%). In the prospective cohort, the D3CCreat had poor accuracy compared to 4-h uCrCl (MPE = 17 ml/min/1.73 m2, p30 = 24%). In a subset of patients (n = 13) where CT-scans were available, combination of CRAFT and D3CCreat improved bias and accuracy (MPE = 8 ml/min/1.73 m2, RMSE = 18 ml/min/1.73 m2) versus D3CCreat alone (MPE = 18 ml/min/1.73 m2, RMSE = 32 ml/min/1.73 m2). The D3CGFR improves assessment of eGFR in ICU patients immediately after CRRT cessation. Although the D3CCreat had poor association with underlying creatinine clearance, inclusion of CT derived biometric parameters in the dynamic renal function algorithm further improved the performance, stressing the role of muscle mass integration into renal function equations in critically ill patients.

Meta-analysis guided development of a standard artificial urine

In this study, we present the first meta-analysis of human urine reported in the literature, drawing data from a total of 35 articles with a combined participant count of 14,021. Through this analysis, we have developed an artificial urine (AU) composition that can be adjusted within typical physiological parameters for in vitro applications. Our findings demonstrate the utility of this AU in determining the solubility of nitrofurantoin, particularly in the context of crystalluria. Notably, we observe that in saline, nitrofurantoin solubility, within the framework of its urinary pharmacokinetics, suggests a risk of crystalluria. However, in AU, this risk is mitigated due to complexation with urea. More broadly, we anticipate that our developed formulation will serve as a foundation for translational studies across biomedical and pharmaceutical sciences.

Myosteatosis and aortic calcium score on abdominal CT as prognostic markers in non-dialysis chronic kidney disease patients

We aimed to examine the relationship between abdominal computed tomography (CT)-based body composition data and both renal function decline and all-cause mortality in patients with non-dialysis chronic kidney disease (CKD). This retrospective study comprised non-dialysis CKD patients who underwent consecutive unenhanced abdominal CT between January 2010 and December 2011. CT-based body composition was measured using semiautomated method that included visceral fat, subcutaneous fat, skeletal muscle area and density, and abdominal aortic calcium score (AAS). Sarcopenia and myosteatosis were defined by decreased skeletal muscle index (SMI) and decreased skeletal muscle density, respectively, each with specific cutoffs. Risk factors for CKD progression and survival were identified using logistic regression and Cox proportional hazard regression models. Survival between groups based on myosteatosis and AAS was compared using the Kaplan-Meier curve. 149 patients (median age: 70 years) were included; 79 (53.0%) patients had sarcopenia and 112 (75.2%) had myosteatosis. The median AAS was 560.9 (interquartile range: 55.7-1478.3)/m2. The prognostic factors for CKD progression were myosteatosis [odds ratio (OR) = 4.31, p = 0.013] and high AAS (OR = 1.03, p = 0.001). Skeletal muscle density [hazard ratio (HR) = 0.93, p = 0.004] or myosteatosis (HR = 4.87, p = 0.032) and high AAS (HR = 1.02, p = 0.001) were independent factors for poor survival outcomes. The presence of myosteatosis and the high burden of aortic calcium were significant factors for CKD progression and survival in patients with non-dialysis CKD.

Body composition radiomic features as a predictor of survival in patients with non-small cellular lung carcinoma: A multicenter retrospective study

OBJECTIVES

This study combined two novel approaches in oncology patient outcome predictions-body composition and radiomic features analysis. The aim of this study was to validate whether automatically extracted muscle and adipose tissue radiomic features could be used as a predictor of survival in patients with non-small cell lung cancer.

METHODS

The study included 178 patients with non-small cell lung cancer receiving concurrent platinum-based chemoradiotherapy. Abdominal imaging was conducted as a part of whole-body positron emission tomography/computed tomography performed before therapy. Methods used included automated assessment of the volume of interest using densely connected convolutional network classification model - DenseNet121, automated muscle and adipose tissue segmentation using U-net architecture implemented in nnUnet framework, and radiomic features extraction. Acquired body composition radiomic features and clinical data were used for overall and 1-y survival prediction using machine learning classification algorithms.

RESULTS

The volume of interest detection model achieved the following metric scores: 0.98 accuracy, 0.89 precision, 0.96 recall, and 0.92 F1 score. Automated segmentation achieved a median dice coefficient >0.99 in all segmented regions. We extracted 330 body composition radiomic features for every patient. For overall survival prediction using clinical and radiomic data, the best-performing feature selection and prediction method achieved areas under the curve-receiver operating characteristic (AUC-ROC) of 0.73 (P < 0.05); for 1-y survival prediction AUC-ROC was 0.74 (P < 0.05).

