Can renal sonography be a reliable diagnostic tool in the assessment of chronic kidney disease?

Just Get in Line: Rural-Urban Differences in Access to Pediatric Renal-Bladder Ultrasounds

INTRODUCTION

Health-care disparities in rural and underserved areas may be exacerbated by the pandemic, personnel challenges, and supply chain limitations. This study aimed to quantify current variation in rural and urban pediatric renal ultrasound availability.

METHODS

We identified all hospitals statewide and contacted radiology departments posing as a parent trying to schedule an appointment for a routine pediatric renal-bladder ultrasound. Intervals between day of contact and first available appointment were compared between rural and urban institutions.

RESULTS

We were able to contact 42/48 (87.5%) rural hospitals, and 20/39 (51.3%) urban hospitals. Scheduling could not be completed in 5 rural and 7 urban hospitals. The median wait time for the 37 remaining rural and 13 remaining urban hospitals was similar: 7 (range: 0-21) days in rural hospitals and 6 (range: 0-17) days in urban hospitals (P = 0.81). If contact was made, the likelihood of scheduling within 7 d was similar in rural and urban areas (odds ratio [OR] = 0.23; 95% confidence interval [CI] 0.03-1.97; P = 0.18). However, patients were much more likely to have a completed call at a rural hospital (OR = 6.65; 95% CI: 2.3-19.2; P = 0.0005), and so in reality, patients were 2.89 times as likely to be able to schedule an renal-bladder ultrasound within 7 d at a rural compared with an urban institution (95% CI: 1.19-7.03; P = 0.019).

CONCLUSIONS

While access to pediatric renal sonograms was similar within a week at rural and urban institutions once telephone contact was made, it was significantly more difficult to schedule appointments at urban institutions.

Diagnostic performance of renal cortical elasticity by supersonic shear wave imaging in pediatric glomerular disease

PURPOSE

To explore the diagnostic performance of renal cortical elasticity expressed by Young's modulus (YM) using the supersonic shear wave imaging (SSI) technique in pediatric glomerular disease.

MATERIALS AND METHODS

Seventy-one children with glomerular disease confirmed by renal biopsy and sixty healthy volunteers were enrolled in this study. Conventional and SSI ultrasound examinations were performed in all individuals for both kidneys. We measured renal length, renal width, renal thickness, parenchyma thickness, interlobar arterial resistive index (RI) and the YM of the middle and lower pole.

RESULTS

Regardless of which pole and which side of the kidney, the YM in the disease group was significantly higher than that in the control group (P < 0.001). The YM of the middle pole in the left kidney demonstrated the largest AUC (0.936, P < 0.001), and the corresponding cut-off value was 15.48 kPa with a sensitivity of 87.3% and a specificity of 86.7%. There was no significant difference in the YM among different pathological types of pediatric glomerular disease in the disease group, and the same in different grades of patients with Immunoglobulin A (IgA) nephropathy by Lee classification and the Oxford Classification as well as Henoch-Schonlein purpura nephritis (HSPN) by International Study of Kidney Disease in Children (ISKDC) classification (P > 0.05). We found positive but weak correlations between the YM and renal length (r = 0.299, P = 0.001), renal width (r = 0.408, P < 0.001), renal thickness (r = 0.299, P = 0.001), and parenchyma thickness (r = 0.212, P = 0.015), whereas the YM had no significant correlations with age, sex, BMI, interlobar arterial RI, and laboratory findings (P > 0.05).

CONCLUSIONS

SSI technology is a non-invasive and feasible method for the diagnosis of pediatric glomerular disease. However, SSI did not show good performance in distinguishing different pathological types and disease grades in our study.

Ultrasound-based radiomics analysis in the assessment of renal fibrosis in patients with chronic kidney disease

PURPOSE

Assessment of renal fibrosis non-invasively in chronic kidney disease (CKD) patients is still a clinical challenge. In this study, we aimed to establish a radiomics model integrating radiomics features derived from ultrasound (US) images with clinical characteristics for the assessment of renal fibrosis severity in CKD patients.

METHODS

A total of 160 patients with CKD who underwent kidney biopsy and renal US examination were prospectively enrolled. Patients were classified into the mild or moderate-severe fibrosis group based on pathology results. Radiomics features were extracted from the US images, and a radiomics signature was constructed using the maximum relevance minimum redundancy (mRMR) and least absolute shrinkage and selection operator (LASSO) regression algorithms. Multivariable logistic regression was employed to construct the radiomics model, which incorporated the radiomics signature and the selected clinical variables. The established model was evaluated for discrimination, calibration, and clinical utility in the derivation cohort and internal cross-validation (CV) analysis, respectively.

RESULTS

The radiomics signature, consisting of nine identified fibrosis-related features, achieved moderate discriminatory ability with an area under the receiver operating characteristic curve (AUC) of 0.72 (95% confidence interval (CI) 0.64-0.79). By combining the radiomics signature with significant clinical risk factors, the radiomics model showed satisfactory discrimination performance, yielding an AUC of 0.85 (95% CI 0.79-0.91) in the derivation cohort and a mean AUC of 0.84 (95% CI 0.77-0.92) in the internal CV analysis. It also demonstrated fine accuracy via the calibration curve. Furthermore, the decision curve analysis indicated that the model was clinically useful.

