Corrigendum to "Erdheim-Chester disease with tendon and muscle involvement: Reports of a rare presentation" [Radiology Case Reports 19 (2024) 1866-1871].

[This corrects the article DOI: 10.1016/j.radcr.2024.02.009.].

Erdheim-Chester disease with tendon and muscle involvement: Reports of a rare presentation.

Erdheim-Chester disease (ECD) is a rare histiocytic disease that affects multiple systems in the body. While it typically targets long bones, cardiovascular structures, the retroperitoneum, and the central nervous system, reports of tendon and skeletal muscle involvement are scarce. This review presents 2 cases: a case of ECD involving the left Achilles tendon and left abductor hallucis, as well as an unusual manifestation of ECD in the thigh musculature. In Case 1, studies involved a 39-year-old man who initially presented with bone and pituitary involvement. An order for 18F-FDG PET/CT imaging was placed by marked swelling in the patient's left ankle and observed soft tissue fullness on foot radiographs, which revealed a soft tissue mass involving the left Achilles tendon, which arose along the tendon-muscle junction and involved the left abductor hallucis muscle. In Case 2, studies involved a 41-year-old man who initially presented with involvement of the cardiovascular system and retroperitoneum. 18F-FDG PET/CT scan showed an infiltrative right atrial mass and hypermetabolic lesion in the left external obturator muscle, extending to the left pectineus and right quadratus femoris muscle. Involvement of the Achilles tendon and skeletal muscle involvement, including left abductor hallucis muscle and medial thigh muscles, is one of the rare manifestations of ECD. Diagnostic delays were frequent due to the condition's rarity and nonspecific multisystemic symptoms. This should be considered in patients who present with myositis, tendinopathy, and bone pain and have other unexplained multisystemic problems.

Kidney scoring surveillance: predictive machine learning models for clear cell renal cell carcinoma growth using MRI.

INTRODUCTION: Classification of clear cell renal cell carcinoma (ccRCC) growth rates in patients with Von Hippel-Lindau (VHL) syndrome has several ramifications for tumor monitoring and surgical planning. Using two separate machine-learning algorithms, we sought to produce models to predict ccRCC growth rate classes based on qualitative MRI-derived characteristics. MATERIAL AND METHODS: We used a prospectively maintained database of patients with VHL who underwent surgical resection for ccRCC between January 2015 and June 2022. We employed a threshold growth rate of 0.5 cm per year to categorize ccRCC tumors into two distinct groups-'slow-growing' and 'fast-growing'. Utilizing a questionnaire of qualitative imaging features, two radiologists assessed each lesion on different MRI sequences. Two machine-learning models, a stacked ensemble technique and a decision tree algorithm, were used to predict the tumor growth rate classes. Positive predictive value (PPV), sensitivity, and F1-score were used to evaluate the performance of the models. RESULTS: This study comprises 55 patients with VHL with 128 ccRCC tumors. Patients' median age was 48 years, and 28 patients were males. Each patient had an average of two tumors, with a median size of 2.1 cm and a median growth rate of 0.35 cm/year. The overall performance of the stacked and DT model had 0.77 ± 0.05 and 0.71 ± 0.06 accuracies, respectively. The best stacked model achieved a PPV of 0.92, a sensitivity of 0.91, and an F1-score of 0.90. CONCLUSION: This study provides valuable insight into the potential of machine-learning analysis for the determination of renal tumor growth rate in patients with VHL. This finding could be utilized as an assistive tool for the individualized screening and follow-up of this population.

Deep learning algorithm (YOLOv7) for automated renal mass detection on contrast-enhanced MRI: a 2D and 2.5D evaluation of results.

INTRODUCTION: Accurate diagnosis and treatment of kidney tumors greatly benefit from automated solutions for detection and classification on MRI. In this study, we explore the application of a deep learning algorithm, YOLOv7, for detecting kidney tumors on contrast-enhanced MRI. MATERIAL AND METHODS: We assessed the performance of YOLOv7 tumor detection on excretory phase MRIs in a large institutional cohort of patients with RCC. Tumors were segmented on MRI using ITK-SNAP and converted to bounding boxes. The cohort was randomly divided into ten benchmarks for training and testing the YOLOv7 algorithm. The model was evaluated using both 2-dimensional and a novel in-house developed 2.5-dimensional approach. Performance measures included F1, Positive Predictive Value (PPV), Sensitivity, F1 curve, PPV-Sensitivity curve, Intersection over Union (IoU), and mean average PPV (mAP). RESULTS: A total of 326 patients with 1034 tumors with 7 different pathologies were analyzed across ten benchmarks. The average 2D evaluation results were as follows: Positive Predictive Value (PPV) of 0.69 ± 0.05, sensitivity of 0.39 ± 0.02, and F1 score of 0.43 ± 0.03. For the 2.5D evaluation, the average results included a PPV of 0.72 ± 0.06, sensitivity of 0.61 ± 0.06, and F1 score of 0.66 ± 0.04. The best model performance demonstrated a 2.5D PPV of 0.75, sensitivity of 0.69, and F1 score of 0.72. CONCLUSION: Using computer vision for tumor identification is a cutting-edge and rapidly expanding subject. In this work, we showed that YOLOv7 can be utilized in the detection of kidney cancers.

