Automated renal cortical volume measurement for assessment of renal function in patients undergoing radical nephrectomy

Using CT images to assist the segmentation of MR images via generalization: Segmentation of the renal parenchyma of renal carcinoma patients

BACKGROUND

Developing deep learning models for segmenting medical images in multiple modalities with less data and annotation is an attractive and challenging task, which was previously discussed as being accomplished by complex external frameworks for bridging the gap between different modalities. Exploring the generalization ability of networks in medical images in different modalities could provide more simple and accessible methods, yet comprehensive testing could still be needed.

PURPOSE

To explore the feasibility and robustness of using computed tomography (CT) images to assist the segmentation of magnetic resonance (MR) images via the generalization, in the segmentation of renal parenchyma of renal cell carcinoma (RCC) patients.

METHODS

Nephrographic CT images and fat-suppressed T2-weighted (fs-T2 W) images were retrospectively collected. The pure CT dataset included 116 CT images. Additionally, 240 MR images were randomly divided into subsets A and B. From subset A, three training datasets were constructed, each containing 40, 80, and 120 images, respectively. Similarly, three datasets were constructed from subset B. Subsequently, datasets with mixed modality were created by combining these pure MR datasets with the 116 CT images. The 3D-UNET models for segmenting the renal parenchyma in two steps were trained using these 13 datasets: segmenting kidneys and then the renal parenchyma. These models were evaluated in internal MR (n = 120), CT (n = 65) validation datasets, and an external validation dataset of CT (n = 79), using the mean of the dice similarity coefficient (DSC). To demonstrate the robustness of generalization ability over different proportions of modalities, we compared the models trained with mixed modality in three different proportions and pure MR, using repeated measures analysis of variance (RM-ANOVA). We developed a renal parenchyma volume quantification tool by the trained models. The mean differences and Pearson correlation coefficients between the model segmentation volume and the ground truth segmentation volume were calculated for its evaluation.

RESULTS

The mean DSCs of models trained with 116 data in CT in the validation of MR were 0.826, 0.842, and 0.953, respectively, for the predictions of kidney segmentation model on whole image, renal parenchymal segmentation model on kidneys with RCC and without RCC. For all models trained with mixed modality, the means of DSC were above 0.9, in all validations of CT and MR. According to the results of the comparison between models trained with mixed modality and pure MR, the means of DSC of the former were significantly greater or equal to the latter, at all three different proportions of modalities. The differences of volumes were all significantly lower than one-third of the volumetric quantification error of a previous method, and the Pearson correlation coefficients of volumes were all above 0.96 on kidneys with and without RCC of three validations.

CONCLUSION

CT images could be used to assist the segmentation of MR images via the generalization, with or without the supervision of MR data. This ability showed acceptable robustness. A tool for accurately measuring renal parenchymal volume on CT and MR images was established.

Studying the effect of donor kidney volume ratios to recipients' body surface area, body mass index, and total body weight on post-transplant graft function

OBJECTIVES

The goal of this study was to retrospectively investigate the relationship between renal transplanted volume indexes (Total kidney volume (TKV)/Body surface area (BSA), Renal parenchymal volume (RPV)/BSA, Renal cortical volume (RCV)/BSA, RPV/Body mass index (BMI), RCV/BMI, RPV/Weight, RCV/Weight), and short- and long-term function of the graft.

METHODS

One-hundred and twelve live donor-recipient pairs from 2017 to 2018, whose donors underwent preoperative renal computed tomography angiography and recipients survived during 12 months of follow-up, were included in this study.

RESULTS

The crude and adjusted linear regressions for the effect of volume measurements by voxel and ellipsoid methods on the estimated glomerular filtration rate (eGFR) at different post-transplantation times demonstrated that the RPV/weight ratio had the most substantial crude effect on the eGFR 12 months and 4 years after renal transplant. Receiver operating characteristic (ROC) curves for six different renal volume ratios demonstrated no significant difference between these ratios in terms of discriminative ability (p value < 0.05). A strong direct correlation between TKV calculated by the ellipsoid formula with RPV and RCV measured using OsiriX software was noted. Analysis of ROC curves for renal volume indices has demonstrated fair to good discriminative ability of our cut-off points to estimate 4-year post-transplantation eGFR > 60 mL/min.