CONCLUSION

Automatically extracted muscle and adipose tissue radiomic features could be used as a predictor of survival in patients with non-small cell lung cancer.

Imaging-based deep learning in kidney diseases: recent progress and future prospects

Kidney diseases result from various causes, which can generally be divided into neoplastic and non-neoplastic diseases. Deep learning based on medical imaging is an established methodology for further data mining and an evolving field of expertise, which provides the possibility for precise management of kidney diseases. Recently, imaging-based deep learning has been widely applied to many clinical scenarios of kidney diseases including organ segmentation, lesion detection, differential diagnosis, surgical planning, and prognosis prediction, which can provide support for disease diagnosis and management. In this review, we will introduce the basic methodology of imaging-based deep learning and its recent clinical applications in neoplastic and non-neoplastic kidney diseases. Additionally, we further discuss its current challenges and future prospects and conclude that achieving data balance, addressing heterogeneity, and managing data size remain challenges for imaging-based deep learning. Meanwhile, the interpretability of algorithms, ethical risks, and barriers of bias assessment are also issues that require consideration in future development. We hope to provide urologists, nephrologists, and radiologists with clear ideas about imaging-based deep learning and reveal its great potential in clinical practice.Critical relevance statement The wide clinical applications of imaging-based deep learning in kidney diseases can help doctors to diagnose, treat, and manage patients with neoplastic or non-neoplastic renal diseases.Key points• Imaging-based deep learning is widely applied to neoplastic and non-neoplastic renal diseases.• Imaging-based deep learning improves the accuracy of the delineation, diagnosis, and evaluation of kidney diseases.• The small dataset, various lesion sizes, and so on are still challenges for deep learning.

Development and validation of a new equation based on plasma creatinine and muscle mass assessed by CT scan to estimate glomerular filtration rate: a cross-sectional study

Background

Inter-individual variations of non-glomerular filtration rate (GFR) determinants of serum creatinine, such as muscle mass, account for the imperfect performance of estimated GFR (eGFR) equations. We aimed to develop an equation based on creatinine and total lumbar muscle cross-sectional area measured by unenhanced computed tomography scan at the third lumbar vertebra.

Methods

The muscle mass-based eGFR (MMB-eGFR) equation was developed in 118 kidney donor candidates (iohexol clearance) using linear regression. Validation cohorts included 114 healthy subjects from another center (⁵¹Cr-EDTA clearance, validation population 1), 55 patients with chronic diseases (iohexol, validation population 2), and 60 patients with highly discordant creatinine and cystatin C-based eGFR, thus presumed to have atypical non-GFR determinants of creatinine (⁵¹Cr-EDTA, validation population 3). Mean bias was the mean difference between eGFR and measured GFR, precision the standard deviation (SD) of the bias, and accuracy the percentage of eGFR values falling within 20% and 30% of measured GFR.

Results

In validation population 1, performance of MMB-eGFR was not different from those of CKD-EPI_Cr2009 and CKD-EPI_Cr2021. In validation population 2, MMB-eGFR was unbiased and displayed better precision than CKD-EPI_Cr2009, CKD-EPI_Cr2021 and EKFC (SD of the biases: 13.1 vs 16.5, 16.8 and 15.9 mL/min/1.73 m²). In validation population 3, MMB-eGFR had better precision and accuracy {accuracy within 30%: 75.0% [95% confidence interval (CI) 64.0-86.0] vs 51.5% (95% CI 39.0-64.3) for CKD-EPI_Cr2009, 43.3% (95% CI 31.0-55.9) for CKD-EPICr2021, and 53.3% (95% CI 40.7-66.0) for EKFC}. Difference in bias between Black and white subjects was -2.1 mL/min/1.73 m² (95% CI -7.2 to 3.0), vs -8.4 mL/min/1.73 m² (95% CI -13.2 to -3.6) for CKD-EPI_Cr2021.

Conclusion

MMB-eGFR displayed better performances than equations based on demographics, and could be applied to subjects of various ethnic backgrounds.

Role of Machine Learning-Based CT Body Composition in Risk Prediction and Prognostication: Current State and Future Directions

CT body composition analysis has been shown to play an important role in predicting health and has the potential to improve patient outcomes if implemented clinically. Recent advances in artificial intelligence and machine learning have led to high speed and accuracy for extracting body composition metrics from CT scans. These may inform preoperative interventions and guide treatment planning. This review aims to discuss the clinical applications of CT body composition in clinical practice, as it moves towards widespread clinical implementation.