CONCLUSION

The proposed radiomics model showed favorable performance in determining the individualized risk of moderate-severe renal fibrosis in patients with CKD, which may facilitate more effective clinical decision-making.

Ultrasound elastography in chronic kidney disease: a systematic review and meta-analysis

Ultrasound elastography (USE) is a noninvasive technique for assessing tissue elasticity, and its application in nephrology has aroused growing interest in recent years. The purpose of this article is to systematically review the clinical application of USE in patients with chronic kidney disease (CKD), including native and transplanted kidneys, and quantitatively investigate differences in elasticity values between healthy individuals and CKD patients. Furthermore, we provide a qualitative analysis of the studies included, discussing the potential interplay between renal stiffness, estimated glomerular filtration rate, and fibrosis. In January 2022, a systematic search was carried out on the MEDLINE (PubMed) database, concerning studies on the application of USE in patients with CKD, including patients with transplanted kidneys. The results of the included studies were extracted by two independent researchers and presented mainly through a formal narrative summary. A meta-analysis of nine study parts from six studies was performed. A total of 647 studies were screened for eligibility and, after applying the exclusion and inclusion criteria, 69 studies were included, for a total of 6728 patients. The studies proved very heterogeneous in terms of design and results. The shear wave velocity difference of - 0.82 m/s (95% CI: - 1.72-0.07) between CKD patients and controls was not significant. This result agrees with the qualitative evaluation of included studies that found controversial results for the relationship between renal stiffness and glomerular filtration rate. On the contrary, a clear relationship seems to emerge between USE values and the degree of fibrosis. At present, due to the heterogeneity of results and technical challenges, large-scale application in the monitoring of CKD patients remains controversial.

Current progress in artificial intelligence-assisted medical image analysis for chronic kidney disease: A literature review

Chronic kidney disease (CKD) causes irreversible damage to kidney structure and function. Arising from various etiologies, risk factors for CKD include hypertension and diabetes. With a progressively increasing global prevalence, CKD is an important public health problem worldwide. Medical imaging has become an important diagnostic tool for CKD through the non-invasive identification of macroscopic renal structural abnormalities. Artificial intelligence (AI)-assisted medical imaging techniques aid clinicians in the analysis of characteristics that cannot be easily discriminated by the naked eye, providing valuable information for the identification and management of CKD. Recent studies have demonstrated the effectiveness of AI-assisted medical image analysis as a clinical support tool using radiomics- and deep learning-based AI algorithms for improving the early detection, pathological assessment, and prognostic evaluation of various forms of CKD, including autosomal dominant polycystic kidney disease. Herein, we provide an overview of the potential roles of AI-assisted medical image analysis for the diagnosis and management of CKD.

Ultrasound Renal Score to Predict the Renal Disease Prognosis in Patients with Diabetic Kidney Disease: An Investigative Study

Renal disease associated with type 2 diabetes mellitus (T2DM) has become the leading cause of chronic kidney disease (CKD). Renal ultrasonography is an imaging examination required in the work-up of renal disease. This study aimed to identify the differences in renal ultrasonographic findings between patients with and without DM, and to evaluate the relationship between renal ultrasound findings and renal prognosis in patients with DM. A total of 252 patients who underwent renal ultrasonography at Chungnam National University Hospital were included. Kidney disease progression was defined as a ≥10% decline in the annual estimated glomerular filtration rate (eGFR), which, in this paper, is referred to as ΔeGFR/year, or the initiation of renal replacement therapy after follow-up. The renal scoring system was evaluated by summing up the following items: the value of renal parenchymal echogenicity (0: normal; 1: mildly increased; and 2: increased) and the shape of the cortical margin (0: normal and 1: irregular; right kidney length/height (RH-0 or 1), mean cortical thickness/renal length/height (CKH-0 or 1), and cortical thickness/parenchymal thickness (CK/PK-0 or 1) based on the median: 0-above median, and 1-below median). Patients with DM had thicker renal PKH than those without, despite having lower eGFRs (0.91 ± 0.15, 0.86 ± 0.14, p = 0.006). In the progression group, the renal scores were significantly higher than those from the non-progression group. In the multivariate logistic regression analysis, the higher renal scores, presence of DM, and younger age were independently predicted for renal disease progression after adjusting for confounding variables, such as the presence of hypertension, serum hemoglobin and albumin levels, and UPCR. In conclusion, patients with high renal scores were significantly associated with renal disease progression. Our results suggest that renal ultrasonography at the time of diagnosis provides useful prognostic information in patients with kidney disease.

