Detection of metallic objects on digital radiographs with convolutional neural networks: A MRI screening tool

IntroductionScreening for metallic implants and foreign bodies before magnetic resonance imaging (MRI) examinations, are crucial for patient safety. History of health are supplied by the patient, a family member, screening of electronic health records or the picture and archive systems (PACS). PACS securely store and transmits digital radiographs (DR) and related reports with patient information. Convolutional neural networks (CNN) can be used to detect metallic objects in DRs stored in PACS. This study evaluates the accuracy of CNNs in the detection of metallic objects on DRs as an MRI screening tool.MethodsThe musculoskeletal radiographs (MURA) dataset consisting of 14.863 upper extremity studies were stratified into datasets with and without metal. For each anatomical region: Elbow, finger, hand, humerus, forearm, shoulder and wrist we trained and validated CNN algorithms to classify radiographs with and without metal. Algorithm performance was evaluated with area under the receiver-operating curve (AUC), sensitivity, specificity, predictive values and accuracies compared with a reference standard of manually labelling.ResultsSensitivities, specificities and area under the ROC-curves (AUC) for the six anatomic regions ranged from 85.33% (95% CI: 78.64%–90.57%) to 100.00% (95% CI: 98.16%–100.00%), 75.44% (95% CI: 62.24%–85.87%) to 93.57% (95% CI: 88.78%–96.75%) and 0.95 to 0.99, respectively.ConclusionCNN algorithms classify DRs with metallic objects for six different anatomic regions with near-perfect accuracy. The rapid and iterative capability of the algorithms allows for scalable expansion and as a substitute MRI screening tool for metallic objects.Implications for practiceAll CNNs would be able to assist in metal detection of digital radiographs prior to MRI, an substantially decrease screening time.     

Deep learning with convolutional neural network for estimation of the characterisation of coronary plaques: Validation using IB-IVUS

IntroductionDeep learning approaches have shown high diagnostic performance in image classifications, such as differentiation of malignant tumors and calcified coronary plaque. However, it is unknown whether deep learning is useful for characterizing coronary plaques without the presence of calcification using coronary computed tomography angiography (CCTA). The purpose of this study was to compare the diagnostic performance of deep learning with a convolutional neural network (CNN) with that of radiologists in the estimation of coronary plaques.MethodsWe retrospectively enrolled 178 patients (191 coronary plaques) who had undergone CCTA and integrated backscatter intravascular ultrasonography (IB-IVUS) studies. IB-IVUS diagnosed 81 fibrous and 110 fatty or fibro-fatty plaques. We manually captured vascular short-axis images of the coronary plaques as Portable Network Graphics (PNG) images (150 × 150 pixels). The display window level and width were 100 and 700 Hounsfield units (HU), respectively. The deep-learning system (CNN; GoogleNet Inception v3) was trained on 153 plaques; its performance was tested on 38 plaques. The area under the curve (AUC) obtained by receiver operating characteristic analysis of the deep learning system and by two board-certified radiologists was compared.ResultsWith the CNN, the AUC and the 95% confidence interval were 0.83 and 0.69–0.96, respectively; for radiologist 1 they were 0.61 and 0.42–0.80; for radiologist 2 they were 0.68 and 0.51–0.86, respectively. The AUC for CNN was significantly higher than for radiologists 1 (p = 0.04); for radiologist 2 it was not significantly different (p = 0.22).ConclusionDL-CNN performed comparably to radiologists for discrimination between fatty and fibro-fatty plaque on CCTA images.Implications for practiceThe diagnostic performance of the CNN and of two radiologists in the assessment of 191 ROIs on CT images of coronary plaques whose type corresponded with their IB-IVUS characterization was comparable.     