Non-Invasive Tumor Grade Evaluation in Von Hippel-Lindau-Associated Clear Cell Renal Cell Carcinoma: A Magnetic Resonance Imaging-Based Study.

BACKGROUND: Pathology grading is an essential step for the treatment and evaluation of the prognosis in patients with clear cell renal cell carcinoma (ccRCC). PURPOSE: To investigate the utility of texture analysis in evaluating Fuhrman grades of renal tumors in patients with Von Hippel-Lindau (VHL)-associated ccRCC, aiming to improve non-invasive diagnosis and personalized treatment. STUDY TYPE: Retrospective analysis of a prospectively maintained cohort. POPULATION: One hundred and thirty-six patients, 84 (61%) males and 52 (39%) females with pathology-proven ccRCC with a mean age of 52.8 ± 12.7 from 2010 to 2023. FIELD STRENGTH AND SEQUENCES: 1.5 and 3 T MRIs. Segmentations were performed on the T1-weighted 3-minute delayed sequence and then registered on pre-contrast, T1-weighted arterial and venous sequences. ASSESSMENT: A total of 404 lesions, 345 low-grade tumors, and 59 high-grade tumors were segmented using ITK-SNAP on a T1-weighted 3-minute delayed sequence of MRI. Radiomics features were extracted from pre-contrast, T1-weighted arterial, venous, and delayed post-contrast sequences. Preprocessing techniques were employed to address class imbalances. Features were then rescaled to normalize the numeric values. We developed a stacked model combining random forest and XGBoost to assess tumor grades using radiomics signatures. STATISTICAL TESTS: The model's performance was evaluated using positive predictive value (PPV), sensitivity, F1 score, area under the curve of receiver operating characteristic curve, and Matthews correlation coefficient. Using Monte Carlo technique, the average performance of 100 benchmarks of 85% train and 15% test was reported. RESULTS: The best model displayed an accuracy of 0.79. For low-grade tumor detection, a sensitivity of 0.79, a PPV of 0.95, and an F1 score of 0.86 were obtained. For high-grade tumor detection, a sensitivity of 0.78, PPV of 0.39, and F1 score of 0.52 were reported. DATA CONCLUSION: Radiomics analysis shows promise in classifying pathology grades non-invasively for patients with VHL-associated ccRCC, potentially leading to better diagnosis and personalized treatment. LEVEL OF EVIDENCE: 1 TECHNICAL EFFICACY: Stage 2.

The kidney imaging surveillance scoring system (KISSS): using qualitative MRI features to predict growth rate of renal tumors in patients with von-Hippel Lindau (VHL) syndrome.

OBJECTIVE: To determine the reliability of an MRI-based qualitative kidney imaging surveillance scoring system (KISSS) and assess which imaging features predict growth rate (GR) of renal tumors in patients with VHL. MATERIALS AND METHODS: We identified 55 patients with VHL with 128 renal tumors who underwent intervention from 2015 to 2020 at the National Cancer Institute. All patients had 2 preoperative MRIs at least 3 months apart. Two fellowship-trained radiologists scored each tumor on location and MR-sequence-specific imaging parameters from the earlier MRI. Weighted kappa was used to determine the degree of agreement between radiologists for each parameter. GR was calculated as the difference in maximum tumor dimension over time (cm/year). Differences in mean growth rate (MGR) within categories of each imaging variable were assessed by ANOVA. RESULTS: Apart from tumor margin and renal sinus, reliability was at least moderate (K > 0.40) for imaging parameters. Median initial tumor size was 2.1 cm, with average follow-up of 1.2 years. Tumor MGR was 0.42 cm/year. T2 hypointense, mixed/predominantly solid, and high restricted diffusion tumors grew faster. When comparing different combinations of these variables, the model with the lowest mean error among both radiologists utilized only solid/cystic and restricted diffusion features. CONCLUSIONS: We demonstrate a novel MR-based scoring system (KISSS) that has good precision with minimal training and can be applied to other qualitative radiology studies. A subset of imaging variables (T2 intensity; restricted diffusion; and solid/cystic) were independently associated with growth rate in VHL renal tumors, with the combination of the latter two most optimal. Additional validation, including in sporadic RCC population, is warranted.

Skeletal involvement in Erdheim-Chester disease: Multimodality imaging features and association with the BRAFV600E mutation.