Changes in CT-Based Morphological Features of the Kidney with Declining Glomerular Filtration Rate in Chronic Kidney Disease

Chronic kidney disease (CKD) progression involves morphological changes in the kidney, such as decreased length and thickness, with associated histopathological alterations. However, the relationship between morphological changes in the kidneys and glomerular filtration rate (GFR) has not been quantitatively and comprehensively evaluated. We evaluated the three-dimensional size and shape of the kidney using computed tomography (CT)-derived features in relation to kidney function. We included 257 patients aged ≥18 years who underwent non-contrast abdominal CT at the Inha University Hospital. The features were quantified using predefined algorithms in the pyRadiomics package after kidney segmentation. All features, except for flatness, significantly correlated with estimated GFR (eGFR). The surface-area-to-volume ratio (SVR) showed the strongest negative correlation (r = -0.75, p < 0.0001). Kidney size features, such as volume and diameter, showed moderate to high positive correlations; other morphological features showed low to moderate correlations. The calculated area under the receiver operating characteristic (ROC) curve (AUC) for different features ranged from 0.51 (for elongation) to 0.86 (for SVR) for different eGFR thresholds. Diabetes patients had weaker correlations between the studied features and eGFR and showed less bumpy surfaces in three-dimensional visualization. We identified alterations in the CKD kidney based on various three-dimensional shape and size features, with their potential diagnostic value.

Prediction of acute kidney injury using a combined model of inflammatory vascular endothelium biomarkers and ultrasound indices

BACKGROUND

Acute kidney injury (AKI) is a common complication of sepsis, with the burden of long hospital admission. Early prediction of AKI is the most effective strategy for intervention and improvement of the outcomes.

OBJECTIVE

In our study, we aimed to investigate the predictive performance of the combined model using ultrasound indices (grayscale and Doppler indieces), endothelium injury (E-selectin, VCAM-1, ICAM1, Angiopoietin 2, syndecan-1, and eNOS) as well as inflammatory biomarkers (TNF-a, and IL-1β) to identify AKI.

METHODS

Sixty albino rats were divided into control and lipopolysaccharide (LPS) groups. Renal ultrasound, biochemical and immunohistological variables were recorded 6 hrs, 24 hrs, and 48 hrs after AKI.

RESULTS

Endothelium injury and inflammatory markers were found to be significantly increased early after AKI, and correlated significantly with kidney size reduction and renal resistance indices elevation.

CONCLUSIONS

Using area under the curve (AUC), the combined model was analyzed based on ultrasound and biochemical variables and provided the highest predictive value for renal injury.

Machine Learning-Aided Chronic Kidney Disease Diagnosis Based on Ultrasound Imaging Integrated with Computer-Extracted Measurable Features

Although ultrasound plays an important role in the diagnosis of chronic kidney disease (CKD), image interpretation requires extensive training. High operator variability and limited image quality control of ultrasound images have made the application of computer-aided diagnosis (CAD) challenging. This study assessed the effect of integrating computer-extracted measurable features with the convolutional neural network (CNN) on the ultrasound image CAD accuracy of CKD. Ultrasound images from patients who visited Severance Hospital and Gangnam Severance Hospital in South Korea between 2011 and 2018 were used. A Mask regional CNN model was used for organ segmentation and measurable feature extraction. Data on kidney length and kidney-to-liver echogenicity ratio were extracted. The ResNet18 model classified kidney ultrasound images into CKD and non-CKD. Experiments were conducted with and without the input of the measurable feature data. The performance of each model was evaluated using the area under the receiver operating characteristic curve (AUROC). A total of 909 patients (mean age, 51.4 ± 19.3 years; 414 [49.5%] men and 495 [54.5%] women) were included in the study. The average AUROC from the model trained using ultrasound images achieved a level of 0.81. Image training with the integration of automatically extracted kidney length and echogenicity features revealed an improved average AUROC of 0.88. This value further increased to 0.91 when the clinical information of underlying diabetes was also included in the model trained with CNN and measurable features. The automated step-wise machine learning-aided model segmented, measured, and classified the kidney ultrasound images with high performance. The integration of computer-extracted measurable features into the machine learning model may improve CKD classification.

Data Augmentation Based on Generative Adversarial Networks to Improve Stage Classification of Chronic Kidney Disease

The prevalence of chronic kidney disease (CKD) is estimated to be 13.4% worldwide and 15% in the United States. CKD has been recognized as a leading public health problem worldwide. Unfortunately, as many as 90% of CKD patients do not know that they already have CKD. Ultrasonography is usually the first and the most commonly used imaging diagnostic tool for patients at risk of CKD. To provide a consistent assessment of the stage classifications of CKD, this study proposes an auxiliary diagnosis system based on deep learning approaches for renal ultrasound images. The system uses the ACWGAN-GP model and MobileNetV2 pre-training model. The images generated by the ACWGAN-GP generation model and the original images are simultaneously input into the pre-training model MobileNetV2 for training. This classification system achieved an accuracy of 81.9% in the four stages of CKD classification. If the prediction results allowed a higher stage tolerance, the accuracy could be improved by up to 90.1%. The proposed deep learning method solves the problem of imbalance and insufficient data samples during training processes for an automatic classification system and also improves the prediction accuracy of CKD stage diagnosis.