The effect of different statistical approaches on image quality data obtained from radiological examinations

IntroductionSelection of optimal image acquisition protocols in medical imaging remains a grey area, the superimposed use of the Likert scale in radiological image quality evaluations creates an additional challenge for the statistical analysis of image quality data.Using a simulation study, we have trialled a novel approach to analysing radiological image quality Likert scale data.MethodsA simulation study was undertaken where simulated datasets were generated based on the distribution of Likert scale values according to varying image acquisition protocols from a real dataset. Simulated Likert scale values were pooled in four different ways; the mean, median, mode and the summation of patient Likert scale values of which the total was assigned a categorical Likert scale value. Estimates of bias, MAPE and RMSPE were then calculated for all four pooling approaches to determine which method most accurately represented an expert's opinion.ResultsWhen compared to an expert's opinion, the method of summation and categorisation of Likert scale values was most accurate 49 times out of the 114 (43.0%) tests. The mean 28 times out of 114 (24.6%), the median 23 times out of 114 (20.2%) and the mode 17 times out of 114 (14.9%).ConclusionWe conclude that our method of summation and categorisation of Likert scale values is most often the best representation of the simulated data compared to the expert's opinion.Implications for practiceThere is scope to reproduce this simulation study with multiple observers to reflect clinical reality more accurately with the dynamic nature of multiple observers. This also prompts future investigation into other anatomical areas, to see if the same methods produce similar results.     

Maskless 2-Dimensional Digital Subtraction Angiography Generation Model for Abdominal Vasculature using Deep Learning

PurposeTo develop a deep learning (DL) model to generate synthetic, 2-dimensional subtraction angiograms free of artifacts from native abdominal angiograms.Materials and MethodsIn this retrospective study, 2-dimensional digital subtraction angiography (2D-DSA) images and native angiograms were consecutively collected from July 2019 to March 2020. Images were divided into motion-free (training, validation, and motion-free test datasets) and motion-artifact (motion-artifact test dataset) sets. A total of 3,185, 393, 383, and 345 images from 87 patients (mean age, 71 years ± 10; 64 men and 23 women) were included in the training, validation, motion-free, and motion-artifact test datasets, respectively. Native angiograms and 2D-DSA image pairs were used to train and validate an image-to-image translation model to generate synthetic DL-based subtraction angiography (DLSA) images. DLSA images were quantitatively evaluated by the peak signal-to-noise ratio (PSNR) and structural similarity (SSIM) using the motion-free dataset and were qualitatively evaluated via visual assessments by radiologists with a numerical rating scale using the motion-artifact dataset.ResultsThe DLSA images showed a mean PSNR (± standard deviation) of 43.05 dB ± 3.65 and mean SSIM of 0.98 ± 0.01, indicating high agreement with the original 2D-DSA images in the motion-free dataset. Qualitative visual evaluation by radiologists of the motion-artifact dataset showed that DLSA images contained fewer motion artifacts than 2D-DSA images. Additionally, DLSA images scored similar to or higher than 2D-DSA images for vascular visualization and clinical usefulness.ConclusionsThe developed DL model generated synthetic, motion-free subtraction images from abdominal angiograms with similar imaging characteristics to 2D-DSA images.     

Validation of deep-learning image reconstruction for coronary computed tomography angiography: Impact on noise,image quality and diagnostic accuracy

BackgroundAdvances in image reconstruction are necessary to decrease radiation exposure from coronary CT angiography (CCTA) further, but iterative reconstruction has been shown to degrade image quality at high levels. Deep-learning image reconstruction (DLIR) offers unique opportunities to overcome these limitations. The present study compared the impact of DLIR and adaptive statistical iterative reconstruction-Veo (ASiR-V) on quantitative and qualitative image parameters and the diagnostic accuracy of CCTA using invasive coronary angiography (ICA) as the standard of reference.MethodsThis retrospective study includes 43 patients who underwent clinically indicated CCTA and ICA. Datasets were reconstructed with ASiR-V 70% (using standard [SD] and high-definition [HD] kernels) and with DLIR at different levels (i.e., medium [M] and high [H]). Image noise, image quality, and coronary luminal narrowing were evaluated by three blinded readers. Diagnostic accuracy was compared against ICA.ResultsNoise did not significantly differ between ASiR-V SD and DLIR-M (37 vs. 37 HU, p = 1.000), but was significantly lower in DLIR-H (30 HU, p < 0.001) and higher in ASiR-V HD (53 HU, p < 0.001). Image quality was higher for DLIR-M and DLIR-H (3.4–3.8 and 4.2–4.6) compared to ASiR-V SD and HD (2.1–2.7 and 1.8–2.2; p < 0.001), with DLIR-H yielding the highest image quality. Consistently across readers, no significant differences in sensitivity (88% vs. 92%; p = 0.453), specificity (73% vs. 73%; p = 0.583) and diagnostic accuracy (80% vs. 82%; p = 0.366) were found between ASiR-V HD and DLIR-H.ConclusionDLIR significantly reduces noise in CCTA compared to ASiR-V, while yielding superior image quality at equal diagnostic accuracy.     

