Diffusion-weighted MR imaging of transplanted kidneys: Preliminary report

A DEEP LEARNING-BASED CAD SYSTEM FOR RENAL ALLOGRAFT ASSESSMENT: DIFFUSION, BOLD, AND CLINICAL BIOMARKERS

Recently, studies for non-invasive renal transplant evaluation have been explored to control allograft rejection. In this paper, a computer-aided diagnostic system has been developed to accommodate with an early-stage renal transplant status assessment, called RT-CAD. Our model of this system integrated multiple sources for a more accurate diagnosis: two image-based sources and two clinical-based sources. The image-based sources included apparent diffusion coefficients (ADCs) and the amount of deoxygenated hemoglobin (R2*). More specifically, these ADCs were extracted from 47 diffusion weighted magnetic resonance imaging (DW-MRI) scans at 11 different b-values (b0, b50, b100, …, b1000 s/mm²), while the R2* values were extracted from 30 blood oxygen level-dependent MRI (BOLD-MRI) scans at 5 different echo times (2ms, 7ms, 12ms, 17ms, and 22ms). The clinical sources included serum creatinine (SCr) and creatinine clearance (CrCl). First, the kidney was segmented through the RT-CAD system using a geometric deformable model called a level-set method. Second, both ADCs and R2* were estimated for common patients (N = 30) and then were integrated with the corresponding SCr and CrCl. Last, these integrated biomarkers were considered the discriminatory features to be used as trainers and testers for future deep learning-based classifiers such as stacked auto-encoders (SAEs). We used a k-fold cross-validation criteria to evaluate the RT-CAD system diagnostic performance, which achieved the following scores: 93.3%, 90.0%, and 95.0% in terms of accuracy, sensitivity, and specificity in differentiating between acute renal rejection (AR) and non-rejection (NR). The reliability and completeness of the RT-CAD system was further accepted by the area under the curve score of 0.92. The conclusions ensured that the presented RT-CAD system has a high reliability to diagnose the status of the renal transplant in a non-invasive way.

A multimodal computer-aided diagnostic system for precise identification of renal allograft rejection: Preliminary results

PURPOSE

Early assessment of renal allograft function post-transplantation is crucial to minimize and control allograft rejection. Biopsy - the gold standard - is used only as a last resort due to its invasiveness, high cost, adverse events (e.g., bleeding, infection, etc.), and the time for reporting. To overcome these limitations, a renal computer-assisted diagnostic (Renal-CAD) system was developed to assess kidney transplant function.

METHODS

The developed Renal-CAD system integrates data collected from two image-based sources and two clinical-based sources to assess renal transplant function. The imaging sources were the apparent diffusion coefficients (ADCs) extracted from 47 diffusion-weighted magnetic resonance imaging (DW-MRI) scans at 11 different b-values (b0, b50, b100, ..., b1000 s/mm 2 ), and the transverse relaxation rate (R2*) extracted from 30 blood oxygen level-dependent MRI (BOLD-MRI) scans at 5 different echo times (TEs = 2, 7, 12, 17, and 22 ms). Serum creatinine (SCr) and creatinine clearance (CrCl) were the clinical sources for kidney function evaluation. The Renal-CAD system initially performed kidney segmentation using the level-set method, followed by estimation of the ADCs from DW-MRIs and the R2* from BOLD-MRIs. ADCs and R2* estimates from 30 subjects that have both types of scans were integrated with their associated SCr and CrCl. The integrated biomarkers were then used as our discriminatory features to train and test a deep learning-based classifier, namely stacked autoencoders (SAEs) to differentiate non-rejection (NR) from acute rejection (AR) renal transplants.