A Physiological Approach to Pharmacokinetics in Chronic Kidney Disease

The conventional approach to approximating the pharmacokinetics of drugs in patients with chronic kidney disease (CKD) only accounts for changes in the estimated glomerular filtration rate. However, CKD is a systemic and multifaceted disease that alters many body systems. Therefore, the objective of this exercise was to develop and evaluate a whole-body mechanistic approach to predicting pharmacokinetics in patients with CKD. Physiologically based pharmacokinetic models were developed in PK-Sim v8.0 (www.open-systems-pharmacology.org) to mechanistically represent the disposition of 7 compounds in healthy human adults. The 7 compounds selected were eliminated by glomerular filtration and active tubular secretion by the organic cation transport system to varying degrees. After a literature search, the healthy adult models were adapted to patients with CKD by numerically accounting for changes in glomerular filtration rate, kidney volume, renal perfusion, hematocrit, plasma protein concentrations, and gastrointestinal transit. Literature-informed interindividual variability was applied to the physiological parameters to facilitate a population approach. Model performance in CKD was evaluated against pharmacokinetic data from 8 clinical trials in the literature. Overall, integration of the CKD parameterization enabled exposure predictions that were within 1.5-fold error across all compounds and patients with varying stages of renal impairment. Notable improvement was observed over the conventional approach to scaling exposure, which failed in all but 1 scenario in patients with advanced CKD. Further research is required to qualify its use for first-in-CKD dose selection and clinical trial planning for a wider selection of renally eliminated compounds, including those subject to anion transport.

Ultrasonographic predictors in chronic kidney disease: A hospital based case control study

BACKGROUND

Prevalence of Chronic Kidney Disease (CKD) is increasing globally with the concomitant upsurge in diabetes mellitus and hypertension. We explored the research question whether Ultrasonographic (US) renal parameters are potential predictors of CKD?

MATERIALS AND METHODS

A case control study was conducted at a tertiary care hospital that included 50 confirmed cases of CKD and 100 nondiseased controls. Renal length, renal parenchymal thickness, and renal cortical thickness were measured in both cases and controls by ultrasound examination. Corticomedullary differentiation and renal cortical echogenicity were also assessed.

RESULTS

US parameters of renal length, renal parenchymal thickness, and renal cortical thickness were found to be significantly and strongly associated with the presence of CKD. The strongest association was observed with renal cortical echogenicity (OR 27.33, 95% CI 8.82-84.63). The association of reduced renal cortical thickness (OR 6.14, 95% CI 1.59-23.62), and renal length (OR 2.72, 95% CI 1.13-8.26) were independent and significant predictors of presence of CKD.

CONCLUSIONS

Specific US parameters of renal cortical echogenicity, cortical thickness, and length of kidney have a strong potential for independently establishing the diagnosis and evaluation of progression of CKD.

Determination of the Accuracy of Supine and Prone Approaches to Sonographically Measured Kidney Dimensions

Objective:

To determine the accuracy of supine and prone approaches to sonographically measured kidney dimensions.

Methods:

The kidney dimensions of 109 participants were sonographically determined in supine and prone patient positions. The two measurements were compared with each other using the intra-class correlation, coefficient of variation for duplicate measurements and Bland-Altman plot. The two sets of measurements were each compared with measurements from computed tomography.

Results:

There was a very strong agreement between kidney dimensions in supine and prone positions. There was also an agreement between kidney dimensions in supine and prone positions and computed tomography measurements.

Conclusion:

The kidney dimensions obtained using patient-in-supine position and patient-in-prone position approaches may be equivalent and the two approaches may be used interchangeably.

ACR Appropriateness Criteria® Renal Failure

Renal failure can be divided into acute kidney injury and chronic kidney disease. Both are common and result in increased patient morbidity and mortality. The etiology is multifactorial and differentiation of acute kidney injury from chronic kidney disease includes clinical evaluation, laboratory tests, and imaging. The main role of imaging is to detect treatable causes of renal failure such as ureteral obstruction or renovascular disease and to evaluate renal size and morphology. Ultrasound is the modality of choice for initial imaging, with duplex Doppler reserved for suspected renal artery stenosis or thrombosis. CT and MRI may be appropriate, particularly for urinary tract obstruction. However, the use of iodinated and gadolinium-based contrast should be evaluated critically depending on specific patient factors and cost-benefit ratio. The American College of Radiology Appropriateness Criteria are evidence-based guidelines for specific clinical conditions that are reviewed annually by a multidisciplinary expert panel. The guideline development and revision include an extensive analysis of current medical literature from peer reviewed journals and the application of well-established methodologies (RAND/UCLA Appropriateness Method and Grading of Recommendations Assessment, Development, and Evaluation or GRADE) to rate the appropriateness of imaging and treatment procedures for specific clinical scenarios. In those instances where evidence is lacking or equivocal, expert opinion may supplement the available evidence to recommend imaging or treatment.

