MRI texture analysis of acetabular cancellous bone can discriminate between normal, cam positive, and cam-FAI hips.

OBJECTIVES: To compare the MRI texture profile of acetabular subchondral bone in normal, asymptomatic cam positive, and symptomatic cam-FAI hips and determine the accuracy of a machine learning model for discriminating between the three hip classes. METHODS: A case-control, retrospective study was performed including 68 subjects (19 normal, 26 asymptomatic cam, 23 symptomatic cam-FAI). Acetabular subchondral bone of unilateral hip was contoured on 1.5 T MR images. Nine first-order 3D histogram and 16 s-order texture features were evaluated using specialized texture analysis software. Between-group differences were assessed using Kruskal-Wallis and Mann-Whitney U tests, and differences in proportions compared using chi-square and Fisher's exact tests. Gradient-boosted ensemble methods of decision trees were created and trained to discriminate between the three groups of hips, with percent accuracy calculated. RESULTS: Sixty-eight subjects (median age 32 (28-40), 60 male) were evaluated. Significant differences among all three groups were identified with first-order (4 features, all p ≤ 0.002) and second-order (11 features, all p ≤ 0.002) texture analyses. First-order texture analysis could differentiate between control and cam positive hip groups (4 features, all p ≤ 0.002). Second-order texture analysis could additionally differentiate between asymptomatic cam and symptomatic cam-FAI groups (10 features, all p ≤ 0.02). Machine learning models demonstrated high classification accuracy of 79% (SD 16) for discriminating among all three groups. CONCLUSION: Normal, asymptomatic cam positive, and cam-FAI hips can be discriminated based on their MRI texture profile of subchondral bone using descriptive statistics and machine learning algorithms. CLINICAL RELEVANCE STATEMENT: Texture analysis can be performed on routine MR images of the hip and used to identify early changes in bone architecture, differentiating morphologically abnormal from normal hips, prior to onset of symptoms. KEY POINTS: • MRI texture analysis is a technique for extracting quantitative data from routine MRI images. • MRI texture analysis demonstrates that there are different bone profiles between normal hips and those with femoroacetabular impingement. • Machine learning models can be used in conjunction with MRI texture analysis to accurately differentiate between normal hips and those with femoroacetabular impingement.

Intraoperative Channeling in Arthroscopic Rotator Cuff Repair: A Multicenter Randomized Controlled Trial.

BACKGROUND: Despite recent advances in arthroscopic rotator cuff repair, the retear rate remains high. New methods to optimize healing rates must be sought. Bone channeling may create a quicker and more vigorous healing response by attracting autologous mesenchymal stem cells, cytokines, and growth factors to the repair site. HYPOTHESIS: Arthroscopic rotator cuff repair with bone channeling would result in a higher healing rate compared with arthroscopic rotator cuff repair without adjuvant channeling. STUDY DESIGN: Randomized controlled trial; Level of evidence, 1. METHODS: Our primary objective was to compare healing rates in patients undergoing arthroscopic rotator cuff repair for degenerative tears, with and without bone channeling. Secondary objectives included comparisons of the Western Ontario Rotator Cuff Index (WORC) score, American Shoulder and Elbow Surgeons (ASES) score, Constant score, Constant strength subscore, and visual analog scale (VAS) for pain score between groups. Patients undergoing arthroscopic rotator cuff repair were recruited at 3 sites and were randomized to receive either bone channeling augmentation or standard repair. Healing was determined via ultrasound at 24 months postoperatively. WORC, ASES, and Constant scores were compared between groups at baseline and at 3, 6, 12, and 24 months postoperatively. RESULTS: A total of 168 patients were enrolled between 2013 and 2018. Intention-to-treat analysis revealed no statistical differences in healing rates between the 2 interventions at 24 months postoperatively. Statistically significant improvements occurred in both groups from preoperatively to all time points for the WORC, the ASES score, the Constant score or Constant strength subscore, and the VAS for pain (P < .0001). No differences were observed between the bone channeling and control groups in WORC, ASES, Constant, and VAS pain scores at any time point. CONCLUSION: This trial did not demonstrate the superiority of intraoperative bone channeling in rotator cuff repair over standard rotator cuff repair at 24 months postoperatively. Healing rates, patient-reported function, and quality-of-life outcomes were similar between groups. REGISTRATION: NCT01877772 (ClinicalTrials.gov identifier).