Deep Learning for Automated Classification of Inferior Vena Cava Filter Types on Radiographs

PurposeTo demonstrate the feasibility and evaluate the performance of a deep-learning convolutional neural network (CNN) classification model for automated identification of different types of inferior vena cava (IVC) filters on radiographs.Materials and MethodsIn total, 1,375 cropped radiographic images of 14 types of IVC filters were collected from patients enrolled in a single-center IVC filter registry, with 139 images withheld as a test set and the remainder used to train and validate the classification model. Image brightness, contrast, intensity, and rotation were varied to augment the training set. A 50-layer ResNet architecture with fixed pre-trained weights was trained using a soft margin loss over 50 epochs. The final model was evaluated on the test set.ResultsThe CNN classification model achieved a F1 score of 0.97 (0.92–0.99) for the test set overall and of 1.00 for 10 of 14 individual filter types. Of the 139 test set images, 4 (2.9%) were misidentified, all mistaken for other filter types that appear highly similar. Heat maps elucidated salient features for each filter type that the model used for class prediction.ConclusionsA CNN classification model was successfully developed to identify 14 types of IVC filters on radiographs and demonstrated high performance. Further refinement and testing of the model is necessary before potential real-world application.     

Comparing Radiomics features of tumour and healthy liver tissue in a limited CT dataset: A machine learning study

IntroductionLiver cancer lesions on Computed Tomography (CT) withholds a great amount of data, which is not visible to the radiologists and radiographer. Radiomics features can be extracted from the lesions and used to train Machine Learning (ML) algorithms to predict between tumour and liver tissue. The purpose of this study was to investigate and classify Radiomics features extracted from liver tumours and normal liver tissue in a limited CT dataset.MethodsThe Liver Tumour Segmentation Benchmark (LiTS) dataset consisting of 131 CT scans of the liver with segmentations of tumour tissue and healthy liver was used to extract Radiomic features. Extracted Radiomic features included size, shape, and location extracted with morphological and statistical techniques according to the International Symposium on Biomedical Imaging manual. Relevant features was selected with chi2 correlation and principal component analysis (PCA) with tumour and healthy liver tissue as outcome according to a consensus between three experienced radiologists. Logistic regression, random forest and support vector machine was used to train and validate the dataset with a 10-fold cross-validation method and the Grid Search as hyper-parameter tuning. Performance was evaluated with sensitivity, specificity and accuracy.ResultsThe performance of the ML algorithms achieved sensitivities, specificities and accuracy ranging from 96.30% (95% CI: 81.03%–99.91%) to 100.00% (95% CI: 86.77%–100.00%), 91.30% (95% CI: 71.96%–98.93%) to 100.00% (95% CI: 83.89%–100.00%)and 94.00% (95% CI: 83.45%–98.75%) to 100.00% (95% CI: 92.45%–100.00%), respectively.ConclusionML algorithms classifies Radiomics features extracted from healthy liver and tumour tissue with perfect accuracy. The Radiomics signature allows for a prognostic biomarker for hepatic tumour screening on liver CT.Implications for practiceDifferentiation between tumour and liver tissue with Radiomics ML algorithms have the potential to increase the diagnostic accuracy, assist in the decision-making of supplementary multiphasic enhanced medical imaging, as well as for developing novel prognostic biomarkers for liver cancer patients.     