CONCLUSION

Renal transplant recipients' volume indices, such as RPV/weight, had strong correlations with eGFR at different points in time, and renal transplant recipients with the volume ratios higher than our cut-off points had a good chance of having a 4-year post-transplantation eGFR higher than 60 mL/min.

A novel rotation method with variable-angle anterior probe for correcting the depth of the kidney to optimize renal dynamic imaging

Purpose

We aimed to investigate the effect and significance of the rotation method with variable-angle anterior probe corrected for the depth of two kidneys on the determination of glomerular filtration rate (GFR) in total and single kidneys by the renal dynamic imaging Gates method.

Methods

Seventy-two patients who underwent dynamic renal imaging by the rotation method and abdominal CT in our hospital were collected in the present study. CT scanning, rotation method, Tonnesen's formula, and Li-Qian's formula were compared in terms of the depth of two kidneys, the depth difference between the two kidneys, and the total renal and single GFR obtained by substituting the renal depth values into Gates' formula.

Results

①The depth of kidneys and GFR: Compared to CT, Tonnesen's formula significantly underestimated the depth of both kidneys and the total and single renal GFR (P < 0.05). No significant differences were found in the depth of both kidneys and the total and single renal GFR between Li-Qian's formula and the rotation method (P > 0.05), with a strong agreement and with the least bias in the values measured by the rotation method. ②Renal depth difference: Compared to CT, Tonnesen's formula and Li-Qian's formula underestimated the difference in depth between the two kidneys (P < 0.05). None of the differences were statistically significant based on the rotation method (P > 0.05). The depth difference was positively correlated with the resulting changes in single renal function (|R(CT)-R(Li-Qian)|) and (|R(Rotation)-R(Li-Qian)|) (r = 0.881, 0.641, P < 0.001). As the depth difference increased, Li-Qian's formula could not visualize changes in single renal function accurately. In contrast, the accuracy of the rotation method in assessing single renal function remains unaffected.

Conclusion

The rotation method obtains an accurate depth and depth difference between the two kidneys without additional CT radiation, enhancing the accuracy of the Gates method for determining total and single renal GFR.

Trial registration Medical Ethics Committee of First Hospital of Shanxi Medical University, 2021BAL0146. Registered 12 January 2021.

Supplementary Information

The online version contains supplementary material available at 10.1186/s40658-022-00511-w.

Application of three-dimensional image reconstruction technology based on high-resolution CT in pyeloplasty

Background

Three-dimensional (3D) image reconstruction technology is widely used in surgical operations for its intuitive visualization. Pyeloplasty requiresprecise cutting and suturing. The reconstruction technology can accurately determine the location and scope of the stenosis at the junction of the renal pelvis and ureter and the relationship with the surrounding vasculature. The purpose of this article is to retrospective evaluate the application value of image reconstruction technology in pyeloplasty based on high-resolution 3D CT images.

Methods

A total of 20 patients with renal pelvic ureteral junction obstruction admitted to our hospital from August 2019 to August 2020 were selected. In this group, left pyeloplasty was performed in 8 patients and right pyeloplasty in 12 patients. In terms of conditions, there was 1 case with secondary pyeloplasty, 6 cases of patients with kidney stones, 2 cases with renal ectopic blood vessels, 1 case with renal prolapse, 1 case with horseshoe kidney, and 1 case with ureteral polyps. There were 12 males and 8 females, with an average age of 34.65±10.67 years and an average body mass index (BMI) of 22.48±3.03 kg/m². In all patients, 3D CT reconstruction technology was used to guide the formulation of robot-assisted laparoscopic pyeloplasty plans; verify the consistency between the actual operation and the preoperative planning; and observe the operation time, blood loss, postoperative exhaust time, indwelling drainage tube time, and follow-up for comorbidities.

Results

The operation was successful in all 20 patients. The actual operation was 100% consistent with the preoperative planning, the operative time was 160.80±63.26 min, the intraoperative blood loss was 47±30.45 mL, the postoperative exhaust time was 1.15±0.37 days, the drainage tube indwelling time was 4.35±1.50 days, and the average follow-up time was 7.95±3.41 months. There were no complications.