Potential value of three-dimensional ultrasonography in diagnosis of diabetic nephropathy in Chinese diabetic population with kidney injury

Background

To explore the potential value of three-dimensional ultrasonography (3DUS) and contrast-enhanced ultrasound (CEUS) in the diagnosis of diabetic nephropathy (DN) in Chinese diabetic patients with kidney injury.

Methods

Patients with type 2 diabetes mellitus and kidney injury in our hospital were enrolled, and the clinical characteristics as well as the laboratory data of patients were collected; 3DUS and CEUS were used to evaluate the morphological structure and blood perfusion of kidneys. Eligible patients were categorized into two groups based on renal biopsy results: DN group and non-diabetic renal diseases (NDRD) group. Correlation analysis and logistic regression analysis were applied to identify the risk factors of DN development.

Results

A total of 115 patients aged from 24 to 78 years old were recruited in the experiment, of which 64 patients (55.65%) and 51 patients (44.35%) were in the DN group and NDRD group, respectively. After correction to CKD stage, BMI and right kidney volume index were retained to identify patients with DN. The ROC of the logistic regression model had an AUC of 0.703 (95% CI: 0.591–0.815) and it was higher than both indicators.

Conclusion

3DUS has potential value in the diagnosis of diabetic nephropathy in Chinese diabetic population with kidney injury and may act as an auxiliary diagnosis for DN. More prospective and well-designed studies with larger samples are needed to verify the result.

Clinical decision support system to predict chronic kidney disease: A fuzzy expert system approach

BACKGROUND AND OBJECTIVES

Diagnosis and early intervention of chronic kidney disease are essential to prevent loss of kidney function and a large amount of financial resources. To this end, we developed a fuzzy logic-based expert system for diagnosis and prediction of chronic kidney disease and evaluate its robustness against noisy data.

METHODS

At first, we identified the diagnostic parameters and risk factors through a literature review and a survey of 18 nephrologists. Depending on the features selected, a set of fuzzy rules for the prediction of chronic kidney disease was determined by reviewing the literature, guidelines and consulting with nephrologists. Fuzzy expert system was developed using MATLAB software and Mamdani Inference System. Finally, the fuzzy expert system was evaluated using data extracted from 216 randomly selected medical records of patients with and without chronic kidney disease. We added noisy data to our dataset and compare the performance of the system on original and noisy datasets.

RESULTS

We selected 16 parameters for the prediction of chronic kidney disease. The accuracy, sensitivity, and specificity of the final system were 92.13 %, 95.37 %, and 88.88 %, respectively. The area under the curve was 0.92 and the Kappa coefficient was 0.84, indicating a very high correlation between the system diagnosis and the final diagnosis recorded in the medical records. The performance of the system on noisy input variables indicated that in the worse scenario, the accuracy, sensitivity, and specificity of the system decreased only by 4.43 %, 7.48 %, and 5.41 %, respectively.

CONCLUSION

Considering the desirable performance of the proposed expert system, the system can be useful in the prediction of chronic kidney disease.

How reliable is renal ultrasound to measure renal length and volume in patients with chronic kidney disease compared with magnetic resonance imaging?

Background

Renal length, volume, and parenchymal thickness are important clinical parameters, yet data concerning the accuracy and reproducibility of ultrasound (US)-based renal length and volume assessment in patients with chronic kidney disease (CKD) are scarce.

Purpose

To establish whether renal length, volume, and parenchymal thickness can be reliably measured with renal US in patients with CKD.

Material and Methods

All participants underwent renal US, immediately followed by 3-T magnetic resonance imaging (MRI). Renal length, width, transverse diameter, and parenchyma thickness were measured with both methods; renal volume was calculated using the ellipsoid formula. A total of 45 patients with CKD (eGFR [mean ± SD] 57.4 ± 4.4 mL/min/1.73 m2) and 46 participants without CKD (eGFR 97.0 ± 2.4 mL/min/1.73 m2) were included.

Results

US-measured renal length correlated strongly with MRI-measured renal length in no-CKD patients (Spearman’s r = 0.83 and 0.85 for the right and left kidney, respectively; P < 0.005) and CKD patients (r = 0.89 and 0.92 for the right and left kidney, respectively; P < 0.005). There was a significant but weaker correlation between MRI- and US-measured right and left renal volume (r = 0.72, P < 0.005) in no-CKD (r = 0.74 and r = 0.72, respectively; for both: P < 0.005) and CKD patients (r = 0.83 and 0.85, P < 0.005). Weak to moderate correlations were found for parenchyma thickness for the right (CKD group: r = 0.29, no-CKD: r = 0.23; for both: P < 0.05) and left kidney (CKD: r = 0.52, no-CKD group: r = 0.37, P < 0.05). Both intra-observer (Pearson’s correlations of 0.82 for the right and 0.89 for the left kidney) and inter-observer (Lin’s correlation coefficient of 0.90 for the right and 0.82 for the left kidney) reproducibility of US-assessed renal length was high.