Cost-Effectiveness of a Guided Peripherally Inserted Central Catheter Placement System: A Single-Center Cohort Study

PurposeTo assess the cost-effectiveness of peripherally inserted central catheter (PICC) placements using an ultrasound and electrocardiogram-guided system versus external measurements and confirmatory chest X-rays (CXRs).Materials and MethodsSixty-eight guided PICC placements were performed in 63 outpatients (mean age, 43 ± 13 years; 50% male) and compared to 68 propensity score-matched PICC placements (mean age, 44 ± 13 years; 56% male) performed using external measurements by the same operators. Post-placement CXRs were used to confirm final catheter tip positioning. Cohorts were compared in terms of repositioning rates, desired tip positioning rates (in the lower third of the superior vena cava or at the cavoatrial junction), and estimated cost per PICC positioned as desired using manufacturer quotes, Medicare reimbursement rates, and hourly wages for staff time. Agreement between tip positioning according to the guided system versus CXR was also assessed.ResultsGuided PICC placements required less repositioning (1.5% vs 10.3%, P = .03) and resulted in more catheters positioned as desired (86.8% vs 67.6%, P = .01) than the external measurement approach. The cost per PICC positioned as desired was lower for guided placements ($318.54 vs $381.44), suggesting that the guided system was cost-effective in this clinical setting. Guided system-CXR agreement for tip position was poor (κ=0.25, P = .002) due to tips being slightly farther from the cavoatrial junction on CXR than indicated by the guided system.ConclusionsThe guided PICC placement system was cost-effective in outpatients treated by a single division of interventional radiology at an academic institution.     

Potential of employing a quantum iterative reconstruction algorithm for ultra-high-resolution photon-counting detector CT of the hip

IntroductionThis study investigated the image quality of a new quantum iterative reconstruction algorithm (QIR) for high resolution photon-counting CT of the hip.MethodsUsing a first-generation photon-counting CT scanner, five cadaveric specimens were examined with ultra-high-resolution protocols matched for radiation dose. Images were post-processed with a sharp convolution kernel and five different strength levels of iterative reconstruction (QIR 0 – QIR 4). Subjective image quality was rated independently by three radiologists on a five-point scale. Intraclass correlation coefficients (ICC) were computed for assessing interrater agreement. Objective image quality was evaluated by means of contrast-to-noise-ratios (CNR) in bone and muscle tissue.ResultsFor osseous tissue, subjective image quality was rated best for QIR 2 reformatting (median 5 [interquartile range 5–5]). Contrarily, for soft tissue, QIR 4 received the highest ratings among compared strength levels (3 [3–4]). Both ICC_bone (0.805; 95% confidence interval 0.711–0.877; p < 0.001) and ICC_muscle (0.885; 0.824–0.929; p < 0.001) suggested good interrater agreement. CNR in bone and muscle tissue increased with ascending strength levels of iterative reconstruction with the highest results recorded for QIR 4 (CNR_bone 29.43 ± 2.61; CNR_muscle 8.09 ± 0.77) and lowest results without QIR (CNR_bone 3.90 ± 0.29; CNR_muscle 1.07 ± 0.07) (all p < 0.001).ConclusionReconstructing photon-counting CT data with an intermediate QIR strength level appears optimal for assessment of osseous tissue, whereas soft tissue analysis benefitted from applying the highest strength level available.Implications for practiceQuantum iterative reconstruction technique can enhance image quality by significantly reducing noise and improving CNR in ultra-high resolution CT imaging of the hip.     

Metal artefact reduction in low-dose computed tomography: Benefits of tin prefiltration versus postprocessing of dual-energy datasets over conventional CT imaging