RESULTS

Using a leave-one-subject-out cross-validation approach along with SAEs, the Renal-CAD system demonstrated 93.3% accuracy, 90.0% sensitivity, and 95.0% specificity in differentiating AR from NR. Robustness of the Renal-CAD system was also confirmed by the area under the curve value of 0.92. Using a stratified tenfold cross-validation approach, the Renal-CAD system demonstrated its reproducibility and robustness by a diagnostic accuracy of 86.7%, sensitivity of 80.0%, specificity of 90.0%, and AUC of 0.88.

CONCLUSION

The obtained results demonstrate the feasibility and efficacy of accurate, noninvasive identification of AR at an early stage using the Renal-CAD system.

EARLY ASSESSMENT OF RENAL TRANSPLANTS USING BOLD-MRI: PROMISING RESULTS

Non-invasive evaluation of renal transplant function is essential to minimize and manage renal rejection. A computer-assisted diagnostic (CAD) system was developed to evaluate kidney function post-transplantation. The developed CAD system utilizes the amount of blood-oxygenation extracted from 3D (2D + time) blood oxygen level-dependent magnetic resonance imaging (BOLD-MRI) to estimate renal function. BOLD-MRI scans were acquired at five different echo-times (2, 7, 12, 17, and 22) ms from 15 transplant patients. The developed CAD system first segments kidneys using the level-sets method followed by estimation of the amount of deoxyhemoglobin, also known as apparent relaxation rate (R2*). These R2* estimates were used as discriminatory features (global features (mean R2*) and local features (pixel-wise R2*)) to train and test state-of-the-art machine learning classifiers to differentiate between non-rejection (NR) and acute renal rejection. Using a leave-one-out cross-validation approach along with an artificial neural network (ANN) classifier, the CAD system demonstrated 93.3% accuracy, 100% sensitivity, and 90% specificity in distinguishing AR from non-rejection . These preliminary results demonstrate the efficacy of the CAD system to detect renal allograft status non-invasively.

A Novel CNN-Based CAD System for Early Assessment of Transplanted Kidney Dysfunction

This paper introduces a deep-learning based computer-aided diagnostic (CAD) system for the early detection of acute renal transplant rejection. For noninvasive detection of kidney rejection at an early stage, the proposed CAD system is based on the fusion of both imaging markers and clinical biomarkers. The former are derived from diffusion-weighted magnetic resonance imaging (DW-MRI) by estimating the apparent diffusion coefficients (ADC) representing the perfusion of the blood and the diffusion of the water inside the transplanted kidney. The clinical biomarkers, namely: creatinine clearance (CrCl) and serum plasma creatinine (SPCr), are integrated into the proposed CAD system as kidney functionality indexes to enhance its diagnostic performance. The ADC maps are estimated for a user-defined region of interest (ROI) that encompasses the whole kidney. The estimated ADCs are fused with the clinical biomarkers and the fused data is then used as an input to train and test a convolutional neural network (CNN) based classifier. The CAD system is tested on DW-MRI scans collected from 56 subjects from geographically diverse populations and different scanner types/image collection protocols. The overall accuracy of the proposed system is 92.9% with 93.3% sensitivity and 92.3% specificity in distinguishing non-rejected kidney transplants from rejected ones. These results demonstrate the potential of the proposed system for a reliable non-invasive diagnosis of renal transplant status for any DW-MRI scans, regardless of the geographical differences and/or imaging protocol.

Noninvasive evaluation of renal tissue oxygenation with blood oxygen level-dependent magnetic resonance imaging early after transplantation has a limited predictive value for the delayed graft function

PURPOSE

The aim of this study was to evaluate the feasibility of renal oxygenation assessment using blood oxygen level-dependent (BOLD) magnetic resonance imaging (MRI) in the early period after kidney transplantation and to estimate its prognostic value for delayed graft function.

MATERIAL AND METHODS

Examinations were performed in 50 subjects: 40 patients within a week after the kidney transplantation and 10 healthy controls, using T2*-weighted sequence. Measurements in transplant patients were correlated to basic laboratory parameters in the early period after transplantation and at follow-up.