Ultrasound-determined healing rates with subscapularis tenotomy versus peel after anatomic shoulder arthroplasty.

BACKGROUND: Several techniques have been described for mobilizing the subscapularis tendon in anatomic total shoulder arthroplasty (TSA). The purpose of this study was to compare subscapularis tendon healing rates, as determined by ultrasound, in patients following anatomic TSA with either a subscapularis tenotomy or subscapularis peel. METHODS: This study was a secondary analysis of patients from a previous randomized controlled trial in which patients underwent anatomic TSA and were randomized to either a tenotomy or peel approach. The primary outcome was postoperative tendon healing rates determined on ultrasound at >12 months after surgery. Secondary outcomes included postoperative tendon thickness measured on ultrasound; elbow position (neutral alignment in the belly-press position vs. posterior); internal rotation function measured with the third and fourth questions of the American Shoulder and Elbow Surgeons questionnaire; and Western Ontario Osteoarthritis of the Shoulder index. Radiographs were analyzed in patients with torn tendons. RESULTS: One hundred patients were randomized to a tenotomy (n = 47) or peel (n = 53) approach. Postoperative ultrasound results were available in 88 patients. Tendon healing rates were 95% for tenotomy vs. 75% for peel (P = .011). The mean postoperative tendon thickness was 4 mm (standard deviation, 1.0 mm) and 4 mm (standard deviation, 1 mm) in the tenotomy and peel groups, respectively (P = .37). Internal rotation function was not associated with healing status (P = .77 and P = .22 for questions 3 and 4, respectively, of the American Shoulder and Elbow Surgeons questionnaire), nor was elbow position (P = .2) in the belly-press position. DISCUSSION: We observed that subscapularis tenotomy had a higher healing rate than peel as determined by ultrasound in TSA patients. There was no statistically significant difference in postoperative tendon thickness in intact tendons as measured on ultrasound when comparing subscapularis mobilization techniques, nor was there any association between healing status and internal rotation function or elbow position.

CT texture analysis of acetabular subchondral bone can discriminate between normal and cam-positive hips.

OBJECTIVES: The purpose of this study was to determine if the CT texture profile of acetabular subchondral bone differs between normal, asymptomatic cam-positive, and symptomatic cam-FAI hips. In addition, the utility of texture analysis to discriminate between the three hip statuses was explored using a machine learning approach. METHODS: IRB-approved, case-control study analyzing CT images in subjects with and without cam morphology from August 2010 to December 2013. Sixty-eight subjects were included: 19 normal controls, 26 asymptomatic cam, and 23 symptomatic cam-FAI. Acetabular subchondral bone was contoured on the sagittal oblique CT images using ImageJ ®. 3D histogram texture features (mean, variance, skewness, kurtosis, and percentiles) were evaluated using MaZda software. Groupwise differences were investigated using Kruskal-Wallis tests and Mann-Whitney U tests. Gradient-boosted decision trees were created and trained to discriminate between control and cam-positive hips. RESULTS: Both asymptomatic and symptomatic cam-FAI hips demonstrated significantly higher values of texture variance (p = 0.0007, p < 0.0001), 90th percentile (p = 0.007, p = 0.006), and 99th percentile (p = 0.009, p = 0.009), but significantly lower values of skewness (p = 0.0001, p = 0.0013) and kurtosis (p = 0.0001, p = 0.0001) compared to normal controls. There were no differences in texture profile between asymptomatic cam and symptomatic cam-FAI hips. Machine learning models demonstrated high classification accuracy for discriminating control hips from asymptomatic cam-positive (82%) and symptomatic cam-FAI (86%) hips. CONCLUSIONS: Texture analysis can discriminate between normal and cam-positive hips using conventional descriptive statistics, regression modeling, and machine learning algorithms. It has the potential to become an important tool in compositional analysis of hip subchondral trabecular bone in the context of FAI, and possibly serve as a biomarker of joint degeneration. KEY POINTS: • The CT texture profile of acetabular subchondral bone is significantly different between normal and cam-positive hips. • Texture analysis can detect changes in subchondral bone in asymptomatic cam-positive hips that are equal to that of symptomatic cam-FAI hips. • Texture analysis has the potential to become an important tool in compositional analysis of hip subchondral bone in the context of FAI and may serve as a biomarker in the study of joint physiology and biomechanics.

