Long-Term Risk of Clinical Progression Utilizing Magnetic Resonance Imaging Findings of Locally Recurrent Prostate Cancer in Patients with Biochemical Recurrence following Radical Prostatectomy.

PURPOSE: Our goal was to evaluate the long-term prognostic value of magnetic resonance imaging of the prostatectomy bed in patients with biochemical recurrence after radical prostatectomy for prostate cancer. MATERIALS AND METHODS: Men with biochemical recurrence after radical prostatectomy who were studied by prostatectomy bed magnetic resonance imaging for suspected local recurrence were retrospectively evaluated. Locally recurrent tumors were noted and measured from imaging reports. Patients with nodal/bone lesions at the time of imaging were excluded. Kaplan-Meier and Cox regression analyses were used to assess systemic progression-free and prostate cancer-specific survival. RESULTS: A total of 896 men were enrolled and the imaging positive and negative groups for local recurrent tumor consisted of 441 and 455 men, respectively. On univariate analysis, preoperative prostate specific antigen (p=0.02), clinical tumor stage (p=0.006), pathological Gleason score from prostatectomy (p=0.02), subsequent salvage radiotherapy (p <0.001), biochemical recurrence to magnetic resonance imaging time interval (p <0.001), age at magnetic resonance imaging (p=0.047) and prostate specific antigen at magnetic resonance imaging (p <0.001) were significantly different between magnetic resonance imaging positive and negative groups. Patients with negative magnetic resonance imaging results had worse systemic progression-free survival rates (p=0.025) and better prostate cancer-specific survival (p=0.016) than those with recurrence. Larger lesion size significantly increased risk of prostate cancer death (hazard ratio: 1.07; p <0.001). On multivariable analysis, pathological Gleason scores ≥7 were independent prognostic factors of systemic progression (p <0.05). CONCLUSIONS: Prostatectomy bed magnetic resonance imaging provides long-term prognostic information for the evaluation of patients with biochemical recurrence after prostatectomy. Post-prostatectomy patients with recurrent lesions on imaging had longer progression-free survival but shorter prostate cancer-specific survival compared to those without lesions. Additionally, those with larger lesions were associated with poorer cancer-specific survival.

Modified acquisition strategy for reduced motion artifact in super resolution T2 FSE multislice MRI: Application to prostate.

PURPOSE: To reduce slice-to-slice motion effects in multislice T2 -weighted fast-spin-echo ( T2 FSE) imaging, manifest as "scalloping" in reformats, by modification of the acquisition strategy and to show applicability in prostate MRI. METHODS: T2 FSE images of contiguous or overlapping slices are typically acquired using multiple passes in which each pass is comprised of multiple slices with slice-to-slice gaps. Combination of slices from all passes provides the desired sampling. For enhancement of through-plane resolution with super resolution or for reformatting into other orientations, subtle ≈1 mm motion between passes can cause objectionable "scalloping" artifact. Here we address this by subdivision of each pass into multiple segments. Interleaving of segments from the multiple passes causes all slices to be acquired over substantially the same time, reducing pass-to-pass motion effects. This was implemented in acquiring 78 overlapped T2 FSE axial slices and studied in phantoms and in 14 prostate MRI patients. Super-resolution axial images and sagittal reformats from the original and new segmented acquisitions were evaluated by 3 uroradiologists. RESULTS: For all criteria of sagittal reformats, the segmented acquisition was statistically superior to the original. For all sharpness criteria of axial images, although the trend preferred the original acquisition, the difference was not significant. For artifact in axial images, the segmented acquisition was significantly superior. CONCLUSIONS: For prostate MRI the new segmented acquisition significantly reduces the scalloping motion artifact that can be present in reformats due to long time lags between the acquisition of adjacent or overlapped slices while retaining image sharpness in the acquired axial slices.

Timing and distribution of early renal cell carcinoma recurrences stratified by pathological nodal status in M0 patients at the time of nephrectomy.