Conclusions

US-based assessment of renal length in CKD patients is comparable to MRI measures. Both intra- and inter-observer reproducibility of US-assessed renal length in CKD patients are high. Measurements of US renal volume and parenchymal thickness should, however, be interpreted with caution.

Renal ultrasonographic abnormalities are associated with low glomerular filtration rate calculated by scintigraphy in dogs

Ultrasound provides information on kidney morphology, but studies relating structural and functional abnormalities in chronic kidney disease (CKD) are lacking. The aim of this descriptive cross-sectional study was to compare individual kidney (IK) B-mode ultrasound abnormalities to IK glomerular filtration rate (GFR) estimated by scintigraphy normalized to plasma volume (PV) in dogs, to evaluate if ultrasonographic findings were associated with low IKGFR/PV. Eighty privately owned dogs with and without clinical suspicion of CKD were prospectively enrolled, and kidney ultrasound and IKGFR/PV were evaluated independently. Ultrasound images were assessed retrospectively for subjective abnormalities (shape, cortical, and medullary hyperechogenicity), and kidney size was measured. The normal IKGFR/PV cutoff was derived from dogs in the study group with no history and clinical signs of kidney disease and normal blood and urine results (n = 28) and was 16.84 mL/min/L. Kidneys were categorized into normal, mild, moderate, and severe ultrasound changes according to subjective ultrasound grades. Associations were found between low IKGFR/PV and abnormal kidney shape (P = .0004), cortical hyperechogenicity (P = .0008), medullary hyperechogenicity (P < .0001), and low kidney volume (P = .0092). Apart from the moderate and severe category comparison, IKGFR/PV value significantly decreased with increasing severity of category. The combination of ultrasonographic subjective abnormalities had a high sensitivity (93.8%) and moderate specificity (65.7%) for detecting low IKGFR/PV. Kidneys with normal IKGFR/PV had a low frequency of mild ultrasound changes. Findings indicate kidneys with increasing number and grade of subjective ultrasound abnormalities are more likely to have a lower IKGFR/PV.

Automation of the kidney function prediction and classification through ultrasound-based kidney imaging using deep learning

Prediction of kidney function and chronic kidney disease (CKD) through kidney ultrasound imaging has long been considered desirable in clinical practice because of its safety, convenience, and affordability. However, this highly desirable approach is beyond the capability of human vision. We developed a deep learning approach for automatically determining the estimated glomerular filtration rate (eGFR) and CKD status. We exploited the transfer learning technique, integrating the powerful ResNet model pretrained on an ImageNet dataset in our neural network architecture, to predict kidney function based on 4,505 kidney ultrasound images labeled using eGFRs derived from serum creatinine concentrations. To further extract the information from ultrasound images, we leveraged kidney length annotations to remove the peripheral region of the kidneys and applied various data augmentation schemes to produce additional data with variations. Bootstrap aggregation was also applied to avoid overfitting and improve the model's generalization. Moreover, the kidney function features obtained by our deep neural network were used to identify the CKD status defined by an eGFR of <60 ml/min/1.73 m2. A Pearson correlation coefficient of 0.741 indicated the strong relationship between artificial intelligence (AI)- and creatinine-based GFR estimations. Overall CKD status classification accuracy of our model was 85.6% -higher than that of experienced nephrologists (60.3%-80.1%). Our model is the first fundamental step toward realizing the potential of transforming kidney ultrasound imaging into an effective, real-time, distant screening tool. AI-GFR estimation offers the possibility of noninvasive assessment of kidney function, a key goal of AI-powered functional automation in clinical practice.

Correlation of Point Shear Wave Velocity and Kidney Function in Chronic Kidney Disease

OBJECTIVES

Point shear wave elastography is a quantitative ultrasound-based imaging method used in the assessment of renal disease. Among point shear wave elastographic options, 2 techniques have been studied considerably: Virtual Touch quantification (VTQ; Siemens AG, Erlangen, Germany) and ElastPQ (EPQ; Philips Healthcare, Bothell, WA). Both rely on the tissue response to an acoustic beam generated by the ultrasound transducer. The data on renal VTQ are more extensive, whereas EPQ has been used less thus far in the assessment of the kidneys. This study aimed to evaluate the performance of EPQ in the kidney and compare it with VTQ.

METHODS

We studied 124 participants using EPQ: 22 with no renal disease and 102 with chronic kidney disease (CKD). Ninety-one were studied with both the EPQ and VTQ methods. We obtained 5 valid measurements in each kidney, expressed in meters per second.

RESULTS

The mean kidney stiffness measurements ± SD obtained with EPQ in the healthy control group were as follows: right kidney, 1.23 ± 0.33 m/s; and left kidney, 1.26 ± 0.32 m/s (P = .6). In the patients with CKD (all stages), the mean kidney stiffness measurements obtained were significantly lower: right kidney, 1.09 ± 0.39 m/s; and left kidney, 1.04 ± 0.38 m/s (P = .4). We observed that, similar to VTQ, EPQ values decreased with CKD progression, based on analysis of variance results using different CKD stages. From a receiver operating characteristic curve analysis, the cutoff value for an estimated glomerular filtration rate of less than 45 mL/min was 1.24 m/s, and the value for an estimated glomerular filtration rate of less than 30 mL/min was 1.07 m/s.