Logistics of Three-dimensional Printing: Primer for Radiologists.

The Association of University Radiologists Radiology Research Alliance Task Force on three-dimensional (3D) printing presents a review of the logistic considerations for establishing a clinical service using this new technology, specifically focused on implications for radiology. Specific topics include printer selection for 3D printing, software selection, creating a 3D model for printing, providing a 3D printing service, research directions, and opportunities for radiologists to be involved in 3D printing. A thorough understanding of the technology and its capabilities is necessary as the field of 3D printing continues to grow. Radiologists are in the unique position to guide this emerging technology and its use in the clinical arena.

Clinical Applications of 3D Printing: Primer for Radiologists.

Three-dimensional (3D) printing refers to a number of manufacturing technologies that create physical models from digital information. Radiology is poised to advance the application of 3D printing in health care because our specialty has an established history of acquiring and managing the digital information needed to create such models. The 3D Printing Task Force of the Radiology Research Alliance presents a review of the clinical applications of this burgeoning technology, with a focus on the opportunities for radiology. Topics include uses for treatment planning, medical education, and procedural simulation, as well as patient education. Challenges for creating custom implantable devices including financial and regulatory processes for clinical application are reviewed. Precedent procedures that may translate to this new technology are discussed. The task force identifies research opportunities needed to document the value of 3D printing as it relates to patient care.

Preoperative planning and tracheal stent design in thoracic surgery: a primer for the 2017 Radiological Society of North America (RSNA) hands-on course in 3D printing.

In this work, we provide specific clinical examples to demonstrate basic practical techniques involved in image segmentation, computer-aided design, and 3D printing. A step-by-step approach using United States Food and Drug Administration cleared software is provided to enhance surgical intervention in a patient with a complex superior sulcus tumor. Furthermore, patient-specific device creation is demonstrated using dedicated computer-aided design software. Relevant anatomy for these tasks is obtained from CT Digital Imaging and Communications in Medicine images, leading to the generation of 3D printable files and delivery of these files to a 3D printer.

MRI evaluation of small (<4cm) solid renal masses: multivariate modeling improves diagnostic accuracy for angiomyolipoma without visible fat compared to univariate analysis.