OBJECTIVE: The aim of this study was to characterize the distribution of skeletal involvement in Erdheim-Chester disease (ECD) by using radiography, computed tomography (CT), 18F-FDG positron emission tomography/computed tomography (PET/CT), and bone scans, as well as looking for associations with the BRAFV600E mutation. MATERIAL AND METHODS: Prospective study of 50 consecutive patients with biopsy-confirmed ECD who had radiographs, CT, 18F-FDG PET/CT, and Tc-99m MDP bone scans. At least two experienced radiologists with expertise in the relevant imaging studies analyzed the images. Summary statistics were expressed as the frequency with percentages for categorical data. Fisher's exact test, as well as odds ratios (OR) with 95 % confidence intervals (CI), were used to link imaging findings to BRAFV600E mutation. The probability for co-occurrence of bone involvement at different locations was calculated and graphed as a heat map. RESULTS: All 50 cases revealed skeletal involvement at different regions of the skeleton. The BRAFV600E mutation, which was found in 24 patients, was correlated with femoral and tibial involvement on 18F-FDG PET/CT and bone scan. The appearance of changes on the femoral, tibial, fibular, and humeral involvement showed correlation with each other based on heat maps of skeletal involvement on CT. CONCLUSION: This study reports the distribution of skeletal involvement in a cohort of patients with ECD. CT is able to detect the majority of ECD skeletal involvement. Considering the complementary nature of information from different modalities, imaging of ECD skeletal involvement is optimized by using a multi-modality strategy.

Longitudinal Natural History Study of Children and Adults with Rare Solid Tumors: Initial Results for First 200 Participants.

Understanding of tumor biology and identification of effective therapies is lacking for many rare tumors. My Pediatric and Adult Rare Tumor (MyPART) network was established to engage patients, advocates, and researchers and conduct a comprehensive longitudinal Natural History Study of Rare Solid Tumors. Through remote or in-person enrollment at the NIH Clinical Center, participants with rare solid tumors ≥4 weeks old complete standardized medical and family history forms, patient reported outcomes, and provide tumor, blood and/or saliva samples. Medical records are extracted for clinical status and treatment history, and tumors undergo genomic analysis. A total of 200 participants (65% female, 35% male, median age at diagnosis 43 years, range = 2-77) enrolled from 46 U.S. states and nine other countries (46% remote, 55% in-person). Frequent diagnoses were neuroendocrine neoplasms (NEN), adrenocortical carcinomas (ACC), medullary thyroid carcinomas (MTC), succinate dehydrogenase (SDH)-deficient gastrointestinal stromal tumors (sdGIST), and chordomas. At enrollment, median years since diagnosis was 3.5 (range = 0-36.6), 63% participants had metastatic disease and 20% had no evidence of disease. Pathogenic germline and tumor mutations included SDHA/B/C (sdGIST), RET (MTC), TP53 and CTNNB1 (ACC), MEN1 (NEN), and SMARCB1 (poorly-differentiated chordoma). Clinically significant anxiety was observed in 20%-35% of adults. Enrollment of participants and comprehensive data collection were feasible. Remote enrollment was critical during the COVID-19 pandemic. Over 30 patients were enrolled with ACC, NEN, and sdGIST, allowing for clinical/genomic analyses across tumors. Longitudinal follow-up and expansion of cohorts are ongoing to advance understanding of disease course and establish external controls for interventional trials. SIGNIFICANCE: This study demonstrates that comprehensive, tumor-agnostic data and biospecimen collection is feasible to characterize different rare tumors, and speed progress in research. The findings will be foundational to developing external controls groups for single-arm interventional trials, where randomized control trials cannot be conducted because of small patient populations.

Evaluating the performance of Generative Pre-trained Transformer-4 (GPT-4) in standardizing radiology reports.

OBJECTIVE: Radiology reporting is an essential component of clinical diagnosis and decision-making. With the advent of advanced artificial intelligence (AI) models like GPT-4 (Generative Pre-trained Transformer 4), there is growing interest in evaluating their potential for optimizing or generating radiology reports. This study aimed to compare the quality and content of radiologist-generated and GPT-4 AI-generated radiology reports. METHODS: A comparative study design was employed in the study, where a total of 100 anonymized radiology reports were randomly selected and analyzed. Each report was processed by GPT-4, resulting in the generation of a corresponding AI-generated report. Quantitative and qualitative analysis techniques were utilized to assess similarities and differences between the two sets of reports. RESULTS: The AI-generated reports showed comparable quality to radiologist-generated reports in most categories. Significant differences were observed in clarity (p = 0.027), ease of understanding (p = 0.023), and structure (p = 0.050), favoring the AI-generated reports. AI-generated reports were more concise, with 34.53 fewer words and 174.22 fewer characters on average, but had greater variability in sentence length. Content similarity was high, with an average Cosine Similarity of 0.85, Sequence Matcher Similarity of 0.52, BLEU Score of 0.5008, and BERTScore F1 of 0.8775. CONCLUSION: The results of this proof-of-concept study suggest that GPT-4 can be a reliable tool for generating standardized radiology reports, offering potential benefits such as improved efficiency, better communication, and simplified data extraction and analysis. However, limitations and ethical implications must be addressed to ensure the safe and effective implementation of this technology in clinical practice. CLINICAL RELEVANCE STATEMENT: The findings of this study suggest that GPT-4 (Generative Pre-trained Transformer 4), an advanced AI model, has the potential to significantly contribute to the standardization and optimization of radiology reporting, offering improved efficiency and communication in clinical practice. KEY POINTS: • Large language model-generated radiology reports exhibited high content similarity and moderate structural resemblance to radiologist-generated reports. • Performance metrics highlighted the strong matching of word selection and order, as well as high semantic similarity between AI and radiologist-generated reports. • Large language model demonstrated potential for generating standardized radiology reports, improving efficiency and communication in clinical settings.