IntroductionThe purpose of this study was to determine the potential for metal artefact reduction in low-dose multidetector CT as these pose a frequent challenge in clinical routine. Investigations focused on whether spectral shaping via tin prefiltration, virtual monoenergetic imaging or virtual blend imaging (VBI) offers superior image quality in comparison with conventional CT imaging.MethodsUsing a third-generation dual-source CT scanner, two cadaveric specimens with different metal implants (dental, cervical spine, hip, knee) were examined with acquisition protocols matched for radiation dose with regards to tube voltage and current. In order to allow for precise comparison, and due to the relatively short scan lengths, automatic tube current modulation was disabled. Specifically, the following scan protocals were examined: conventional CT protocols (100/120 kVp), tin prefiltration (Sn 100/Sn 150 kVp), VBI and virtual monoenergetic imaging (VME 100/120/150 keV). Mean attenuation and image noise were measured in hyperdense and hypodense artefacts, in artefact-impaired and artefact-free soft tissue. Subjective image quality was rated independently by three radiologists.ResultsObjectively, Sn 150 kVp allowed for the best reduction of hyperdense streak artefacts (p < 0.001), while VME 150 keV and Sn 150 kVp protocols facilitated equally good reduction of hypodense artefacts (p = 0.173). Artefact-impaired soft tissue attenuation was lowest in Sn 150 kVp protocols (p ≤ 0.011), whereas all VME showed significantly less image noise compared to conventional or tin-filtered protocols (p ≤ 0.001). Subjective assessment favoured Sn 150 kVp regarding hyperdense streak artefacts and delineation of cortical bone (p ≤ 0.005). The intraclass correlation coefficient was 0.776 (95% confidence interval: 0.712–0.831; p < 0.001) indicating good interrater reliability.ConclusionIn the presence of metal implants in our cadaveric study, tin prefiltration with 150 kVp offers superior artefact reduction for low-dose CT imaging of osseous tissue compared with virtual monoenergetic images of dual-energy datasets. The delineation of cortical boundaries seems to benefit particularly from spectral shaping.Implications for practiceLow-dose CT imaging of osseous tissue in combination with tin prefiltration allows for superior metal artefact reduction when compared to virtual monoenergetic images of dual-energy datasets. Employing this technique ought to be considered in daily routine when metal implants are present within the scan volume as findings suggest it allows for radiation dose reduction and facilitates diagnosis relevant to further treatment.     

The potential of iodinated contrast reduction in dual-energy CT thoracic angiography; an evaluation of image quality

IntroductionThe purpose of this study was to compare a dual energy CT (DECT) protocol with 50% reduction of iodinated contrast to a single energy CT (SECT) protocol using standard contrast dose in imaging of the thoracic aorta.MethodsDECT with a 50% reduction in iodinated contrast was compared with SECT. For DECT, monoenergetic images at 50, 55, 60, 65, 68, 70, and 74 keV were reconstructed with adaptive statistical iterative reconstruction (ASiR-V) of 50% and 80%. Objective image quality parameters included intravascular attenuation (HU), image noise (SD), contrast-to-noise ratio (CNR), and signal-to-noise ratio (SNR). Two independent radiologists subjectively assessed the image quality for the 55 and 68 keV DECT reconstructions and SECT on a five-point Likert scale.ResultsAcross 14 patients, the intravascular attenuation at 50–55 keV was comparable to SECT (p > 0.05). The CNRs were significantly lower for DECT with ASIR-V 50% compared to SECT for all keV-values (p < 0.05 for all). For ASIR-V 80%, CNR was comparable to SECT at energies below 60 keV (p > 0.05). The subjective image quality was comparable between DECT and SECT independent of keV level.ConclusionThis study indicates that a 50% reduction in iodinated contrast may result in adequate image quality using DECT with monoenergetic reconstructions at lower energy levels for the imaging of the thoracic aorta. The best image quality was obtained for ASiR-V 80% image reconstructions at 55 keV.Implications of practiceDual energy CT with a reduction in iodinated contrast may result in adequate image quality in imaging of the thoracic aorta. However, increased radiation dose may limit the use to patients in which a reduction in fluid and iodinated contrast volume may outweigh this risk.     