OBJECTIVE: To assess MRI for diagnosis of angiomyolipoma without visible fat (AMLwvf). MATERIAL AND METHODS: With IRB approval, a retrospective study in consecutive patients with contrast-enhanced (CE)-MRI and <4 cm solid renal masses from 2002-2013 was performed. Ten AMLwvf were compared to 77 RCC; 33 clear cell (cc), 35 papillary (p), 9 chromophobe (ch). A blinded radiologist measured T2W signal-intensity ratio (SIR), chemical-shift (CS) SI-index and area under CE-MRI curve (CE-AUC). Regression modeling and ROC analysis was performed. RESULTS: T2W-SIR was lower in AMLwvf (0.64 ± 0.12) compared to cc-RCC (1.37 ± 0.30, p < 0.001), ch-RCC (0.94 ± 0.19, p = 0.005) but not p-RCC (0.74 ± 0.17, p = 0.2). CS-SI index was higher in AMLwvf (16.1 ± 31.5 %) compared to p-RCC (-5.2 ± 26.1 %, p = 0.02) but not ch-RCC (3.0 ± 12.5 %, p = 0.1) or cc-RCC (7.7 ± 17.9 %,p = 0.1). CE-AUC was higher in AMLwvf (515.7 ± 144.7) compared to p-RCC (154.5 ± 92.8, p < 0.001) but not ch-RCC (341.5 ± 202.7, p = 0.07) or cc-RCC (520.9 ± 276.9, p = 0.95). Univariate ROC-AUC were: T2SIR = 0.86 (CI 0.77-0.96); CE-AUC = 0.76 (CI 0.65-0.87); CS-SI index = 0.66 (CI 0.4.3-0.85). Logistic regression models improved ROC-AUC, A) T2 SIR + CE-AUC = 0.97 (CI 0.93-1.0) and T2 SIR + CS-SI index = 0.92 (CI 0.84-0.99) compared to univariate analyses (p < 0.05). The optimal sensitivity/specificity of T2SIR + CE-AUC and T2SIR + CS-SI index were 100/88.8 % and 60/97.4 %. CONCLUSION: MRI, using multi-variate modelling, is accurate for diagnosis of AMLwvf. KEY POINTS: • AMLwvf are difficult to prospectively diagnose with imaging. • MRI findings associated with AMLwvf overlap with various RCC subtypes. • T2W-SI combined with chemical-shift SI-index is specific for AMLwvf but lacks sensitivity. • T2W-SI combined with AUC CE-MRI is sensitive and specific for AMLwvf. • Models incorporating two or more findings are more accurate than univariate analysis.

Angiomyolipoma (AML) without visible fat: Ultrasound, CT and MR imaging features with pathological correlation.

OBJECTIVE: To compare imaging findings with histopathology in AML without visible fat (AMLwvf). MATERIAL AND METHODS: With IRB approval, we identified 18 AMLwvf that underwent CT between 2002-2014. A radiologist measured NECT-attenuation, corticomedullary (CM) and nephrographic (NG) enhancement, echogenicity relative to renal cortex (RC) (N = 5), T2W (T2AML/T2RC) signal-intensity (SI), and chemical-shift SI ([SIIN-PHASE - SIOPPOSED-PHASE]/SIIN-PHASE) indices (N = 6). A pathologist re-evaluated 15/18 AMLwvf for 1) < or > 25% adipocytes/high-power-field (HPF), 2) "many or few" blood vessels. Comparisons were performed using chi-square and independent t-tests. RESULTS: 73.3%(11/15) of AMLwvf had <25% adipocytes/HPF and 86.7%(13/15) had "many" blood vessels. NECT-attenuation was 41.8(±6.9) HU. 61.1 %(11/18) of AMLwvf were hyper-attenuating and 38.9%(7/18) iso-attenuating; attenuation was associated with %-adipocytes/HPF, (p = 0.01). CM/NG enhancement were 63.3(±20.8)/51.7(±15.5) HU. 72.2%(13/18) of AMLwvf had wash-out enhancement, with no association with amount of blood vessels at pathology, (p = 0.68). No difference in echogenicity was noted by histology (p > 0.05). All AMLwvf were T2-hypointense (SI ratio = 0.61 [±0.1]). 2/6 AMLwvf showed SI drop on chemical-shift MRI; both were iso-attenuating and were associated with >25% adipocytes/HPF (p = 0.04). CONCLUSIONS: AMLwvf are typically T2-hypointense and hyper-attenuating with wash-out enhancement due to abundant smooth muscle and vessels respectively. Iso-attenuating AMLwvf with microscopic fat on MRI contain more adipocytes/HPF. KEY POINTS: • Five percent of AML do not demonstrate detectable fat on imaging • These AML are hyperattenuating and T2-hypointense due to abundant smooth muscle • These AML show washout enhancement without association to vessel count at histopathology • Iso-attenuating AML with microscopic fat on MRI show >25% adipocytes/HPF • The term "AML without visible fat" is proposed to reduce ambiguity.