Conclusions

Three-dimensional image reconstruction technology based on high-resolution CT has high clinical application value in the treatment of ureteropelvic junction obstruction (UPJO), which simplifies the operation process and shortens the operation time, and is a valuable tool for auxiliary surgeons in devising the operation plan.

Renal function after partial nephrectomy following intra-arterial embolization of renal tumors

Laparoscopic Partial Nephrectomy (LPN) after intra-arterial Embolization of renal tumors (LPNE) in a hybrid operating room allows renal tumor enucleation without dissection and clamping of the renal pedicle. The purpose was to assess the potential negative impact of embolization on the renal function. This prospective monocentric study included all patients treated with LPNE between May 2015 and June 2019. Clinical data was collected and incorporated into the UroCCR database (NCT03293563). Glomerular Filtration Rate (GFR) and Computed Tomography Renal Volume (CTRV) were compared before and after 6 months following LPNE. The mean post-operative GFR was 86.6 mL/min (SD 22.9). The mean GFR loss was 9.4% (SD 15.1) and the median renal parenchyma loss was 21 mL (SD 20.6). Using a threshold of 25% GFR loss, age was the only significant predictive factor of renal function impairment according to bivariate (59.5 vs 69.3 years, p = 0.017) and multivariable analysis (OR 1.075, CI 1–1.2], p = 0.05). Significant renal function impairment was not correlated with the renal parenchymal volume loss (OR 0.987, CI [0.95–1.02], p = 0.435). Renal function impairment after LPNE seems to be comparable to other techniques of partial nephrectomy.

Semiautomated Renal Cortex Volumetry in Multislice Computed Tomography: Effect of Slice Thickness and Iterative Reconstruction Algorithms

OBJECTIVE

The aim of the study was to evaluate the effect of slice thickness, iterative reconstruction (IR) algorithm, and kernel selection on measurement accuracy and interobserver variability for semiautomated renal cortex volumetry (RCV) with multislice computed tomography (CT).

METHODS

Ten patients (62.4 ± 17.2 years) undergoing abdominal biphasic multislice computed tomography were enrolled in this retrospective study. Computed tomography data sets were reconstructed at 1-, 2-, and 5-mm slice thickness with 2 different IR algorithms (iDose, IMRST) and 2 different kernels (IMRS and IMRR) (Philips, the Netherlands). Two readers independently performed semiautomated RCV for each reconstructed data set to calculate left kidney volume (LKV) and split renal function (SRF). Statistics were calculated using analysis of variance with Geisser-Greenhouse correction, followed by Tukey multiple comparisons post hoc test. Statistical significance was defined as P ≤ 0.05.

RESULTS

Semiautomated RCV of 120 data sets (240 kidneys) was successfully performed by both readers. Semiautomated RCV provides comparable results for LKV and SRF with 3 different slice thicknesses, 2 different IR algorithms, and 2 different kernels. Only the 1-mm slice thickness showed significant differences for LKV between IMRR and IMRS (P = 0.02, mean difference = 4.28 bb) and IMRST versus IMRS (P = 0.02, mean difference = 4.68 cm) for reader 2. Interobserver variability was low between both readers irrespective of slice thickness and reconstruction algorithm (0.82 ≥ P ≥ 0.99).

CONCLUSIONS

Semiautomated RCV measurements of LKV and SRF are independent of slice thickness, IR algorithm, and kernel selection. These findings suggest that comparisons between studies using different slice thicknesses and reconstruction algorithms for RCV are valid.

Preserved Kidney Volume, Body Mass Index, and Age Are Significant Preoperative Factors for Predicting Estimated Glomerular Filtration Rate in Living Kidney Donors at 1 Year After Donation

BACKGROUND

Securing postdonation renal function in the lifetime of donors is a consequential subject for physicians, and precise prediction of postdonation renal function would be considerably beneficial when judging the feasibility of kidney donation. The aim of this study was to investigate the optimum model for predicting eGFR at 1 year after kidney donation.

METHODS

We enrolled 101 living-related kidney donors for the development cohort and 44 for the external validation cohort. All patients in each cohort underwent thin-sliced (1 mm) enhanced computed tomography (CT) scans. We excluded individuals with diabetes, glucose intolerance, or albuminuria from this study. We evaluated preoperative factors including age, sex, hypertension, body mass index (BMI), serum uric acid, baseline eGFR, and body surface area (BSA)-adjusted preserved kidney volume (PKV) by using 3-dimensional reconstruction of thin-sliced enhanced CT images. To detect independent predictors, we performed multivariable regression analysis.