Preoperative Renal Parenchyma Volume as a Predictor of Kidney Function Following Nephrectomy of Complex Renal Masses.

Background: The von Hippel-Lindau disease (VHL) is a hereditary cancer syndrome with multifocal, bilateral cysts and solid tumors of the kidney. Surgical management may include multiple extirpative surgeries, which ultimately results in parenchymal volume loss and subsequent renal function decline. Recent studies have utilized parenchyma volume as an estimate of renal function prior to surgery for renal cell carcinoma; however, it is not yet validated for surgically altered kidneys with multifocal masses and complex cysts such as are present in VHL. Objective: We sought to validate a magnetic resonance imaging (MRI)-based volumetric analysis with mercaptoacetyltriglycine (MAG-3) renogram and postoperative renal function. Design setting and participants: We identified patients undergoing renal surgery at the National Cancer Institute from 2015 to 2020 with preoperative MRI. Renal tumors, cysts, and parenchyma of the operated kidney were segmented manually using ITK-SNAP software. Outcome measurements and statistical analysis: Serum creatinine and urinalysis were assessed preoperatively, and at 3- and 12-mo follow-up time points. Estimated glomerular filtration rate (eGFR) was calculated using serum creatinine-based CKD-EPI 2021 equation. A statistical analysis was conducted on R Studio version 4.1.1. Results and limitations: Preoperative MRI scans of 113 VHL patients (56% male, median age 48 yr) were evaluated between 2015 and 2021. Twelve (10.6%) patients had a solitary kidney at the time of surgery; 59 (52%) patients had at least one previous partial nephrectomy on the renal unit. Patients had a median of three (interquartile range [IQR]: 2-5) tumors and five (IQR: 0-13) cysts per kidney on imaging. The median preoperative GFR was 70 ml/min/1.73 m2 (IQR: 58-89). Preoperative split renal function derived from MAG-3 studies and MRI split renal volume were significantly correlated (r = 0.848, p < 0.001). On the multivariable analysis, total preoperative parenchymal volume, solitary kidney, and preoperative eGFR were significant independent predictors of 12-mo eGFR. When only considering patients with two kidneys undergoing partial nephrectomy, preoperative parenchymal volume and eGFR remained significant predictors of 12-mo eGFR. Conclusions: A parenchyma volume analysis on preoperative MRI correlates well with renogram split function and can predict long-term renal function with added benefit of anatomic detail and ease of application. Patient summary: Prior to kidney surgery, it is important to understand the contribution of each kidney to overall kidney function. Nuclear medicine scans are currently used to measure split kidney function. We demonstrated that kidney volumes on preoperative magnetic resonance imaging can also be used to estimate split kidney function before surgery, while also providing essential details of tumor and kidney anatomy.

Deep Learning-Based Interpretable AI for Prostate T2W MRI Quality Evaluation.