OBJECTIVES: To evaluate the timing and distribution of first renal cell carcinoma metastasis after nephrectomy stratified by nodal status. METHODS: We evaluated patients treated with nephrectomy for sporadic, unilateral renal cell carcinoma between 1970 and 2011 who subsequently developed distant metastasis to three or fewer sites. Site-specific metastases-free 2-year survival rates were estimated using the Kaplan-Meier method. Associations of nodal status with time to metastasis were evaluated using multivariable Cox regression models. RESULTS: A total of 1049 patients met the inclusion criteria (135 pN1, 914 pN0/x patients). The median time to identification of first distant metastasis for pN1 patients was 0.4 years (interquartile range 0.2-1.1 years) versus 2.2 years (interquartile range 0.6-6.0 years) in pN0/x patients. The most common site of metastasis was to the lung, but this occurred earlier in pN1 patients (median 0.3 years vs 2.0 years). pN1 was associated with significantly lower site-specific 2-year metastases-free survival when compared with pN0/x for lung (37% vs 70%, P < 0.001), bone (63% vs 87%, P < 0.001), non-regional lymph nodes (60% vs 96%, P < 0.001) and liver metastases (79% vs 91%, P < 0.001). On multivariable analysis, pN1 status remained significantly associated with lung, bone, and non-regional lymph node (all P < 0.001) metastases, but it was no longer associated with liver metastases (P = 0.3). CONCLUSIONS: pN1 nodal status in M0 patients treated with nephrectomy for renal cell carcinoma is associated with more frequent early metastasis to sites conferring poor prognosis when compared with pN0/x. Our findings highlight the importance of rigorous, early surveillance though the multimodal use of a comprehensive history, physical, laboratory and radiological studies, as outlined in societal guidelines.

Use of kZ -space for high through-plane resolution in multislice MRI: Application to prostate.

PURPOSE: The goal of this work is to demonstrate 1 mm through-plane resolution in multislice T2SE MRI using k Z -space processing of overlapping slices and show applicability in prostate MRI. METHODS: Multiple overlapped slices are acquired and Fourier transformed in the slice-select direction. The slice profile is incorporated into a Tikhonov-regularized reconstruction. Through-plane resolution is tested in a resolution phantom. An anthropomorphic prostate phantom is used to study the SNR, and results are compared with theoretical prediction. The proposed method is tested in 16 patients indicated for clinical prostate MRI who gave written informed consent as overseen by our IRB. The "proposed" vs. "reference" multislice images are compared using multiple evaluation criteria for through-plane resolution. RESULTS: The modulation transfer function (MTF) plots of the resolution phantom show good modulation at frequency 0.5 lp/mm, demonstrating 1 mm through-plane resolution restoration. The SNR measurements experimentally match the theoretically predicted values. The radiological evaluation shows that the proposed method is superior to the reference method for five criteria of sharpness but inferior with respect to artifacts. CONCLUSIONS: In conjunction with overlapped slices a k Z -space-based reconstruction approach can be used to improve through-plane resolution in multislice T2SE MRI. 1 mm resolution is demonstrated from 3.2 mm thick slices. The in vivo results from prostate MRI show improved sharpness when compared to the standard multislice method.

Multiple unilateral subcapsular cortical hemorrhagic cystic disease of the kidney: CT and MRI findings and clinical characteristic.