RESULTS

The multivariable regression analysis revealed that age, BMI, predonation eGFR, and BSA-adjusted PKV were independent predictors of eGFR at 1 year after kidney donation (correlation coefficient: -0.15, -0.476, 0.521, 0.127, respectively). A strong correlation between predicted eGFR and observed eGFR was obtained in the development cohort (r = 0.839, P < .0001). The significance of this predictive model was also confirmed with the external validation cohort (r = 0.797, P < .0001).

CONCLUSIONS

Age, BMI, predonation eGFR, and BSA-adjusted PKV may be useful for precisely predicting eGFR at 1 year after living kidney donation and be helpful to determine the feasibility of kidney donation from marginal donors.

Estimation of split renal function using different volumetric methods: inter- and intraindividual comparison between MRI and CT

PURPOSE

This study aims to determine whether contrast-enhanced (CE)-magnetic resonance imaging (MRI) is comparable to CE-computed tomography (CT) for estimation of split renal function (SRF). For this purpose, two different kidney volumetry methods, the renal cortex volumetry (RCV) and modified ellipsoid volume (MELV), are compared for both acquisition types (CT vs. MRI) with regard to accuracy and reliability, subsequently referred to as RCV_CT/RCV_MRI and MELV_CT/MELV_MRI.

METHODS

This retrospective study included 29 patients (18 men and 11 women; mean age 62.8 ± 12.4 years) who underwent CE-MRI and CE-CT of the abdomen within a period of 3 months. Two independent readers (R1/R2) performed RCV and MELV in all datasets with corresponding semiautomated software tools. RCV was performed with datasets in the arterial phase and MELV in the venous phase. Statistics were calculated using one-way ANOVA, two-tailed Student's t test, Pearson´s correlation, and Bland-Altman plots with p ≤ 0.05 being considered statistically significant.

RESULTS

In all datasets, SRF was almost identical for both volumetry methods with a mean difference of < 1%. Bland-Altman analysis comparing RCV in CT and MRI showed very good agreement for R1/R2. Interreader agreement was strong for RCV_CT and good for RCV_MRI (r = 0.89; r = 0.69). MELV_CT/MRI interreader agreement was only moderate (r = 0.54; r = 0.50) with a high range of values. Intrareader agreement was excellent for all measurements, except MELV_MRI which showed a high mean bias and range of values (RCV_CT: r = 0.93, RCV_MRI: r = 0.98, MELV_CT: r = 0.89, MELV_MRI: r = 0.54).

CONCLUSION

Renal volumetric estimates of SRF are almost as accurate and reliable with CE-MRI as with CE-CT using RCV method. In distinction, the calculation of SRF using MELV was inferior to RCV with respect to accuracy and reliability. Thus, RCV method is recommended to estimate SRF, primarily using CT datasets. However, RCV with MRI datasets for kidney volumetry allows for comparable accuracy and reliability while sparing patients and healthy donors of unnecessary radiation exposure.

The 3-D Volumetric Measurement Including Resected Specimen for Predicting Renal Function AfterRobot-assisted Partial Nephrectomy

OBJECTIVE

To investigate the relationship between postoperative renal function and resected cortex margin volume calculated by a 3-dimensional reconstruction technique based on the resected specimen, and to determine predictors of renal function after robot-assisted partial nephrectomy.

METHODS

A total of 114 patients underwent robot-assisted partial nephrectomy from 2014 to 2018. Patients without a 1 mm slice computed tomography or renal scintigraphy were excluded. We identified the margins of the tumor from each resected specimen with 2 mm margin being added as the ischemic margin. The volume of the renal cortex was calculated automatically using 3-dimensional volume analyzer software. The total margin volume was excluded from the ipsilateral cortex volume to calculate the cortex volume split. Predicted estimated glomerular filtration rate (eGFR) was calculated using the change in cortex volume and then compared with the actual eGFR.