Pregnancy Rate and Outcomes following Uterine Artery Embolization for Uterine Arteriovenous Malformations: A Systematic Review and Meta-Analysis

PurposeTo systematically review published studies on the pregnancy rate and outcomes after uterine artery embolization (UAE) for uterine arteriovenous malformations (UAVMs).Materials and MethodsInternational medical databases were searched for all English-language studies published between 2000 and 2022 on patients with UAVMs who had undergone embolization and had a subsequent pregnancy. Data on the pregnancy rate, pregnancy complications, and physiologic status of newborns were extracted from the articles. Ten case series were included in the meta-analysis, and 18 case reports on pregnancy following UAE were reviewed.ResultsIn the case series, 44 pregnancies were reported in 189 patients. The pooled estimate of pregnancy rate was 23.3% (95% confidence interval [CI], 17.3%–29.3%). The pregnancy rate was higher in studies of women with a mean age of ≤30 years (50.6% vs 22.2%; P < .05). The pooled estimate of live birth rate was 88.6% (95% CI, 78.6%–98.7%).ConclusionsAll published series report preservation of fertility and successful pregnancies after embolization of UAVMs. The live birth rate in these series does not differ substantially from that of the general population.     

Adrenal Vein Sampling Results and Surgical Outcomes in Patients with a Normal Plasma Aldosterone Concentration

PurposeTo determine the utility of adrenal vein sampling (AVS) and outcomes after adrenalectomy in patients with normal plasma aldosterone concentration (PAC) and elevated aldosterone-to-renin ratio (ARR).Materials and MethodsThe study sample included 106 patients with ARR greater than 20 and PAC between 5 and 15 ng/dL (normal PAC group) who underwent AVS from 2005 to 2021. These patients were compared with a cohort of 106 patients with ARR >20 and PAC >15 ng/dL (high PAC group) who underwent AVS during the same period. Data regarding baseline clinical characteristics, lateralization indices from AVS, and outcomes after adrenalectomy were analyzed.ResultsAVS was technically successful in 210 patients (210/212, 99%). A smaller proportion of patients in the normal PAC group showed a lateralization index of >4 compared with those in the high PAC group (44% vs 64%, P <.01). A similar proportion of patients in the normal PAC group experienced improved or cured hypertension after adrenalectomy compared with that in the high PAC group (94% vs 88%, P =.31). Hypokalemia was cured in all patients in the normal PAC group after adrenalectomy compared with 98% of patients in the high PAC group (100% vs 98%, P = 1).ConclusionsAlthough lateralization is less frequent for patients with normal PAC, patients who do lateralize show similar blood pressure response and correction of hypokalemia after adrenalectomy, regardless of initial plasma aldosterone levels. Therefore, patients with PAC <15 ng/dL should still be considered for AVS provided the ARR is elevated.     

Endobronchial Forceps Retrieval of Embedded Inferior Vena Cava Filters: Retrieval of 535 Filters at a Single Center

PurposeTo report results of 16 years of using the endobronchial forceps technique to remove embedded inferior vena cava (IVC) filters.Materials and MethodsBetween January 2005 and June 2021, 534 patients (310 women and 224 men) with a mean age of 52 years (standard deviation [SD] ± 16 years) presented for complex filter retrieval of 535 tip- or strut-embedded IVC filters. Tip-embedded filters were diagnosed on rotational venography. Filters were considered strut-embedded if they were closed-cell filters with dwell times of >6 months. The filter was dissected from the IVC using rigid bronchoscopy forceps and removed through a vascular sheath.ResultsThe endobronchial forceps technique was successful in 530 of 537 retrieval attempts on an intention-to-treat basis (98.7%); a total of 530 filters were retrieved. There were 7 failures: (a) 5 failed retrieval attempts (2 that were retrieved successfully in subsequent procedures) and (b) 2 for which retrieval was not attempted. The mean filter dwell time was 1,459 days (SD ± 1,617 days). Laser sheaths were not used for any removal. Filters included herein were 137 Celect (94 Celect and 43 Celect Platinum), 99 Günther Tulip, 72 Option (48 Option and 24 Option Elite), 68 G2, 45 G2X/Eclipse, 42 Denali, 30 OptEase, 29 Recovery, 7 Meridian, and 6 ALN with Hook filters. Thirty-four minor (6.3%) and 11 major (2%) adverse events (AEs) occurred, which did not result in permanent sequelae.ConclusionsUse of endobronchial forceps for removal of tip- and strut-embedded retrievable IVC filters is effective and has low AE rates.     