Solid Renal Cell Carcinoma Measuring Water Attenuation (-10 to 20 HU) on Unenhanced CT.

OBJECTIVE: The objectives of this study were to evaluate the incidence of solid renal cell carcinoma (RCC) with attenuation similar to that of water (-10 to 20 HU) on unenhanced CT and to examine imaging features that can allow RCC to be differentiated from simple cysts. MATERIALS AND METHODS: We performed an enriched quantitative and qualitative retrospective analysis of consecutive patients who underwent unenhanced CT before resection of solid (noncystic and nonnecrotic) renal masses measuring < 5 cm from 2008 to 2013. In all, 93 patients with 96 tumors (55 clear cell, 27 papillary, and 14 chromophobe) were evaluated with unenhanced CT. Attenuation was measured at three standardized levels and compared using the Kruskal-Wallis test. Two radiologists independently assessed calcification, margin (smooth or irregular), and heterogeneity (homogeneous or heterogeneous). Results were compared using the chi-square test with Bonferroni correction. RESULTS: Mean ± SD attenuation values were 31.8 ± 9.6 HU (range, 11-63 HU) overall, 29.9 ± 8.8 HU (range, 11.0-49.0 HU) for clear cell tumors, 34.6 ± 10.3 HU (range, 20.3-63.0 HU) for papillary tumors, and 35.5 ± 9.2 HU (range, 20.7-47 HU) for chromophobe tumors (p = 0.06). Eight clear cell RCCs had attenuation similar to that of water (15.7 ± 2.4 HU; range, 11-18.7 HU). There was no significant difference in calcification or margin among different types of tumors (p = 0.91 and p = 0.55, respectively). Chromophobe tumors were more likely to be homogeneous (p < 0.001). Interobserver agreement was moderate to very good (κ = 0.91 for calcification, κ = 0.55 for margin, and κ = 0.44 for heterogeneity). All eight clear cell RCCs with attenuation similar to that of water were considered heterogeneous by both readers. Irregular margins were identified in three of these eight tumors by reader 1 and four of eight tumors by reader 2. CONCLUSION: A minority of solid RCCs have attenuation similar to that of water on unenhanced CT. In this study, all such tumors were of the clear cell subtype and qualitatively heterogeneous, suggesting they can be distinguished from simple cysts on unenhanced CT.

Can Quantitative CT Texture Analysis be Used to Differentiate Fat-poor Renal Angiomyolipoma from Renal Cell Carcinoma on Unenhanced CT Images?

PURPOSE: To determine the accuracy of texture analysis to differentiate fat-poor angiomyolipoma (fp-AML) from renal cell carcinoma (RCC) on unenhanced computed tomography (CT) images. MATERIALS AND METHODS: In this institutional review board-approved retrospective case-control study, patients with AML and RCC were identified from the pathology database: there were 16 patients with fp-AML (no visible fat at unenhanced CT) and 84 patients with RCC. Axial unenhanced CT images were contoured manually by two independent analysts. Texture analysis was performed for each lesion, and reproducibility was assessed. Texture features related to the gray-level histogram, gray-level co-occurrence, and run-length matrix statistics were evaluated. The most discriminative features were used to generate support vector machine (SVM) classifiers. Diagnostic accuracy of textural features was assessed and 10-fold cross validation was performed. Unenhanced CT images for each patient were independently reviewed by two blinded radiologists who subjectively graded lesion heterogeneity on a five-point scale. Differences in area under the receiver operating characteristic curve (AUC) between subjective heterogeneity ratings and textural features were evaluated by using the DeLong method. RESULTS: There was lower lesion homogeneity and higher lesion entropy in RCCs (P ≤ .01). A model incorporating several texture features resulted in an AUC of 0.89 ± 0.04. The average SVM accuracy of textural features ranged from 83% to 91% (after 10-fold cross validation). An optimal subjective heterogeneity rating of 2 or higher was identified as a predictor of RCC for both readers, with no significant difference in AUC between readers (P = .06). Each of the three textural-based classifiers was more accurate than either radiologists' subjective heterogeneity ratings for the models incorporating a subset of the top three textural features (difference in AUC between textural features and subjective visual heterogeneity, 0.25; 95% confidence interval: 0.02, 0.47; P = .03). CONCLUSION: CT texture analysis can be used to accurately differentiate fp-AML from RCC on unenhanced CT images.