CONCLUSIONS

When using EPQ, the kidney shear wave velocity is decreased in patients with CKD, an observation similar to that obtained by using the VTQ method.

Ultrasound and color Doppler applications in chronic kidney disease

Chronic kidney disease (CKD) includes all clinical features and complications during the progression of various kidney conditions towards end-stage renal disease (ESRD). These conditions include immune and inflammatory disease such as: primary and hepatitis C virus (HCV)-related glomerulonephritis; infectious disease such as pyelonephritis with or without reflux and tuberculosis; vascular disease such as chronic ischemic nephropathy; hereditary and congenital disease such as polycystic disease and congenital cystic dysplasia; metabolic disease including diabetes and hyperuricemia; and systemic disease (collagen disease, vasculitis, myeloma). During the progression of CKD, ultrasound imaging and color Doppler imaging (US-CDI) can differentiate the etiology of the renal damage in only 50-70% of cases. Indeed, the end-stage kidney appears shrunken, reduced in volume (Ø < 9 cm), unstructured, amorphous, and with acquired cystic degeneration (small and multiple cysts involving the cortex and medulla) or nephrocalcinosis, but there are rare exceptions, such as polycystic kidney disease, diabetic nephropathy, and secondary inflammatory nephropathies. The main difficulties in the differential diagnosis are encountered in multifactorial CKD, which is commonly presented to the nephrologist at stage 4-5, when the kidney is shrunken, unstructured and amorphous. As in acute renal injury and despite the lack of sensitivity, US-CDI is essential for assessing the progression of renal damage and related complications, and for evaluating all conditions that increase the risk of CKD, such as lithiasis, recurrent urinary tract infections, vesicoureteral reflux, polycystic kidney disease and obstructive nephropathy. The timing and frequency of ultrasound scans in CKD patients should be evaluated case by case. In this review, we will consider the morpho-functional features of the kidney in all nephropathies that may lead to progressive CKD.

Shear wave elastography in the evaluation of renal parenchymal stiffness in patients with chronic kidney disease

OBJECTIVE

To investigate the use of shear wave elastography (SWE)-derived estimates of Young's modulus (YM) as an indicator to detect abnormal renal tissue diagnosed by estimated glomerular filtration rate (eGFR).

METHODS

The study comprised 106 chronic kidney disease (CKD) patients and 203 control subjects. Conventional ultrasound was performed to measure the kidney length and cortical thickness. SWE imaging was performed to measure renal parenchymal stiffness. Diagnostic performance of SWE and conventional ultrasound were correlated with serum creatinine, urea levels and eGFR.

RESULTS

Pearson's correlation coefficient revealed a negative correlation between YM measurements and eGFR (r = -0.576, p < 0.0001). Positive correlations between YM measurements and age (r = 0.321, p < 0.05), serum creatinine (r = 0.375, p < 0.0001) and urea (r = 0.287, p < 0.0001) were also observed. The area under the receiver operating characteristic curve for SWE (0.87) was superior to conventional ultrasound alone (0.35-0.37). The cut-off value of less or equal to 4.31 kPa suggested a non-diseased kidney (80.3% sensitivity, 79.5% specificity).

CONCLUSION

SWE was superior to renal length and cortical thickness in detecting CKD. A value of 4.31 kPa or less showed good accuracy in determining whether a kidney was diseased or not. Advances in knowledge: On SWE, CKD patients show greater renal parenchymal stiffness than non-CKD patients. Determining a cut-off value between normal and diseased renal parenchyma may help in early non-invasive detection and management of CKD.

A novel ultrasound technique to detect early chronic kidney disease

Chronic kidney disease (CKD) of unknown etiology is recognized as a major public health challenge and a leading cause of morbidity and mortality in the dry zone in Sri Lanka. CKD is asymptomatic and are diagnosed only in late stages. Evidence points to strong correlation between progression of CKD and kidney fibrosis. Several biochemical markers of renal fibrosis have been associated with progression of CKD. However, no marker is able to predict CKD consistently and accurately before being detected with traditional clinical tests (serum creatinine, and cystatin C, urine albumin or protein, and ultrasound scanning). In this paper, we hypothesize that fibrosis in the kidney, and therefore the severity of the disease, is reflected in the frequency spectrum of the scattered ultrasound from the kidney. We present a design of a simple ultrasound system, and a set of clinical and laboratory studies to identify spectral characteristics of the scattered ultrasound wave from the kidney that correlates with CKD. We believe that spectral parameters identified in these studies can be used to detect and stratify CKD at an earlier stage than what is possible with current markers of CKD.