RESULTS

Examinations of seven patients (18%) were rejected due to their poor technical quality. Mean R2* values in transplant recipients were lower than in controls (11.6 vs. 15.9 Hz; p = 0.0001). An R2* value of 0.28 Hz was calculated as the minimal detectable change. There was no relation between R2* values and laboratory parameters. However, patients eGFR ≥ 40 ml/min/1.73 m2 presented higher R2* values than recipients eGFR < 40 ml/min/1.73 m2 (12.0 vs. 11.1 Hz; p = 0.0189). In ROC analysis R2* of ≤ 11.7 predicted an early reduced graft function with 0.82 sensitivity and 56% specificity (AUC = 0.708; p = 0.024) but was not useful for delayed graft function prediction (p > 0.7).

CONCLUSIONS

Evaluation of renal graft oxygenation using BOLD MRI is technically challenging in the early period after transplantation. An R2* value of 0.28 Hz may in practice be considered as the minimal detectable change. The delayed graft function seems not to be dependent on early oxygenation values. Further, large-scale studies are necessary to confirm the latter observation.

3D kidney segmentation from abdominal diffusion MRI using an appearance-guided deformable boundary

A new technique for more accurate automatic segmentation of the kidney from its surrounding abdominal structures in diffusion-weighted magnetic resonance imaging (DW-MRI) is presented. This approach combines a new 3D probabilistic shape model of the kidney with a first-order appearance model and fourth-order spatial model of the diffusion-weighted signal intensity to guide the evolution of a 3D geometric deformable model. The probabilistic shape model was built from labeled training datasets to produce a spatially variant, independent random field of region labels. A Markov-Gibbs random field spatial model with up to fourth-order interactions was adequate to capture the inhomogeneity of renal tissues in the DW-MRI signal. A new analytical approach estimated the Gibbs potentials directly from the DW-MRI data to be segmented, in order that the segmentation procedure would be fully automatic. Finally, to better distinguish the kidney object from the surrounding tissues, marginal gray level distributions inside and outside of the deformable boundary were modeled with adaptive linear combinations of discrete Gaussians (first-order appearance model). The approach was tested on a cohort of 64 DW-MRI datasets with b-values ranging from 50 to 1000 s/mm2. The performance of the presented approach was evaluated using leave-one-subject-out cross validation and compared against three other well-known segmentation methods applied to the same DW-MRI data using the following evaluation metrics: 1) the Dice similarity coefficient (DSC); 2) the 95-percentile modified Hausdorff distance (MHD); and 3) the percentage kidney volume difference (PKVD). High performance of the new approach was confirmed by the high DSC (0.95±0.01), low MHD (3.9±0.76) mm, and low PKVD (9.5±2.2)% relative to manual segmentation by an MR expert (a board certified radiologist).

Rejection evaluation after renal transplantation using MR diffusion tensor imaging

Background Renal allograft dysfunction monitoring is mainly performed using the serum creatinine (SC) level, Doppler ultrasound (US), or renal biopsy. Recently proposed diffusion-based magnetic resonance imaging (MRI) methods have been explored as new, non-invasive tools for assessing renal function after transplantation. Purpose To investigate the value of fractional anisotropy (FA) measurements in the evaluation of acute rejection cases after renal transplant. Material and Methods Doppler US and MRI diffusion tensor imaging (DTI) were performed in 21 patients with graft dysfunction requiring graft biopsy after renal transplantation and in 21 patients with normal graft function. The MR examinations were performed on a 1.5-T MRI using two b-values (0 and 800 s/mm²). FA values were measured from the cortex and medulla of the transplanted kidney at the upper, middle, and lower poles. Results Twenty-one transplant patients diagnosed with acute rejection (Group 1) were compared to the control group of 21 transplant patients with normal graft function (Group 2). The measured FA values of the medulla were 0.19 ± 0.02 and 0.22 ± 0.05 ( P = 0.017) for Groups 1 and 2, respectively. On the other hand, the measured FA values of the renal cortex were 0.18 ± 0.04 and 0.18 ± 0.04 ( P = 0.97) for Groups 1 and 2, respectively. Conclusion The good correlation between the renal medulla FA values and allograft function shows that MR DTI has potential for non-invasive functional assessment of transplanted kidneys. On the other hand, the renal cortex FA values had no correlation with the allograft function.

Statistical analysis of ADCs and clinical biomarkers in detecting acute renal transplant rejection

OBJECTIVE

The main goal of this study is to determine which parameters [e.g. clinical biomarkers, demographics and image-markers using 4D (3D + b-value) diffusion-weighted MRI (DW-MRI)] are more correlated with transplanted kidney status in patients who have undergone kidney transplantation, and can be used for early assessment of acute renal rejection.

METHODS

The study included 16 patients with stable graft function and 37 patients with acute rejection (AR), determined by renal biopsy post-transplantation. 3D DW-MRI of each allograft had been acquired using a series of b-values 50 and 100-1000 in steps of 100 smm^-2. The kidney was automatically segmented and co-aligned across series for motion correction using geometric deformable models. Volume-averaged apparent diffusion coefficients (ADCs) at each b-value were calculated. All possible subsets of ADC were used, along with patient age, sex, serum plasma creatinine (SPCr) and creatinine clearance (CrCl), as predictors in 2¹¹ logistic regression models where AR was the outcome variable. Predictive value of ADC at each b-value was assessed using its Akaike weight.

RESULTS

ANOVA of the saturated model found that odds of AR depended significantly on SPCr, CrCl and ADC at b = 500, 600, 700 and 900 smm^-2. The model incorporating ADC at b = 100 and700 smm^-2 had the lowest value of the Akaike information criterion; the same two b-values also had the greatest Akaike weights. For comparison, the top 10 submodels and the full model were reported.

CONCLUSION

Preliminary findings suggest that ADC provides improved detection of AR than lab values alone. At least two non-zero gradient strengths should be used for optimal results. Advances in knowledge: This paper investigated possible correlations between image-based and clinical biomarkers, and the fusion of both with respect to biopsy diagnosis of AR.

Usefulness of Apparent Diffusion Coefficient of Diffusion-Weighted Imaging for Differential Diagnosis of Primary Solid and Cystic Renal Masses

BACKGROUND

To evaluate the value of diffusion-weighted imaging (DWI) for distinguishing between benign and malignant renal masses.

MATERIAL/METHODS

Seventy-five patients with 75 unilateral renal lesions were included, and 75 normal contralateral kidneys served as controls. The lesions were categorized into four groups as malignant cystic, malignant solid, benign cystic and benign solid. The apparent diffusion coefficients (ADCs) were evaluated for two different b values (b=600 s/mm² and b=1000 s/mm²). Receiving operating characteristic analysis was performed to identify threshold ADCs.

RESULTS

Sensitivity and specificity were 67% and 77% (p=0.003) at the cutoff value of 1.5 for b=600 s/mm², and 79% and 62% (p=0.004) at the cutoff value of 1.99 for b=1000 s/mm² as regards the differentiation between solid benign and malignant renal lesions. Sensitivity and specificity were 78% and 79% (p=0.001) at the cutoff value of 3.1 for b=600 s/mm², and 86% and 61% (p=0.003) at the cutoff value of 2.9 for b=1000 s/mm² as regrads the differentiation between benign and malignant cystic renal lesions.

CONCLUSIONS

DWI can be an effective diagnostic method for distinguishing between benign and malignant renal masses.

Evaluation of Free Breathing Versus Breath Hold Diffusion Weighted Imaging in Terms Apparent Diffusion Coefficient (ADC) and Signal-to-Noise Ratio (SNR) Values for Solid Abdominal Organs

Background

Our aim was to compare the apparent diffusion coefficient (ADC) values of normal abdominal parenchymal organs and signal-to-noise ratio (SNR) measurements in the same patients with breath hold (BH) and free breathing (FB) diffusion weighted imaging (DWI).

Material/Methods

Forty-eight patients underwent both BH and FB DWI. Spherical region of interest (ROI) was placed on the right hepatic lobe, spleen, pancreas, and renal cortices. ADC values were calculated for each organ on each sequence using an automated software. Image noise, defined as the standard deviation (SD) of the signal intensities in the most artifact-free area of the image background was measured by placing the largest possible ROI on either the left or the right side of the body outside the object in the recorded field of view. SNR was calculated using the formula: SNR=signal intensity (SI)_(organ)/standard deviation (SD)_(noise).

Results

There were no statistically significant differences in ADC values of the abdominal organs between BH and FB DWI sequences (p>0.05). There were statistically significant differences between SNR values of organs on BH and FB DWIs. SNRs were found to be better on FB DWI than BH DWI (p<0.001).

Conclusions

Free breathing DWI technique reduces image noise and increases SNR for abdominal examinations. Free breathing technique is therefore preferable to BH DWI in the evaluation of abdominal organs by DWI.

Acute pyelonephritis in transplanted kidneys: can diffusion-weighted magnetic resonance imaging be useful for diagnosis and follow-up?

PURPOSE

To assess reliability of diffusion-weighted magnetic resonance imaging (DW-MRI) in the management of acute pyelonephritis (APN) foci in transplanted kidneys.

MATERIALS AND METHODS

In the 2012-2014 period, 24 kidney-transplanted patients underwent MR screening for clinical suspicion of APN. Two readers independently analyzed all images, establishing presence and location of APN foci. The 22 patients who were positive at the MR exam constituted the study population. For each patient the apparent diffusion coefficient (ADC) was measured in the APN foci and in three sites of the healthy parenchyma (case-control comparison). The data were matched to the laboratory measurements for white blood cell, C-reactive protein, and serum creatinine.

RESULTS

Forty-six APN foci were found in 22/24 patients. At the acute stage, the difference in ADC between healthy parenchyma and APN foci was significant (2.06 ± 0.16 vs. 1.43 ± 0.32 × 10(-3) mm(2)/s; p < 0.0001). The performance of ADC as APN indicator was tested by the receiving operating characteristics (ROC) curve: the area under curve AUC = 0.99 witnessed an excellent discriminatory ability, with threshold APN/normal parenchyma 1.9 × 10(-3) mm(2)/s. At the 1-month follow-up 43/46 APN foci were no longer visible, with ADC values significantly higher than at the acute stage; all laboratory data were physiological, with WBC significantly reduced from the acute phase (5.2 ± 1.6 × 10(9)/L vs. 10.6 ± 4.8 × 10(9)/L; p < 0.0001). The other 3 patients underwent further therapy and exams, including a third MR.

CONCLUSIONS

DW-MRI with ADC measurement seems to be a reliable tool in diagnosing and monitoring APN foci in transplanted kidneys, with clinical impact on patient management.

Thoracic splenosis after a gunshot: diffusion-weighted MRI findings

BACKGROUND

Intrathoracic splenosis is a rare condition resulting from concomitant rupture of the spleen and left hemidiaphragm after a traumatic event involving the spleen and the diaphragma and is defined as autotransplantation of splenic tissue in thorax.

CASE REPORT

The aim of this study was to present a case report of a combined intrathoracic and subcutaneous splenosis in a patient 19 years after penetrating trauma. She has left dorsal side pain and routine chest roentgenogram shows pleural nodular masses. The patient was referred to us for radiologic work up.

CONCLUSIONS

The MRI scans revealed the intrathoracic and subcutan masses as mainly hypointense on T1-weighted images and hyperintense on T2-weighted images and significant restriction in diffusion-weighted images. Scintigraphy revealed abnormal hot spots in subcutaneous tissue and diaphragmatic pleura of the left hemithorax.