The Effect of Smoking Cessation on the Technical Success of Endovascular Treatment for Thromboangiitis Obliterans

PurposeTo show that smoking cessation improves the technical success of lower extremity endovascular treatment in patients with thromboangiitis obliterans (TAO), or Buerger disease.Materials and MethodsOne hundred two patients with TAO who underwent endovascular treatment for chronic limb-threatening ischemia or severe claudication of lower extremities in a tertiary hospital between 2015 and 2022 were included in this retrospective study. Data on serum cotinine levels were available for the last 45 patients, and 38 patients constituted the study population. Per the institution’s protocol, patients were instructed to quit smoking 15 days before the intervention. However, cotinine levels showed that some of the patients continued smoking. Technical and recanalization successes were assessed as the primary end points. The secondary end point was the improvement in Rutherford scores at the 1-month follow-up. The McNemar test was used to compare the proportion of recanalized arteries after the intervention.ResultsThirty-seven men and 1 woman (mean age, 42.9 years ± 10.1) were evaluated. The overall technical success rate was 86.8% in the study group. The technical success rate was significantly higher in the nonsmoker group (n = 24 [96%]) than in the smoker group (n = 8 [61.5%]; P = .006). One-month clinical data were available for 100% of the patients. The Rutherford category of the nonsmoker group was significantly lower at the 1-month follow-up. In addition, the Wilcoxon signed-rank test revealed lower Rutherford scores after the intervention in the nonsmoker group. The adverse event rate was 8%. One (2.7%) patient in the smoker group underwent a minor amputation.ConclusionsCessation of smoking before endovascular therapy improved technical success and recanalization rates in patients with TAO.     

Machine learning to identify lymph node metastasis from thyroid cancer in patients undergoing contrast-enhanced CT studies

IntroductionWe compared the diagnostic performance of morphological methods such as the major axis, the minor axis, the volume and sphericity and of machine learning with texture analysis in the identification of lymph node metastasis in patients with thyroid cancer who had undergone contrast-enhanced CT studies.MethodsWe sampled 772 lymph nodes with histology defined tissue types (84 metastatic and 688 benign lymph nodes) that were visualised on CT images of 117 patients. A support vector machine (SVM), free programming software (Python), and the scikit-learn machine learning library were used to discriminate metastatic-from benign lymph nodes. We assessed 96 texture and 4 morphological features (major axis, minor axis, volume, sphericity) that were reported useful for the differentiation between metastatic and benign lymph nodes on CT images. The area under the curve (AUC) obtained by receiver operating characteristic analysis of univariate logistic regression and SVM classifiers were calculated for the training and testing datasets.ResultsThe AUC for all classifiers in training and testing datasets was 0.96 and 0.86, at the SVM for machine learning. When we applied conventional methods to the training and testing datasets, the AUCs were 0.63 and 0.48 for the major axis, 0.70 and 0.44 for the minor axis, 0.66 and 0.43 for the volume, and 0.69 and 0.54 for sphericity, respectively. The SVM using texture features yielded significantly higher AUCs than univariate logistic regression models using morphological features (p = 0.001).ConclusionFor the identification of metastatic lymph nodes from thyroid cancer on contrast-enhanced CT images, machine learning combined with texture analysis was superior to conventional diagnostic methods with the morphological parameters.Implications for practiceOur findings suggest that in patients with thyroid cancer and suspected lymph node metastasis who undergo contrast-enhanced CT studies, machine learning using texture analysis is high diagnostic value for the identification of metastatic lymph nodes.     

Diagnostic Performance of CT-Guided Bone Biopsies in Patients with Suspected Osteomyelitis of the Appendicular and Axial Skeleton with a Focus on Clinical and Technical Factors Associated with Positive Microbiology Culture Results

