Dataset Information

An Artificial Intelligence Generated Automated Algorithm to Measure Total Kidney Volume in ADPKD.

ABSTRACT:

Introduction

Accurate tools to inform individual prognosis in patients with autosomal dominant polycystic kidney disease (ADPKD) are lacking. Here, we report an artificial intelligence (AI)-generated method for routinely measuring total kidney volume (TKV).

Methods

An ensemble U-net algorithm was created using the nnUNet approach. The training and internal cross-validation cohort consisted of all 1.5T magnetic resonance imaging (MRI) data acquired using 5 different MRI scanners (454 kidneys, 227 scans) in the CYSTic consortium, which was first manually segmented by a single human operator. As an independent validation cohort, we utilized 48 sequential clinical MRI scans with reference results of manual segmentation acquired by 6 individual analysts at a single center. The tool was then implemented for clinical use and its performance analyzed.

Results

The training or internal validation cohort was younger (mean age 44.0 vs. 51.5 years) and the female-to-male ratio higher (1.2 vs. 0.94) compared to the clinical validation cohort. The majority of CYSTic patients had PKD1 mutations (79%) and typical disease (Mayo Imaging class 1, 86%). The median DICE score on the clinical validation data set between the algorithm and human analysts was 0.96 for left and right kidneys with a median TKV error of -1.8%. The time taken to manually segment kidneys in the CYSTic data set was 56 (±28) minutes, whereas manual corrections of the algorithm output took 8.5 (±9.2) minutes per scan.

Conclusion

Our AI-based algorithm demonstrates performance comparable to manual segmentation. Its rapidity and precision in real-world clinical cases demonstrate its suitability for clinical application.

SUBMITTER: Taylor J

PROVIDER: S-EPMC10851006 | biostudies-literature | 2024 Feb

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

An Artificial Intelligence Generated Automated Algorithm to Measure Total Kidney Volume in ADPKD.

Taylor Jonathan J Thomas Richard R Metherall Peter P van Gastel Marieke M Cornec-Le Gall Emilie E Caroli Anna A Furlano Monica M Demoulin Nathalie N Devuyst Olivier O Winterbottom Jean J Torra Roser R Perico Norberto N Le Meur Yannick Y Schoenherr Sebastian S Forer Lukas L Gansevoort Ron T RT Simms Roslyn J RJ Ong Albert C M ACM

Kidney international reports 20231104 2

<h4>Introduction</h4>Accurate tools to inform individual prognosis in patients with autosomal dominant polycystic kidney disease (ADPKD) are lacking. Here, we report an artificial intelligence (AI)-generated method for routinely measuring total kidney volume (TKV).<h4>Methods</h4>An ensemble U-net algorithm was created using the nnUNet approach. The training and internal cross-validation cohort consisted of all 1.5T magnetic resonance imaging (MRI) data acquired using 5 different MRI scanners (4 ...[more]

PMID: 38344736

Similar Datasets

Project description:ObjectivesTo develop a high-performance, rapid semi-automated method (Sheffield TKV Tool) for measuring total kidney volume (TKV) from magnetic resonance images (MRI) in patients with autosomal dominant polycystic kidney disease (ADPKD).MethodsTKV was initially measured in 61 patients with ADPKD using the Sheffield TKV Tool and its performance compared to manual segmentation and other published methods (ellipsoidal, mid-slice, MIROS). It was then validated using an external dataset of MRI scans from 65 patients with ADPKD.ResultsSixty-one patients (mean age 45 ± 14 years, baseline eGFR 76 ± 32 ml/min/1.73 m2) with ADPKD had a wide range of TKV (258-3680 ml) measured manually. The Sheffield TKV Tool was highly accurate (mean volume error 0.5 ± 5.3% for right kidney, - 0.7 ± 5.5% for left kidney), reproducible (intra-operator variability - 0.2 ± 1.3%; inter-operator variability 1.1 ± 2.9%) and outperformed published methods. It took less than 6 min to execute and performed consistently with high accuracy in an external MRI dataset of T2-weighted sequences with TKV acquired using three different scanners and measured using a different segmentation methodology (mean volume error was 3.45 ± 3.96%, n = 65).ConclusionsThe Sheffield TKV Tool is operator friendly, requiring minimal user interaction to rapidly, accurately and reproducibly measure TKV in this, the largest reported unselected European patient cohort with ADPKD. It is more accurate than estimating equations and its accuracy is maintained at larger kidney volumes than previously reported with other semi-automated methods. It is free to use, can run as an independent executable and will accelerate the application of TKV as a prognostic biomarker for ADPKD into clinical practice.Key points• This new semi-automated method (Sheffield TKV Tool) to measure total kidney volume (TKV) will facilitate the routine clinical assessment of patients with ADPKD. • Measuring TKV manually is time consuming and laborious. • TKV is a prognostic indicator in ADPKD and the only imaging biomarker approved by the FDA and EMA.