RESULTS

Eighty-two patients were included in this retrospective study. Sixty-six patients (80%) were cT1a. A strong correlation was observed between renal scintigraphy split and pre- and postoperative cortex volume split (Pearson correlation coefficient r = 0.9330 and 0.8742, respectively). The predicted eGFR correlated strongly with post 1, 3, 6, and 12 months eGFR (r = 0.8929, 0.9294, 0.9320, and 0.8952, respectively). Preoperative relative renal function and total cortex margin volume were independent risk factors for decreasing postoperative renal function.

CONCLUSION

This precise volumetric assessment that includes the resected margin is an alternative to renal scintigraphy for predicting postoperative relative renal function. The healthy cortex margin volume calculated by the reconstruction technique is an independent risk factor of decreasing postoperative renal function.

Accuracy of implementing principles of fusion imaging in the follow up and surveillance of complex aneurysm repair

Fusion imaging is standard for the endovascular treatment of complex aortic aneurysms, but its role in follow up has not been explored. A critical issue is renal function deterioration over time. Renal volume has been used as a marker of renal impairment; however, it is not reproducible and remains a complex and resource-intensive procedure. The aim of this study is to determine the accuracy of a fusion-based software to automatically calculate the renal volume changes during follow up. In this study, computerized tomography (CT) scans of 16 patients who underwent complex aortic endovascular repair were analysed. Preoperative, 1-month and 1-year follow-up CT scans have been analysed using a conventional approach of semi-automatic segmentation, and a second approach with automatic segmentation. For each kidney and at each time point the percentage of change in renal volume was calculated using both techniques. After review, volume assessment was feasible for all CT scans. For the left kidney, the intraclass correlation coefficient (ICC) was 0.794 and 0.877 at 1 month and 1 year, respectively. For the right side, the ICC was 0.817 at 1 month and 0.966 at 1 year. The automated technique reliably detected a decrease in renal volume for the eight patients with occluded renal arteries during follow up. This is the first report of a fusion-based algorithm to detect changes in renal volume during postoperative surveillance using an automated process. Using this technique, the standardized assessment of renal volume could be implemented with greater ease and reproducibility and serve as a warning of potential renal impairment.

Renal cortical volume: High correlation with pre- and post-operative renal function in living kidney donors

BACKGROUND

CT volumetry has previously been proposed as an alternative to scintigraphy for the evaluation of pre-donation split renal function and the prediction of post-donation renal function in living kidney donors. The aim of our study was to retrospectively assess the relevance of three CT volumetry techniques for estimating pre-donation kidney function and predicting the risk for chronic kidney disease (CKD) at 1-year post-nephrectomy in a French cohort of living donors using isotopic measures of kidney function.

METHODS

Kidney volume was quantified pre-donation for 105 donors using three methods total parenchymal three-dimensional renal volume (3DRV), total parenchymal renal volume contouring (RVCt), and renal cortical volume (RCoV). Subjects also had a 51Cr-EDTA scintigraphy to measure glomerular filtration rate (mGFR) pre-donation and 1-year after donation. For each volume, we tested for association with mGFR using univariate regression models, and computed receiver operating characteristics analyses to assess their predictive potential of post-donation CKD.

RESULTS

Our population was composed of healthy subjects, who were predominantly female (69%) with a median age at donation of 51yo. Median mGFR was 102 mL/min/1.73 m² at pre-donation and 66 mL/min/1.73 m² 1-year after nephrectomy. The pre-donation median volume of the preserved kidney was 156 cm³, 163 cm³ and 99 cm³ for the 3DRV, RVCt and RCoV methods respectively, with a high correlation observed between each technique (R > 0.84). For all methods, total kidney volume was significantly associated with pre-donation mGFR (P < 0.001). Preserved kidney volume was also strongly correlated with post-donation mGFR (P < 0.0001), with the strongest correlation observed for RCoV (R = 0.60 vs. R = 0.39 and R = 0.51 for 3DRV and RVCt, respectively). Finally, the RCoV method yielded the best predictive value of 1-year post-donation CKD (AUC = 0.80 vs. AUC = 0.76 and 0.70 for RVCt and 3DRV, respectively).

CONCLUSIONS

In our cohort of healthy donors with measured kidney function, cortical volumetry (RCoV) appears as the best volumetric technique to use as a surrogate to scintigraphy for estimating pre-donation split renal function and predicting post-donation renal outcomes.