RATIONALE AND OBJECTIVES: Prostate MRI quality is essential in guiding prostate biopsies. However, assessment of MRI quality is subjective with variation. Quality degradation sources exert varying impacts based on the sequence under consideration, such as T2W versus DWI. As a result, employing sequence-specific techniques for quality assessment could yield more advantageous outcomes. This study aims to develop an AI tool that offers a more consistent evaluation of T2W prostate MRI quality, efficiently identifying suboptimal scans while minimizing user bias. MATERIALS AND METHODS: This retrospective study included 1046 patients from three cohorts (ProstateX [n = 347], All-comer in-house [n = 602], enriched bad-quality MRI in-house [n = 97]) scanned between January 2011 and May 2022. An expert reader assigned T2W MRIs a quality score. A train-validation-test split of 70:15:15 was applied, ensuring equal distribution of MRI scanners and protocols across all partitions. T2W quality AI classification model was based on 3D DenseNet121 architecture using MONAI framework. In addition to multiclassification, binary classification was utilized (Classes 0/1 vs. 2). A score of 0 was given to scans considered non-diagnostic or unusable, a score of 1 was given to those with acceptable diagnostic quality with some usability but with some quality distortions present, and a score of 2 was given to those considered optimal diagnostic quality and usability. Partial occlusion sensitivity maps were generated for anatomical correlation. Three body radiologists assessed reproducibility within a subgroup of 60 test cases using weighted Cohen Kappa. RESULTS: The best validation multiclass accuracy of 77.1% (121/157) was achieved during training. In the test dataset, multiclassification accuracy was 73.9% (116/157), whereas binary accuracy was 84.7% (133/157). Sub-class sensitivity for binary quality distortion classification for class 0 was 100% (18/18), and sub-class specificity for T2W classification of absence/minimal quality distortions for class 2 was 90.5% (95/105). All three readers showed moderate to substantial agreement with ground truth (R1-R3 κ = 0.588, κ = 0.649, κ = 0.487, respectively), moderate to substantial agreement with each other (R1-R2 κ = 0.599, R1-R3 κ = 0.612, R2-R3 κ = 0.685), fair to moderate agreement with AI (R1-R3 κ = 0.445, κ = 0.410, κ = 0.292, respectively). AI showed substantial agreement with ground truth (κ = 0.704). 3D quality heatmap evaluation revealed that the most critical non-diagnostic quality imaging features from an AI perspective related to obscuration of the rectoprostatic space (94.4%, 17/18). CONCLUSION: The 3D AI model can assess T2W prostate MRI quality with moderate accuracy and translate whole sequence-level classification labels into 3D voxel-level quality heatmaps for interpretation. Image quality has a significant downstream impact on ruling out clinically significant cancers. AI may be able to help with reproducible identification of MRI sequences requiring re-acquisition with explainability.

Virtual monoenergetic imaging in photon-counting CT of the head and neck.

PURPOSE: Advantages of virtual monoenergetic images (VMI) have been reported for dual energy CT of the head and neck, and more recently VMIs derived from photon-counting (PCCT) angiography of the head and neck. We report image quality metrics of VMI in a PCCT angiography dataset, expanding the anatomical regions evaluated and extending observer-based qualitative methods further than previously reported. METHODS: In a prospective study, asymptomatic subjects underwent contrast enhanced PCCT of the head and neck using an investigational scanner. Image sets of low, high, and full spectrum (Threshold-1) energies; linear mix of low and high energies (Mix); and 23 VMIs (40-150 keV, 5 keV increments) were generated. In 8 anatomical locations, SNR and radiologists' preferences for VMI energy levels were measured using a forced-choice rank method (4 observers) and ratings of image quality using visual grading characteristic (VGC) analysis (2 observers) comparing VMI to Mix and Threshold-1 images. RESULTS: Fifteen subjects were included (7 men, 8 women, mean 57 years, range 46-75). Among all VMIs, SNRs varied by anatomic location. The highest SNRs were observed in VMIs. Radiologists preferred 50-60 keV VMIs for vascular structures and 75-85 keV for all other structures. Cumulative ratings of image quality averaged across all locations were higher for VMIs with areas under the curve of VMI vs Mix and VMI vs Threshold-1 of 0.67 and 0.68 for the first reader and 0.72 and 0.76 for the second, respectively. CONCLUSION: Preferred keV level and quality ratings of VMI compared to mixed and Threshold-1 images varied by anatomical location.

Quantitative CT Metrics Associated with Variability in the Diffusion Capacity of the Lung of Post-COVID-19 Patients with Minimal Residual Lung Lesions.

(1) Background: A reduction in the diffusion capacity of the lung for carbon monoxide is a prevalent longer-term consequence of COVID-19 infection. In patients who have zero or minimal residual radiological abnormalities in the lungs, it has been debated whether the cause was mainly due to a reduced alveolar volume or involved diffuse interstitial or vascular abnormalities. (2) Methods: We performed a cross-sectional study of 45 patients with either zero or minimal residual lesions in the lungs (total volume < 7 cc) at two months to one year post COVID-19 infection. There was considerable variability in the diffusion capacity of the lung for carbon monoxide, with 27% of the patients at less than 80% of the predicted reference. We investigated a set of independent variables that may affect the diffusion capacity of the lung, including demographic, pulmonary physiology and CT (computed tomography)-derived variables of vascular volume, parenchymal density and residual lesion volume. (3) Results: The leading three variables that contributed to the variability in the diffusion capacity of the lung for carbon monoxide were the alveolar volume, determined via pulmonary function tests, the blood vessel volume fraction, determined via CT, and the parenchymal radiodensity, also determined via CT. These factors explained 49% of the variance of the diffusion capacity, with p values of 0.031, 0.005 and 0.018, respectively, after adjusting for confounders. A multiple-regression model combining these three variables fit the measured values of the diffusion capacity, with R = 0.70 and p < 0.001. (4) Conclusions: The results are consistent with the notion that in some post-COVID-19 patients, after their pulmonary lesions resolve, diffuse changes in the vascular and parenchymal structures, in addition to a low alveolar volume, could be contributors to a lingering low diffusion capacity.