Medical 3D printing for vascular interventions and surgical oncology: a primer for the 2016 radiological society of North America (RSNA) hands-on course in 3D printing.

Medical 3D printing holds the potential of transforming personalized medicine by enabling the fabrication of patient-specific implants, reimagining prostheses, developing surgical guides to expedite and transform surgical interventions, and enabling a growing multitude of specialized applications. In order to realize this tremendous potential in frontline medicine, an understanding of the basic principles of 3D printing by the medical professionals is required. This primer underlines the basic approaches and tools in 3D printing, starting from patient anatomy acquired through cross-sectional imaging, in this case Computed Tomography (CT). We describe the basic principles using the relatively simple task of separation of the relevant anatomy to guide aneurysm repair. This is followed by exploration of more advanced techniques in the creation of patient-specific surgical guides and prostheses for a patient with extensive pleomorphic sarcoma using Computer Aided Design (CAD) software.

Unenhanced CT for the diagnosis of minimal-fat renal angiomyolipoma.

OBJECTIVE: Minimal-fat angiomyolipoma resembles renal cell carcinoma (RCC) on imaging. The purpose of this study was to evaluate the diagnostic accuracy of unenhanced CT to diagnose minimal-fat angiomyolipoma. MATERIALS AND METHODS: Consecutive patients who underwent unenhanced CT before resection of a solid renal mass from 2002 to 2012 were included in this study. Sixteen patients with minimal-fat angiomyolipoma and 48 patients with RCC (18 clear cell, 17 papillary, and 13 chromophobe RCCs) were studied. The mean (±SD) age of patients was 53.4±11.7 years for minimal-fat angiomyolipoma and 56.4±13.2 years for RCC. There were five male patients in the minimal-fat angiomyolipoma group and 26 male patients in the RCC group. Absolute density and relative density ratios were calculated and compared by the Kruskal-Wallis test and univariate regression analysis. Two radiologists, blinded to the diagnosis, independently assessed for coexisting classic angiomyolipoma, calcification within the mass, and the "hypodense rim" sign (defined as a low-density rim at the interface of the tumor and normal kidney). Radiologists evaluated both the axial and the coronal reformatted images. Diagnostic accuracy was compared using the Fisher exact test. RESULTS: Mean attenuation of minimal-fat angiomyolipoma was 43.1±9.8 HU, which was significantly higher when compared with RCC overall (p=0.004) as well as with clear cell RCC (33.1±10.2 HU; p=0.003), papillary RCC (33.0±6.6 HU; p=0.003), and chromophobe RCC (34.3±9.9 HU; p=0.01). The density ratio of minimal-fat angiomyolipoma to normal kidney was higher when compared with RCC overall (p=0.002) and the respective RCC subtypes (p<0.001, p=0.006, and p=0.002). The hypodense rim sign was identified in three patients with minimal-fat angiomyolipoma and five patients with RCC by both radiologists. There was no difference in the rate of detection of the hypodense rim sign in minimal-fat angiomyolipoma compared with RCC (p=0.14), and interobserver variability was fair (κ=0.32). Classic angiomyolipomas were identified only in patients with minimal-fat angiomyolipoma (p=0.003). Calcification was not detected in any minimal-fat angiomyolipoma. CONCLUSION: Coexisting classic angiomyolipomas and the absence of calcification are associated with minimal-fat angiomyolipomas; by contrast, the hypodense rim sign is not useful for diagnosis. Minimal-fat angiomyolipomas have increased absolute and relative (normalized to renal parenchyma) density compared with RCC; however, overlap in density values limit diagnostic utility.