PURPOSE: The aim of this study was to clarify the radiologic and clinical characteristics of multiple unilateral subcapsular cortical hemorrhagic cystic disease of the kidney. METHOD: Fourteen patients with unique and characteristic multiple hemorrhagic subcapsular cortical cysts of the kidney, not categorized in any existing renal cystic diseases, were retrospectively reviewed. The clinical information including age, sex, symptom, family history of renal or renal cystic disease, and laboratory data were collected. CT and MRI findings including distribution, number and size of cysts, and CT attenuation and signal intensity on T1- and T2-weighted MRI of cysts were analyzed. RESULTS: All patients except one were young and none had a family history of renal or renal cystic disease. Common clinical symptoms were flank or abdominal pain and hematuria. In all cases, only the left kidney was involved at initial presentation. Cysts were small (median cyst size, 4-15 mm), numerous, and distributed mainly along the subcapsular cortex of the kidney. Cysts were hyper-attenuated on unenhanced CT, extremely hypointense on T2-weighted MRI, and mildly hyperintense on T1-weighted MRI. All patients except one had normal renal function. Imaging follow-up revealed stable or mildly progressive disease in seven patients. Two patients developed several hemorrhagic subcapsular cortical cysts in the right kidney at follow-up. Three of five patients with a renal pathology specimen showed concurrent IgA nephropathy. CONCLUSION: We have identified a unique renal cystic disease with multiple unilateral subcapsular cortical hemorrhagic cystic disease of the kidney that has a characteristic manifestation both radiologically and clinically. KEY POINTS: • Multiple unilateral subcapsular cortical hemorrhagic cystic disease of the kidney is a unique non-familial renal cystic disease with a characteristic manifestation both radiologically and clinically. • Most cases of multiple unilateral subcapsular cortical hemorrhagic cystic disease of the kidney are stable or slowly progressive, and do not require invasive intervention.

Safety and Image Quality of 1.5-T Endorectal Coil Multiparametric MRI of the Prostate or Prostatectomy Fossa for Patients With Pacemaker or Implantable Cardioverter-Defibrillator.

OBJECTIVE: The purpose of this study is to report the patient safety and image quality of 1.5-T multiparametric MRI of the prostate in patients with cardiac implantable electronic devices (CIEDs). MATERIALS AND METHODS: In this retrospective study, a database was searched to identify prostate multiparametric 1.5-T MRI examinations performed with endorectal coils for patients with CIEDs from 2012 to 2016 (study group) and matched patients without CIEDs (control group). Clinical safety in the study group was reviewed. The specific absorption rate (SAR) and signal-to-noise ratio (SNR) were measured in both groups. Imaging quality and artifact on T2-weighted images, DW images, and dynamic contrast-enhanced images were rated on a 5-point scale by two independent readers. RESULTS: The study group consisted of total 28 multiparametric MRI examinations in 25 patients. There were no serious device-related adverse effects observed (0/28; 0%), and the estimated whole-body SAR in the study group was never greater than 1.5 W/kg. The SNR values tended to be lower in the study group than in the control group. However, overall perceived image preferences and influences of artifacts on image quality for the study group were not significantly different from those for the control group (p > 0.05), which were rated above average (rating 3) by both readers 1 and 2. CONCLUSION: Multiparametric 1.5-T MRI examination of the prostate can be safely performed in selected patients with CIEDs under controlled conditions with applicable image quality while maintaining a SAR less than 1.5 W/kg.

Complications of Intravesical BCG Immunotherapy for Bladder Cancer.

An earlier incorrect version appeared online. This article was corrected on December 14, 2018.

Image texture features predict renal function decline in patients with autosomal dominant polycystic kidney disease.

Magnetic resonance imaging (MRI) examinations provide high-resolution information about the anatomic structure of the kidneys and are used to measure total kidney volume (TKV) in patients with Autosomal Dominant Polycystic Kidney Disease (ADPKD). Height-adjusted TKV (HtTKV) has become the gold-standard imaging biomarker for ADPKD progression at early stages of the disease when estimated glomerular filtration rate (eGFR) is still normal. However, HtTKV does not take advantage of the wealth of information provided by MRI. Here we tested whether image texture features provide additional insights into the ADPKD kidney that may be used as complementary information to existing biomarkers. A retrospective cohort of 122 patients from the Consortium for Radiologic Imaging Studies of Polycystic Kidney Disease (CRISP) study was identified who had T2-weighted MRIs and eGFR values over 70 mL/min/1.73m2 at the time of their baseline scan. We computed nine distinct image texture features for each patient. The ability of each feature to predict subsequent progression to CKD stage 3A, 3B, and 30% reduction in eGFR at eight-year follow-up was assessed. A multiple linear regression model was developed incorporating age, baseline eGFR, HtTKV, and three image texture features identified by stability feature selection (Entropy, Correlation, and Energy). Including texture in a multiple linear regression model (predicting percent change in eGFR) improved Pearson correlation coefficient from -0.51 (using age, eGFR, and HtTKV) to -0.70 (adding texture). Thus, texture analysis offers an approach to refine ADPKD prognosis and should be further explored for its utility in individualized clinical decision making and outcome prediction.