CT radiomics for differentiating fat poor angiomyolipoma from clear cell renal cell carcinoma: Systematic review and meta-analysis.

PURPOSE: Differentiation of fat-poor angiomyolipoma (fp-AMLs) from renal cell carcinoma (RCC) is often not possible from just visual interpretation of conventional cross-sectional imaging, typically requiring biopsy or surgery for diagnostic confirmation. However, radiomics has the potential to characterize renal masses without the need for invasive procedures. Here, we conducted a systematic review on the accuracy of CT radiomics in distinguishing fp-AMLs from RCCs. METHODS: We conducted a search using PubMed/MEDLINE, Google Scholar, Cochrane Library, Embase, and Web of Science for studies published from January 2011-2022 that utilized CT radiomics to discriminate between fp-AMLs and RCCs. A random-effects model was applied for the meta-analysis according to the heterogeneity level. Furthermore, subgroup analyses (group 1: RCCs vs. fp-AML, and group 2: ccRCC vs. fp-AML), and quality assessment were also conducted to explore the possible effect of interstudy differences. To evaluate CT radiomics performance, the pooled sensitivity, specificity, and diagnostic odds ratio (DOR) were assessed. This study is registered with PROSPERO (CRD42022311034). RESULTS: Our literature search identified 10 studies with 1456 lesions in 1437 patients. Pooled sensitivity was 0.779 [95% CI: 0.562-0.907] and 0.817 [95% CI: 0.663-0.910] for groups 1 and 2, respectively. Pooled specificity was 0.933 [95% CI: 0.814-0.978]and 0.926 [95% CI: 0.854-0.964] for groups 1 and 2, respectively. Also, our findings showed higher sensitivity and specificity of 0.858 [95% CI: 0.742-0.927] and 0.886 [95% CI: 0.819-0.930] for detecting ccRCC from fp-AML in the unenhanced phase of CT scan as compared to the corticomedullary and nephrogenic phases of CT scan. CONCLUSION: This study suggested that radiomic features derived from CT has high sensitivity and specificity in differentiating RCCs vs. fp-AML, particularly in detecting ccRCCs vs. fp-AML. Also, an unenhanced CT scan showed the highest specificity and sensitivity as compared to contrast CT scan phases. Differentiating between fp-AML and RCC often is not possible without biopsy or surgery; radiomics has the potential to obviate these invasive procedures due to its high diagnostic accuracy.

Complications after Nephron-sparing Interventions for Renal Tumors: Imaging Findings and Management.

The two primary nephron-sparing interventions for treating renal masses such as renal cell carcinoma are surgical partial nephrectomy (PN) and image-guided percutaneous thermal ablation. Nephron-sparing surgery, such as PN, has been the standard of care for treating many localized renal masses. Although uncommon, complications resulting from PN can range from asymptomatic and mild to symptomatic and life-threatening. These complications include vascular injuries such as hematoma, pseudoaneurysm, arteriovenous fistula, and/or renal ischemia; injury to the collecting system causing urinary leak; infection; and tumor recurrence. The incidence of complications after any nephron-sparing surgery depends on many factors, such as the proximity of the tumor to blood vessels or the collecting system, the skill or experience of the surgeon, and patient-specific factors. More recently, image-guided percutaneous renal ablation has emerged as a safe and effective treatment option for small renal tumors, with comparable oncologic outcomes to those of PN and a low incidence of major complications. Radiologists must be familiar with the imaging findings encountered after these surgical and image-guided procedures, especially those indicative of complications. The authors review cross-sectional imaging characteristics of complications after PN and image-guided thermal ablation of kidney tumors and highlight the respective management strategies, ranging from clinical observation to interventions such as angioembolization or repeat surgery. Work of the U.S. Government published under an exclusive license with the RSNA. Online supplemental material and the slide presentation from the RSNA Annual Meeting are available for this article. Quiz questions for this article are available in the Online Learning Center. See the invited commentary by Chung and Raman in this issue.

PRDM10 RCC: A Birt-Hogg-Dubé-like Syndrome Associated With Lipoma and Highly Penetrant, Aggressive Renal Tumors Morphologically Resembling Type 2 Papillary Renal Cell Carcinoma.

OBJECTIVE: To characterize the clinical manifestations and genetic basis of a familial cancer syndrome in patients with lipomas and Birt-Hogg-Dubé-like clinical manifestations including fibrofolliculomas and trichodiscomas and kidney cancer. METHODS: Genomic analysis of blood and renal tumor DNA was performed. Inheritance pattern, phenotypic manifestations, and clinical and surgical management were documented. Cutaneous, subcutaneous, and renal tumor pathologic features were characterized. RESULTS: Affected individuals were found to be at risk for a highly penetrant and lethal form of bilateral, multifocal papillary renal cell carcinoma. Whole genome sequencing identified a germline pathogenic variant in PRDM10 (c.2029 T>C, p.Cys677Arg), which cosegregated with disease. PRDM10 loss of heterozygosity was identified in kidney tumors. PRDM10 was predicted to abrogate expression of FLCN, a transcriptional target of PRDM10, which was confirmed by tumor expression of GPNMB, a TFE3/TFEB target and downstream biomarker of FLCN loss. In addition, a sporadic papillary RCC from the TCGA cohort was identified with a somatic PRDM10 mutation. CONCLUSION: We identified a germline PRDM10 pathogenic variant in association with a highly penetrant, aggressive form of familial papillary RCC, lipomas, and fibrofolliculomas/trichodiscomas. PRDM10 loss of heterozygosity and elevated GPNMB expression in renal tumors indicate that PRDM10 alteration leads to reduced FLCN expression, driving TFE3-induced tumor formation. These findings suggest that individuals with Birt-Hogg-Dubé-like manifestations and subcutaneous lipomas, but without a germline pathogenic FLCN variant, should be screened for germline PRDM10 variants. Importantly, kidney tumors identified in patients with a pathogenic PRDM10 variant should be managed with surgical resection instead of active surveillance.

Deep-Learning-Based Whole-Lung and Lung-Lesion Quantification Despite Inconsistent Ground Truth: Application to Computerized Tomography in SARS-CoV-2 Nonhuman Primate Models.

RATIONALE AND OBJECTIVES: Animal modeling of infectious diseases such as coronavirus disease 2019 (COVID-19) is important for exploration of natural history, understanding of pathogenesis, and evaluation of countermeasures. Preclinical studies enable rigorous control of experimental conditions as well as pre-exposure baseline and longitudinal measurements, including medical imaging, that are often unavailable in the clinical research setting. Computerized tomography (CT) imaging provides important diagnostic, prognostic, and disease characterization to clinicians and clinical researchers. In that context, automated deep-learning systems for the analysis of CT imaging have been broadly proposed, but their practical utility has been limited. Manual outlining of the ground truth (i.e., lung-lesions) requires accurate distinctions between abnormal and normal tissues that often have vague boundaries and is subject to reader heterogeneity in interpretation. Indeed, this subjectivity is demonstrated as wide inconsistency in manual outlines among experts and from the same expert. The application of deep-learning data-science tools has been less well-evaluated in the preclinical setting, including in nonhuman primate (NHP) models of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection/COVID-19, in which the translation of human-derived deep-learning tools is challenging. The automated segmentation of the whole lung and lung lesions provides a potentially standardized and automated method to detect and quantify disease. MATERIALS AND METHODS: We used deep-learning-based quantification of the whole lung and lung lesions on CT scans of NHPs exposed to SARS-CoV-2. We proposed a novel multi-model ensemble technique to address the inconsistency in the ground truths for deep-learning-based automated segmentation of the whole lung and lung lesions. Multiple models were obtained by training the convolutional neural network (CNN) on different subsets of the training data instead of having a single model using the entire training dataset. Moreover, we employed a feature pyramid network (FPN), a CNN that provides predictions at different resolution levels, enabling the network to predict objects with wide size variations. RESULTS: We achieved an average of 99.4 and 60.2% Dice coefficients for whole-lung and lung-lesion segmentation, respectively. The proposed multi-model FPN outperformed well-accepted methods U-Net (50.5%), V-Net (54.5%), and Inception (53.4%) for the challenging lesion-segmentation task. We show the application of segmentation outputs for longitudinal quantification of lung disease in SARS-CoV-2-exposed and mock-exposed NHPs. CONCLUSION: Deep-learning methods should be optimally characterized for and targeted specifically to preclinical research needs in terms of impact, automation, and dynamic quantification independently from purely clinical applications.

Deep learning-based decision forest for hereditary clear cell renal cell carcinoma segmentation on MRI.