Project description:Background and objectivesTotal kidney volume is a validated prognostic biomarker for autosomal dominant polycystic kidney disease. Total kidney volume by magnetic resonance imaging (MRI) and manual segmentation is considered the "reference standard," but it is time consuming and not readily accessible. By contrast, three-dimensional (3D) ultrasound provides a promising technology for total kidney volume measurements with unknown potential. Here, we report a comparative study of total kidney volume measurements by 3D ultrasound versus the conventional methods by ultrasound ellipsoid and MRI ellipsoid.Design, setting, participants, & measurementsThis single-center prospective study included 142 patients who completed a standardized 3D ultrasound and MRI. Total kidney volumes by 3D ultrasound and ultrasound ellipsoid were compared with those by MRI. We assessed the agreement of total kidney volume measurements by Bland-Altman plots and misclassification of the Mayo Clinic imaging classes between the different imaging methods, and we assessed prediction of Mayo Clinic imaging classes 1C-1E by average ultrasound kidney length >16.5 cm.ResultsCompared with MRI manual segmentation, MRI ellipsoid, 3D ultrasound, and ultrasound ellipsoid underestimated total kidney volume (mean difference: -3%, -9%, and -11%, respectively), with Mayo Clinic imaging classes misclassified in 11%, 21%, and 22% of patients, respectively; most misclassified cases by MRI ellipsoid (11 of 16), 3D ultrasound (23 of 30), and ultrasound ellipsoid (26 of 31) were placed into a lower Mayo Clinic imaging class. Predictions of the high-risk Mayo Clinic imaging classes (1C-1E) by MRI ellipsoid, 3D ultrasound, and ultrasound ellipsoid all yielded high positive predictive value (96%, 95%, and 98%, respectively) and specificity (96%, 96%, and 99%, respectively). However, both negative predictive value (90%, 88%, and 95%, respectively) and sensitivity (88%, 85%, and 94%, respectively) were lower for 3D ultrasound and ultrasound ellipsoid compared with MRI ellipsoid. An average ultrasound kidney length >16.5 cm was highly predictive of Mayo Clinic imaging classes 1C-1E only in patients aged ≤45 years.ConclusionsTotal kidney volume measurements in autosomal dominant polycystic kidney disease by 3D ultrasound and ultrasound ellipsoid displayed similar bias and variability and are less accurate than MRI ellipsoid. Prediction of high-risk Mayo Clinic imaging classes (1C-1E) by all three methods provides high positive predictive value, but ultrasound ellipsoid is simpler to use and more readily available.

Project description:Background and objectivesThis study aimed to compare Total kidney volume (TKV) measurements using US-ellipsoid (US-EL) and MRI-ellipsoid (MRI-EL) in patients with autosomal-dominant-polycystic-kidney-disease (ADPKD). It also evaluated whether the agreement between right (RKV) and left (LKV) kidney volume measurements differed.MethodsRetrospective analysis of a prospective data-base that included consecutive patients diagnosed with ADPKD. Total kidney volumes by 3D-US-EL were compared with those by MRI-EL. Bland-Altman-plots, Passing-Bablok-regression, and the concordance-correlation-coefficient (CCC) were used to compare right (RKV), left (LKV), and TKV measurements.ResultsThirty-two ADPKD patients, 14(43.7%) women, were included. Mean measured (mGFR) and estimated (eGFR) glomerular-filtration-rate (GFR) were 86.5 ± 23.9 mL/min and 78.9 ± 23.6 mL/min, respectively. Compared with MRI-EL, TKV (Mean difference: - 85.9 ± 825.6 mL; 95%CI - 498.5 to 326.7 mL; p = 0.6787), RKV (Mean difference: - 58.5 ± 507.7 mL; 95%CI - 312.2 to 195.2 mL; p = 0.6466), and LKV (Mean difference: - 27.4 ± 413.5 mL; 95%CI - 234.1 to 179.2 mL; p = 0.7918) were lower with US-EL than with MRI-EL, although without significant differences. According to Passing and Bablok-regression analysis, the Spearman correlation-coefficient was 0.96 (95%CI 0.92 to 0.98); 0.91 (95%CI 0.82 to 0.96), and 0.94 (95%CI 0.87 to 0.97) in the RKV, LKV, and TKV, respectively; p < 0.0001 each, respectively. CCC of RKV, LKV, and TKV measurements were 0.95, 0.89, and 0.94, respectively. The mGFR and eGFR showed statistically significant negative correlations with TKV measured by both MRI-EL (p = 0.0281 and p = 0.0054, respectively) and US-EL (p = p = 0.0332 and p = 0.0040, respectively).ConclusionsThis study found that ultrasound-based ellipsoid kidney volume measurements strongly correlated with MRI-based measurements, suggesting that ultrasound is a reliable, accessible alternative for assessing kidney volume, particularly when MRI is unavailable.