CT-guided transgluteal biopsy for systematic sampling of the prostate in patients without rectal access: a 13-year single-center experience.

OBJECTIVES: The purpose of our study was to examine the safety and diagnostic utility of transgluteal CT-guided prostate biopsy for prostate sampling in patients without rectal access. METHODS: Seventy-three biopsies were performed in 65 patients over a 13-year period (2002-2015). Mean prostate-specific antigen (PSA) at biopsy was 7.8 ng/mL (range 0.37-31.5). Electronic medical records were reviewed for procedural details and complications. Mean PSA and number of cores in malignant and benign cohorts were compared with Student's t test. RESULTS: Technical success rate was 97.3% (71/73; mean cores 8, range 3-28). Of these, 43.6% (31/71) yielded malignancy (mean Gleason score 7, range 6-10) and 56.3% (40/71) yielded benign tissue. The only complication was an asymptomatic periprostatic hematoma (1/73; 1.4%). In 14 patients who underwent surgery, Gleason scores were concordant in 71.4% (10/14) and discordant in 28.6% (4/14; Gleason 6 on biopsy but Gleason 7 on surgical specimen). Mean effective radiation dose was 18.5 mSv (median 15.0, range 4.4-86.2). There was no significant difference in either mean PSA (p = 0.06) or number of core specimens (p = 0.33) between malignant and benign cohorts. CONCLUSION: CT-guided transgluteal prostate biopsy is highly safe and reliable for the detection of prostate cancer in men without rectal access. KEY POINTS: • Prostate cancer detection in men without rectal access is challenging. • CT-guided transgluteal prostate biopsy is safe and effective in these patients. • CT-guided biopsy may be particularly effective in diagnosing high-grade prostate cancer. • Unilateral CT-guided biopsy may be effective in patients with focal lesions. • The radiation exposure with this technique is acceptable.

Performance of an Artificial Multi-observer Deep Neural Network for Fully Automated Segmentation of Polycystic Kidneys.

Deep learning techniques are being rapidly applied to medical imaging tasks-from organ and lesion segmentation to tissue and tumor classification. These techniques are becoming the leading algorithmic approaches to solve inherently difficult image processing tasks. Currently, the most critical requirement for successful implementation lies in the need for relatively large datasets that can be used for training the deep learning networks. Based on our initial studies of MR imaging examinations of the kidneys of patients affected by polycystic kidney disease (PKD), we have generated a unique database of imaging data and corresponding reference standard segmentations of polycystic kidneys. In the study of PKD, segmentation of the kidneys is needed in order to measure total kidney volume (TKV). Automated methods to segment the kidneys and measure TKV are needed to increase measurement throughput and alleviate the inherent variability of human-derived measurements. We hypothesize that deep learning techniques can be leveraged to perform fast, accurate, reproducible, and fully automated segmentation of polycystic kidneys. Here, we describe a fully automated approach for segmenting PKD kidneys within MR images that simulates a multi-observer approach in order to create an accurate and robust method for the task of segmentation and computation of TKV for PKD patients. A total of 2000 cases were used for training and validation, and 400 cases were used for testing. The multi-observer ensemble method had mean ± SD percent volume difference of 0.68 ± 2.2% compared with the reference standard segmentations. The complete framework performs fully automated segmentation at a level comparable with interobserver variability and could be considered as a replacement for the task of segmentation of PKD kidneys by a human.

Utilizing magnetization transfer imaging to investigate tissue remodeling in a murine model of autosomal dominant polycystic kidney disease.

PURPOSE: Noninvasive imaging techniques that quantify renal tissue composition are needed to more accurately ascertain prognosis and monitor disease progression in polycystic kidney disease (PKD). Given the success of magnetization transfer (MT) imaging to characterize various tissue remodeling pathologies, it was tested on a murine model of autosomal dominant PKD. METHODS: C57Bl/6 Pkd1 R3277C mice at 9, 12, and 15 months were imaged with a 16.4T MR imaging system. Images were acquired without and with RF saturation in order to calculate MT ratio (MTR) maps. Following imaging, the mice were euthanized and kidney sections were analyzed for cystic and fibrotic indices, which were compared with statistical parameters of the MTR maps. RESULTS: The MTR-derived mean, median, 25th percentile, skewness, and kurtosis were all closely related to indices of renal pathology, including kidney weight/body weight, cystic index, and percent of remaining parenchyma. The correlation between MTR and histology-derived cystic and fibrotic changes was R(2) = 0.84 and R(2) = 0.70, respectively. CONCLUSION: MT imaging provides a new, noninvasive means of measuring tissue remodeling PKD changes and may be better suited for characterizing renal impairment compared with conventional MR techniques.

Automatic total kidney volume measurement on follow-up magnetic resonance images to facilitate monitoring of autosomal dominant polycystic kidney disease progression.

BACKGROUND: Renal imaging examinations provide high-resolution information about the anatomic structure of the kidneys and are used to measure total kidney volume (TKV) in autosomal dominant polycystic kidney disease (ADPKD) patients. TKV has become the gold-standard image biomarker for ADPKD progression at early stages of the disease and is used in clinical trials to characterize treatment efficacy. Automated methods to segment the kidneys and measure TKV are desirable because of the long time requirement for manual approaches such as stereology or planimetry tracings. However, ADPKD kidney segmentation is complicated by a number of factors, including irregular kidney shapes and variable tissue signal at the kidney borders. METHODS: We describe an image processing approach that overcomes these problems by using a baseline segmentation initialization to provide automatic segmentation of follow-up scans obtained years apart. We validated our approach using 20 patients with complete baseline and follow-up T1-weighted magnetic resonance images. Both manual tracing and stereology were used to calculate TKV, with two observers performing manual tracings and one observer performing repeat tracings. Linear correlation and Bland-Altman analysis were performed to compare the different approaches. RESULTS: Our automated approach measured TKV at a level of accuracy (mean difference ± standard error = 0.99 ± 0.79%) on par with both intraobserver (0.77 ± 0.46%) and interobserver variability (1.34 ± 0.70%) of manual tracings. All approaches had excellent agreement and compared favorably with ground-truth manual tracing with interobserver, stereological and automated approaches having 95% confidence intervals ∼ ± 100 mL. CONCLUSIONS: Our method enables fast, cost-effective and reproducible quantification of ADPKD progression that will facilitate and lower the costs of clinical trials in ADPKD and other disorders requiring accurate, longitudinal kidney quantification. In addition, it will hasten the routine use of TKV as a prognostic biomarker in ADPKD.

Renal hemodynamic effects of the HMG-CoA reductase inhibitors in autosomal dominant polycystic kidney disease.

BACKGROUND: To determine the effect of statins on renal hemodynamics in normal volunteers and those with autosomal dominant polycystic kidney disease either with mild or moderate renal dysfunction. METHODS: Thirty-two study subjects were enrolled in this study: 11 normal volunteers, 11 study subjects with autosomal dominant polycystic kidney disease (ADPKD) and mild kidney disease and 10 study subjects with ADPKD and moderate kidney disease. Subjects in each group received simvastatin 40 mg once daily for a period of 4 weeks. Renal blood flow was measured based on para-amino-hippurate (PAH) clearance and with the use of a magnetic resonance (MR) scanner at the beginning and following 4 weeks of therapy with statins. RESULTS: At the end of the study, except for the lipid profile, which was significantly lower in all groups, other laboratory results showed no change. Four weeks of therapy with simvastatin resulted in no change in serum creatinine, 24-h urinary protein, sodium, iothalamate clearance, PAH clearance or renal blood flow as measured by MRI or based on PAH clearance. CONCLUSIONS: Four weeks of therapy with simvastatin did not change renal blood flow in the study subjects with ADPKD with mild-to-moderate renal dysfunction or in healthy volunteers. CLINICAL TRIAL REGISTRATION NUMBER: NCT02511418.

Imaging classification of autosomal dominant polycystic kidney disease: a simple model for selecting patients for clinical trials.

The rate of renal disease progression varies widely among patients with autosomal dominant polycystic kidney disease (ADPKD), necessitating optimal patient selection for enrollment into clinical trials. Patients from the Mayo Clinic Translational PKD Center with ADPKD (n=590) with computed tomography/magnetic resonance images and three or more eGFR measurements over ≥6 months were classified radiologically as typical (n=538) or atypical (n=52). Total kidney volume (TKV) was measured using stereology (TKVs) and ellipsoid equation (TKVe). Typical patients were randomly partitioned into development and internal validation sets and subclassified according to height-adjusted TKV (HtTKV) ranges for age (1A-1E, in increasing order). Consortium for Radiologic Imaging Study of PKD (CRISP) participants (n=173) were used for external validation. TKVe correlated strongly with TKVs, without systematic underestimation or overestimation. A longitudinal mixed regression model to predict eGFR decline showed that log2HtTKV and age significantly interacted with time in typical patients, but not in atypical patients. When 1A-1E classifications were used instead of log2HtTKV, eGFR slopes were significantly different among subclasses and, except for 1A, different from those in healthy kidney donors. The equation derived from the development set predicted eGFR in both validation sets. The frequency of ESRD at 10 years increased from subclass 1A (2.4%) to 1E (66.9%) in the Mayo cohort and from 1C (2.2%) to 1E (22.3%) in the younger CRISP cohort. Class and subclass designations were stable. An easily applied classification of ADPKD based on HtTKV and age should optimize patient selection for enrollment into clinical trials and for treatment when one becomes available.

Role of kidney Doppler ultrasonography in the diagnosis and management of anuric kidney failure.

Kidney perfusion can be acutely compromised by many factors including reduced systemic blood pressure and elevated intra-abdominal pressure. We present a case of near complete absence of kidney perfusion in a 57-year-old man with heart failure and new onset ascites. The renal perfusion defect was directly detected by Doppler ultrasonography. Immediate decompression with large-volume paracentesis restored the kidney perfusion and kidney function. This case illustrates that renal ultrasonography with Doppler flow analysis in appropriate settings can serve as an important adjunct in the diagnosis and treatment of acute oligoanuric kidney failure. Timely reversal of the perfusion defect can rescue kidney function.

Early localization of recurrent prostate cancer after prostatectomy by endorectal coil magnetic resonance imaging.

INTRODUCTION: To evaluate the ability of endorectal coil (e-coil) magnetic resonance imaging (MRI) to identify early prostatic fossa recurrence after radical prostatectomy. MATERIALS AND METHODS: We identified 187 patients from 2005-2011 who underwent e-coil MRI with dynamic gadolinium-contrast enhancement followed by transrectal ultrasound (TRUS) guided prostatic fossa biopsy for possible local prostate cancer recurrence. For analysis, local recurrence was defined as a negative evaluation for distant metastatic disease with a positive prostatic fossa biopsy, decreased prostate-specific antigen (PSA) following salvage radiation therapy, or increased lesion size on serial imaging. RESULTS: Local recurrence was identified in 132 patients, with 124 (94%) detected on e-coil MRI. The median PSA was 0.59 ng/mL (range < 0.1-13.1), and median lesion size on MRI was 1 cm. The sensitivity of MRI was 91%, with a specificity of 45%. The positive predictive value was 85%, with a negative predictive value of 60%. For patients with a PSA < 0.4 ng/mL the sensitivity of e-coil MRI was 86%. When a lesion was identified on MRI, the positive biopsy rate was 65% and lesion size was a significant predictor of positive biopsies. The positive biopsy rates were 51%, 74%, and 88% when the lesion was < 1 cm, 1 cm-2 cm, or > 2 cm, respectively (p = 0.0006). CONCLUSIONS: E-coil MRI has a high level of sensitivity in identifying local recurrence of prostate cancer following radical prostatectomy, even at low PSA levels. E-coil MRI should be considered as the first imaging evaluation for biochemical recurrence for identifying patients suitable for localized salvage therapy.

Supervised segmentation of polycystic kidneys: a new application for stereology data.

Stereology is a volume estimation method, typically applied to diagnostic imaging examinations in population studies where planimetry is too time-consuming (Chapman et al. Kidney Int 64:1035-1045, 2003), to obtain quantitative measurements (Nyengaard J Am Soc Nephrol 10:1100-1123, 1999, Michel and Cruz-Orive J Microsc 150:117-136, 1988) of certain structures or organs. However, true segmentation is required in order to perform advanced analysis of the tissues. This paper describes a novel method for segmentation of region(s) of interest using stereology data as prior information. The result is an efficient segmentation method for structures that cannot be easily segmented using other methods.

Mutations in SEC63 cause autosomal dominant polycystic liver disease.

Mutations in PRKCSH, encoding the beta-subunit of glucosidase II, an N-linked glycan-processing enzyme in the endoplasmic reticulum (ER), cause autosomal dominant polycystic liver disease. We found that mutations in SEC63, encoding a component of the protein translocation machinery in the ER, also cause this disease. These findings are suggestive of a role for cotranslational protein-processing pathways in maintaining epithelial luminal structure and implicate noncilial ER proteins in human polycystic disease.

Clinical Implementation of an Artificial Intelligence Algorithm for Magnetic Resonance-Derived Measurement of Total Kidney Volume.

OBJECTIVE: To evaluate the performance of an internally developed and previously validated artificial intelligence (AI) algorithm for magnetic resonance (MR)-derived total kidney volume (TKV) in autosomal dominant polycystic kidney disease (ADPKD) when implemented in clinical practice. PATIENTS AND METHODS: The study included adult patients with ADPKD seen by a nephrologist at our institution between November 2019 and January 2021 and undergoing an MR imaging examination as part of standard clinical care. Thirty-three nephrologists ordered MR imaging, requesting AI-based TKV calculation for 170 cases in these 161 unique patients. We tracked implementation and performance of the algorithm over 1 year. A radiologist and a radiology technologist reviewed all cases (N=170) for quality and accuracy. Manual editing of algorithm output occurred at radiology or radiology technologist discretion. Performance was assessed by comparing AI-based and manually edited segmentations via measures of similarity and dissimilarity to ensure expected performance. We analyzed ADPKD severity class assignment of algorithm-derived vs manually edited TKV to assess impact. RESULTS: Clinical implementation was successful. Artificial intelligence algorithm-based segmentation showed high levels of agreement and was noninferior to interobserver variability and other methods for determining TKV. Of manually edited cases (n=84), the AI-algorithm TKV output showed a small mean volume difference of -3.3%. Agreement for disease class between AI-based and manually edited segmentation was high (five cases differed). CONCLUSION: Performance of an AI algorithm in real-life clinical practice can be preserved if there is careful development and validation and if the implementation environment closely matches the development conditions.