BACKGROUND: von Hippel-Lindau syndrome (VHL) is an autosomal dominant hereditary syndrome with an increased predisposition of developing numerous cysts and tumors, almost exclusively clear cell renal cell carcinoma (ccRCC). Considering the lifelong surveillance in such patients to monitor the disease, patients with VHL are preferentially imaged using MRI to eliminate radiation exposure. PURPOSE: Segmentation of kidney and tumor structures on MRI in VHL patients is useful in lesion characterization (e.g., cyst vs. tumor), volumetric lesion analysis, and tumor growth prediction. However, automated tasks such as ccRCC segmentation on MRI is sparsely studied. We develop segmentation methodology for ccRCC on T1 weighted precontrast, corticomedullary, nephrogenic, and excretory contrast phase MRI. METHODS: We applied a new neural network approache using a novel differentiable decision forest, called hinge forest (HF), to segment kidney parenchyma, cyst, and ccRCC tumors in 117 images from 115 patients. This data set represented an unprecedented 504 ccRCCs with 1171 cystic lesions obtained at five different MRI scanners. The HF architecture was compared with U-Net on 10 randomized splits with 75% used for training and 25% used for testing. Both methods were trained with Adam using default parameters ( α = 0.001 , ß 1 = 0.9 , ß 2 = 0.999 $\alpha = 0.001,\ \beta _1 = 0.9,\ \beta _2 = 0.999$ ) over 1000 epochs. We further demonstrated some interpretability of our HF method by exploiting decision tree structure. RESULTS: The HF achieved an average kidney, cyst, and tumor Dice similarity coefficient (DSC) of 0.75 ± 0.03, 0.44 ± 0.05, 0.53 ± 0.04, respectively, while U-Net achieved an average kidney, cyst, and tumor DSC of 0.78 ± 0.02, 0.41 ± 0.04, 0.46 ± 0.05, respectively. The HF significantly outperformed U-Net on tumors while U-Net significantly outperformed HF when segmenting kidney parenchymas ( α < 0.01 $\alpha < 0.01$ ). CONCLUSIONS: For the task of ccRCC segmentation, the HF can offer better segmentation performance compared to the traditional U-Net architecture. The leaf maps can glean hints about deep learning features that might prove to be useful in other automated tasks such as tumor characterization.

Automatic segmentation of clear cell renal cell tumors, kidney, and cysts in patients with von Hippel-Lindau syndrome using U-net architecture on magnetic resonance images.

We demonstrate automated segmentation of clear cell renal cell carcinomas (ccRCC), cysts, and surrounding normal kidney parenchyma in patients with von Hippel-Lindau (VHL) syndrome using convolutional neural networks (CNN) on Magnetic Resonance Imaging (MRI). We queried 115 VHL patients and 117 scans (3 patients have two separate scans) with 504 ccRCCs and 1171 cysts from 2015 to 2021. Lesions were manually segmented on T1 excretory phase, co-registered on all contrast-enhanced T1 sequences and used to train 2D and 3D U-Net. The U-Net performance was evaluated on 10 randomized splits of the cohort. The models were evaluated using the dice similarity coefficient (DSC). Our 2D U-Net achieved an average ccRCC lesion detection Area under the curve (AUC) of 0.88 and DSC scores of 0.78, 0.40, and 0.46 for segmentation of the kidney, cysts, and tumors, respectively. Our 3D U-Net achieved an average ccRCC lesion detection AUC of 0.79 and DSC scores of 0.67, 0.32, and 0.34 for kidney, cysts, and tumors, respectively. We demonstrated good detection and moderate segmentation results using U-Net for ccRCC on MRI. Automatic detection and segmentation of normal renal parenchyma, cysts, and masses may assist radiologists in quantifying the burden of disease in patients with VHL.

Role of ultra-high b-value DWI in the imaging of hereditary leiomyomatosis and renal cell carcinoma (HLRCC).

PURPOSE: Hereditary leiomyomatosis and renal cell carcinoma (HLRCC) syndrome is associated with an aggressive form of renal cell carcinoma with high risk of metastasis, even in small primary tumors with unequivocal imaging findings. In this study, we compare the performance of ultra-high b-value diffusion-weighted imaging (DWI) sequence (b = 2000 s/mm2) to standard DWI (b = 800 s/mm2) sequence in identifying malignant lesions in patients with HLRCC. METHODS: Twenty-eight patients (n = 18 HLRCC patients with 22 lesions, n = 10 controls) were independently evaluated by three abdominal radiologists with different levels of experience using four combinations of MRI sequences in two separate sessions (session 1: DWI with b-800, session 2: DWI with b-2000). T1 precontrast, T2-weighted (T2WI), and apparent diffusion coefficient (ADC) sequences were similar in both sessions. Each identified lesion was subjectively assessed using a six-point cancer likelihood score based on individual sequences and overall impression. RESULTS: The ability to distinguish benign versus malignant renal lesions improved with the use of b-2000 for more experienced radiologists (Reader 1 AUC: Session 1-0.649 and Session 2-0.938, p = 0.017; Reader 2 AUC: Session 1-0.781 and Session 2-0.921, p = 0.157); whereas no improvement was observed for the less experienced reader (AUC: Session 1-0.541 and Session 2-0.607, p = 0.699). CONCLUSION: The inclusion of ultra-high b-value DWI sequence improved the ability of classification of renal lesions in patients with HLRCC for experienced radiologists. Consideration should be given toward incorporation of DWI with b-2000 s/mm2 into existing renal MRI protocols.