Project description:Background/Objectives: Although tolvaptan efficacy in ADPKD has been demonstrated in randomized clinical trials, there is no definitive method for assessing its efficacy in the individual patient in the clinical setting. In this exploratory feasibility study, we report a method to quantify the change in total kidney volume (TKV) growth rate to retrospectively evaluate tolvaptan efficacy for individual patients. Treatment-related changes in estimated glomerular filtration rate (eGFR) are also assessed. Methods: MRI scans covering at least 1 year prior to and during treatment with tolvaptan were performed, with deep learning facilitated kidney segmentation and fitting multiple imaging timepoints to exponential growth in 32 ADPKD patients. Clustering analysis differentiated tolvaptan treatment "responders" and "non-responders" based upon the magnitude of change in TKV growth rate. Differences in rate of eGFR decline, urine osmolality, and other parameters were compared between responders and non-responders. Results: Eighteen (56%) tolvaptan responders (mean age 42 ± 8 years) were identified by k-means clustering, with an absolute reduction in annual TKV growth rate of >2% (mean = -5.1% ± 2.5% per year). Thirteen (44%) non-responders were identified, with <1% absolute reduction in annual TKV growth rate (mean = +2.4% ± 2.7% per year) during tolvaptan treatment. Compared to non-responders, tolvaptan responders had significantly higher mean TKV growth rates prior to tolvaptan treatment (7.1% ± 3.6% per year vs. 3.7% ± 2.4% per year; p = 0.003) and higher median pretreatment spot urine osmolality (Uosm, 393 mOsm/kg vs. 194 mOsm/kg, p = 0.03), confirmed by multivariate analysis. Mean annual rate of eGFR decline was less in responders than in non-responders (-0.25 ± 0.04, CI: [-0.27, -0.23] mL/min/1.73 m2 per year vs. -0.40 ± 0.06, CI: [-0.43, -0.37] mL/min/1.73 m2 per year, p = 0.036). Conclusions: In this feasibility study designed to assess predictors of tolvaptan treatment efficacy in individual patients with ADPKD, we found that high pretreatment levels of annual TKV growth rate and higher pretreatment spot urine osmolality were associated with a responder phenotype.

Project description:Background and objectiveHydronephrosis is essential in the diagnosis of renal colic. We automated the detection of hydronephrosis from ultrasound images to standardize the therapy and reduce the misdiagnosis of renal colic.MethodsAnonymously collected ultrasound images of human kidneys, both normal and hydronephrotic, were preprocessed for neural networks. Six "state of the art" models were trained and cross-validated for the detection of hydronephrosis, and two convolutional networks were used for kidney segmentation. In the testing phase, performance metrics included true positives, true negatives, false positives, false negatives, accuracy, and F1 score, while the evaluation of the segmentation task involved accuracy, precision, dice, jaccard, recall, and ASSD.Key findings and limitationsA total of 523 sonographic kidney images (423 nonhydronephrotic and 100 hydronephrotic) were collected from three different ultrasound devices. After training on this dataset, all models were used to evaluate 200 new ultrasound kidney images (142 nonhydronephrotic and 58 hydronephrotic kidneys). The highest validation accuracy (98.5%) was achieved by the AlexNet model (GoogLeNet 97%, AlexNet_v2 96%, ResNet50 96%, ResNet101 97.5%, and ResNet152 95%). The deeplabv3_resnet50 and deeplabv3_resnet101 reached a dice coefficient of 94.74% and 94.48%, respectively, on the task of automated kidney segmentation. The study is limited by analyzing only hydronephrosis, but this specific focus enabled high detection accuracy.Conclusions and clinical implicationsWe show that our automated ultrasound deep learning model can be trained and used to interpret and segmentate ultrasound images from different sources with high accuracy. This method will serve as an automated tool in the diagnostic algorithm of acute renal failure in the future.Patient summaryHydronephrosis is crucial in the diagnosis of renal colic. Recent advances in artificial intelligence allow automated detection of hydronephrosis in ultrasound images with high accuracy. These methods will help standardize the diagnosis and treatment renal colic.

Dataset Information

An Artificial Intelligence Generated Automated Algorithm to Measure Total Kidney Volume in ADPKD.

Introduction

Methods

Results

Conclusion

Publications

An Artificial Intelligence Generated Automated Algorithm to Measure Total Kidney Volume in ADPKD.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets