Radiomics-Based Image Phenotyping of Kidney Apparent Diffusion Coefficient Maps: Preliminary Feasibility & Efficacy

Noninvasive diagnosis of interstitial fibrosis in chronic kidney disease: a systematic review and meta-analysis

RATIONALE AND OBJECTIVES

Researchers have delved into noninvasive diagnostic methods of renal fibrosis (RF) in chronic kidney disease, including ultrasound (US), magnetic resonance imaging (MRI), and radiomics. However, the value of these diagnostic methods in the noninvasive diagnosis of RF remains contentious. Consequently, the present study aimed to systematically delineate the accuracy of the noninvasive diagnosis of RF.

MATERIALS AND METHODS

A systematic search covering PubMed, Embase, Cochrane Library, and Web of Science databases for all data available up to 28 July 2023 was conducted for eligible studies.

RESULTS

We included 21 studies covering 4885 participants. Among them, nine studies utilized US as a noninvasive diagnostic method, eight studies used MRI, and four articles employed radiomics. The sensitivity and specificity of US for detecting RF were 0.81 (95% CI: 0.76-0.86) and 0.79 (95% CI: 0.72-0.84). The sensitivity and specificity of MRI were 0.77 (95% CI: 0.70-0.83) and 0.92 (95% CI: 0.85-0.96). The sensitivity and specificity of radiomics were 0.69 (95% CI: 0.59-0.77) and 0.78 (95% CI: 0.68-0.85).

CONCLUSIONS

The current early noninvasive diagnostic methods for RF include US, MRI, and radiomics. However, this study demonstrates that US has a higher sensitivity for the detection of RF compared to MRI. Compared to US, radiomics studies based on US did not show superior advantages. Therefore, challenges still exist in the current radiomics approaches for diagnosing RF, and further exploration of optimized artificial intelligence (AI) algorithms and technologies is needed.

Multi-parametric MRI-based machine learning model for prediction of pathological grade of renal injury in a rat kidney cold ischemia-reperfusion injury model

BACKGROUND

Renal cold ischemia-reperfusion injury (CIRI), a pathological process during kidney transplantation, may result in delayed graft function and negatively impact graft survival and function. There is a lack of an accurate and non-invasive tool for evaluating the degree of CIRI. Multi-parametric MRI has been widely used to detect and evaluate kidney injury. The machine learning algorithms introduced the opportunity to combine biomarkers from different MRI metrics into a single classifier.

OBJECTIVE

To evaluate the performance of multi-parametric magnetic resonance imaging for grading renal injury in a rat model of renal cold ischemia-reperfusion injury using a machine learning approach.

METHODS

Eighty male SD rats were selected to establish a renal cold ischemia -reperfusion model, and all performed multiparametric MRI scans (DWI, IVIM, DKI, BOLD, T1mapping and ASL), followed by pathological analysis. A total of 25 parameters of renal cortex and medulla were analyzed as features. The pathology scores were divided into 3 groups using K-means clustering method. Lasso regression was applied for the initial selecting of features. The optimal features and the best techniques for pathological grading were obtained. Multiple classifiers were used to construct models to evaluate the predictive value for pathology grading.

RESULTS

All rats were categorized into mild, moderate, and severe injury group according the pathologic scores. The 8 features that correlated better with the pathologic classification were medullary and cortical Dp, cortical T2*, cortical Fp, medullary T2*, ∆T1, cortical RBF, medullary T1. The accuracy(0.83, 0.850, 0.81, respectively) and AUC (0.95, 0.93, 0.90, respectively) for pathologic classification of the logistic regression, SVM, and RF are significantly higher than other classifiers. For the logistic model and combining logistic, RF and SVM model of different techniques for pathology grading, the stable and perform are both well. Based on logistic regression, IVIM has the highest AUC (0.93) for pathological grading, followed by BOLD(0.90).

CONCLUSION

The multi-parametric MRI-based machine learning model could be valuable for noninvasive assessment of the degree of renal injury.

Characterization of motion dependent magnetic field inhomogeneity for DWI in the kidneys

PURPOSE

Diffusion-weighted imaging (DWI) of the abdomen has increased dramatically for both research and clinical purposes. Motion and static field inhomogeneity related challenges limit image quality of abdominopelvic imaging with the most conventional echo-planar imaging (EPI) pulse sequence. While reversed phase encoded imaging is increasingly used to facilitate distortion correction, it typically assumes one motion independent magnetic field distribution. In this study, we describe a more generalized workflow for the case of kidney DWI in which the field inhomogeneity at multiple respiratory phases is mapped and used to correct all images in a multi-contrast DWI series.

METHODS

In this HIPAA-compliant and IRB-approved prospective study, 8 volunteers (6 M, ages 28-51) had abdominal imaging performed in a 3 T MRI system (MAGNETOM Prisma; Siemens Healthcare, Erlangen, Germany) with ECG gating. Coronal oblique T2-weighted HASTE images were collected for anatomical reference. Sagittal phase-contrast (PC) MRI images through the left renal artery were collected to determine systolic and diastolic phases. Cardiac triggered oblique coronal DWI were collected at 10 b-values between 0 and 800 s/mm2 and 12 directions. DWI series were distortion corrected using field maps generated by forward and reversed phase encoded b = 0 images collected over the full respiratory cycle and matched by respiratory phase. Morphologic accuracy, intraseries spatial variability, and diffusion tensor imaging (DTI) metrics mean diffusivity (MD) and fractional anisotropy (FA) were compared for results generated with no distortion correction, correction with only one respiratory bin, and correction with multiple respiratory bins across the breathing cycle.

RESULTS

Computed field maps showed significant variation in static field with kidney laterality, region, and respiratory phase. Distortion corrected images showed significantly better registration to morphologic images than uncorrected images; for the left kidney, the multiple bin correction outperformed one bin correction. Line profile analysis showed significantly reduced spatial variation with multiple bins than one bin correction. DTI metrics were mostly similar between correction methods, with some differences observed in MD between uncorrected and corrected datasets.

CONCLUSIONS

Our results indicate improved morphology of kidney DWI and derived parametric maps as well as reduced variability over the full image series using the motion-resolved distortion correction. This work highlights some morphologic and quantitative metric improvements can be obtained for kidney DWI when distortion correction is performed in a respiratory-resolved manner.

Diffusion-Weighted Magnetic Resonance Imaging: Clinical Potential and Applications

Since its discovery in the 1980s [...].