A novel ultrasound technique to detect early chronic kidney disease

Chronic kidney disease (CKD) of unknown etiology is recognized as a major public health challenge and a leading cause of morbidity and mortality in the dry zone in Sri Lanka. CKD is asymptomatic and are diagnosed only in late stages. Evidence points to strong correlation between progression of CKD and kidney fibrosis. Several biochemical markers of renal fibrosis have been associated with progression of CKD. However, no marker is able to predict CKD consistently and accurately before being detected with traditional clinical tests (serum creatinine, and cystatin C, urine albumin or protein, and ultrasound scanning).

In this paper, we hypothesize that fibrosis in the kidney, and therefore the severity of the disease, is reflected in the frequency spectrum of the scattered ultrasound from the kidney. We present a design of a simple ultrasound system, and a set of clinical and laboratory studies to identify spectral characteristics of the scattered ultrasound wave from the kidney that correlates with CKD. We believe that spectral parameters identified in these studies can be used to detect and stratify CKD at an earlier stage than what is possible with current markers of CKD.

Role of ultrasonographic chronic kidney disease score in the assessment of chronic kidney disease

PURPOSE

Ultrasonography (US) is an inexpensive, noninvasive and easy imaging procedure to comment on the kidney disease. Data are limited about the relation between estimated glomerular filtration rate (e-GFR) and all 3 renal US parameters, including kidney length, parenchymal thickness and parenchymal echogenicity, in chronic kidney disease (CKD). In this study, we aimed to investigate the association between e-GFR and ultrasonographic CKD score calculated via these ultrasonographic parameters.

METHODS

One hundred and twenty patients with stage 1-5 CKD were enrolled in this study. The glomerular filtration rate was estimated by the Chronic Kidney Disease Epidemiology Collaboration equation. US was performed by the same radiologist who was blinded to patients' histories and laboratory results. US parameters including kidney length, parenchymal thickness and parenchymal echogenicity were obtained from both kidneys. All 3 parameters were scored for each kidney, separately. The sum of the average scores of these parameters was used to calculate ultrasonographic CKD score.

RESULTS

The mean age of patients was 63.34 ± 14.19 years. Mean kidney length, parenchymal thickness, ultrasonographic CKD score and median parenchymal echogenicity were found as 96.2 ± 12.3, 10.97 ± 2.59 mm, 6.28 ± 2.52 and 1.0 (0-3.5), respectively. e-GFR was positively correlated with kidney length (r = 0.343, p < 0.001), parenchymal thickness (r = 0.37, p < 0.001) and negatively correlated with CKD score (r = -0.587, p < 0.001) and parenchymal echogenicity (r = -0.683, p < 0.001). Receiver operating characteristic curve analysis for distinction of e-GFR lower than 60 mL/min showed that the ultrasonographic CKD score higher than 4.75 was the best parameter with the sensitivity of 81% and positive predictivity of 92% (AUC, 0.829; 95% CI, 0.74-0.92; p < 0.001).

CONCLUSION

We found correlation between e-GFR and ultrasonographic CKD score via using all ultrasonographic parameters. Also, our study showed that ultrasonographic CKD score can be useful for distinction of CKD stage 3-5 from stage 1 and 2. We suggested that the ultrasonographic CKD score provided more objective data in the assessment of CKD.

Multi-Parametric Evaluation of Chronic Kidney Disease by MRI: A Preliminary Cross-Sectional Study

Background

The current clinical classification of chronic kidney disease (CKD) is not perfect and may be overestimating both the prevalence and the risk for progressive disease. Novel markers are being sought to identify those at risk of progression. This preliminary study evaluates the feasibility of magnetic resonance imaging based markers to identify early changes in CKD.

Methods

Fifty-nine subjects (22 healthy, 7 anemics with no renal disease, 30 subjects with CKD) participated. Data using 3D volume imaging, blood oxygenation level dependent (BOLD) and Diffusion MRI was acquired. BOLD MRI acquisition was repeated after 20 mg of iv furosemide.

Results

Compared to healthy subjects, those with CKD have lower renal parenchymal volumes (329.6±66.4 vs. 257.1±87.0 ml, p<0.005), higher cortical R2* values (19.7±3.2 vs. 23.2±6.3 s⁻¹, p = 0.013) (suggesting higher levels of hypoxia) and lower response to furosemide on medullary R2* (6.9±3.3 vs. 3.1±7.5 s⁻¹, p = 0.02). All three parameters showed significant correlation with estimated glomerular filtration rate (eGFR). When the groups were matched for age and sex, cortical R2* and kidney volume still showed significant differences between CKD and healthy controls. The most interesting observation is that a small number of subjects (8 of 29) contributed to the increase in mean value observed in CKD. The difference in cortical R2* between these subjects compared to the rest were highly significant and had a large effect size (Cohen’s d = 3.5). While highly suggestive, future studies may be necessary to verify if such higher levels of hypoxia are indicative of progressive disease. Diffusion MRI showed no differences between CKD and healthy controls.

Conclusions

These data demonstrate that BOLD MRI can be used to identify enhanced hypoxia associated with CKD and the preliminary observations are consistent with the chronic hypoxia model for disease progression in CKD. Longitudinal studies are warranted to further verify these findings and assess their predictive value.