PurposeTo assess diagnostic performance of CT-guided percutaneous needle bone biopsy (CTNBB) in patients with suspected osteomyelitis and analyze whether certain clinical or technical factors were associated with positive microbiology results.Materials and MethodsAll CTNBBs performed in a single center for suspected osteomyelitis of the appendicular and axial skeleton during 2003–2018 were retrospectively reviewed. Specific inclusion criteria were clinical and radiologic suspicion of osteomyelitis. Standard of reference was defined using outcome of surgical histopathology and microbiology culture and clinical and imaging follow-up. Technical and clinical data (needle size, comorbidities, clinical factors, laboratory values, blood cultures) were collected. Logistic regression was performed to assess associations between technical and clinical data and microbiology biopsy outcome.ResultsA total of 142 CTNBBs were included (46.5% female patients; age ± SD 46.10 y ± 22.8), 72 (50.7%) from the appendicular skeleton and 70 (49.3%) from the axial skeleton. CTNBB showed a sensitivity of 42.5% (95% confidence interval [CI], 32.0%–53.6%) in isolating the causative pathogen. A higher rate of positive microbiology results was found in patients with intravenous drug use (odds ratio [OR] = 5.15; 95% CI, 1.2–21.0; P = .022) and elevated white blood cell count ≥ 10 × 10⁹/L (OR = 3.9; 95% CI, 1.62–9.53; P = .002). Fever (≥ 38°C) was another clinical factor associated with positive microbiology results (OR = 3.6; 95% CI, 1.3–9.6; P = .011).ConclusionsCTNBB had a low sensitivity of 42.5% for isolating the causative pathogen. Rate of positive microbiology samples was significantly higher in patients with IV drug use, elevated white blood cell count, and fever.     

Effectiveness of Thermal Ablation for Aldosterone-Producing Adrenal Adenoma: A Systematic Review and Meta-Analysis of Clinical and Biochemical Parameters

PurposeTo assess the effectiveness of thermal ablation for aldosterone-producing adrenal adenoma.Materials and MethodsA systematic search of the PubMed and CINAHL databases was performed to identify studies of thermal ablation for adrenal adenomas. Random effects meta-analysis models were used to compare pre- and post-treatment values of the following outcomes: systolic blood pressure (SBP), diastolic blood pressure (DBP), use of antihypertensive medications, and biochemical parameters (plasma aldosterone levels, aldosterone-to-renin ratio, and potassium levels). The rate of hypertension (HTN) resolution and improvement were also evaluated.ResultsA total of 89 patients from 7 studies were included in the analysis. The mean postablation follow-up duration was 45.8 months. Pooled data analysis revealed a statistically significant decrease in SBP (−29.06 mm Hg; 95% confidence interval [CI], −33.93 to −24.19), DBP (−16.03 mm Hg; 95% CI, −18.33 to −13.73), and the number of antihypertensive medications used (−1.43; 95% CI, −1.97 to −0.89) after ablation. Biochemical parameters had returned to normal ranges after ablation in all studies. The cumulative rate of resolution or improvement in HTN status was 75.3%. On metaregression analysis, there was no statistically significant association between postablation blood pressure changes or serum aldosterone levels and study follow-up duration.ConclusionsThermal ablation for aldosterone-producing adrenal adenoma can be effective in controlling blood pressure, reducing the need for antihypertensive medications, and normalizing hormone secretion. Further higher-quality evidence is needed to confirm these results.     

The Value of Preprocedural MR Imaging in Genicular Artery Embolization for Patients with Osteoarthritic Knee Pain

PurposeTo determine the value of preprocedural MR imaging in genicular artery embolization (GAE) for patients with osteoarthritic knee pain.Materials and MethodsThis single-center study retrospectively analyzed 28 knees in 18 patients who underwent GAE for intractable knee pain < 1 month after MR imaging. The pain experienced in each knee was evaluated on a 100-mm visual analog scale (VAS) at baseline and 1- and 3-month after GAE. “GAE responders” were defined as knees that exhibited greater than 30% reduction of VAS pain scores from baseline at both follow-up visits. Musculoskeletal radiologists evaluated MR images of the affected knee compartment regarding cartilage defects, osteophytes, subchondral cysts, bone marrow lesions (BMLs), meniscal injury, and joint effusion. The performances of Kellgren–Lawrence (KL) grading and MR findings in predicting GAE responders was estimated based on receiver operating characteristic curves.ResultsThe mean VAS pain score was 84.3 mm. BML (area under the curve [AUC], 0.860; P < .001), meniscal injury (AUC, 0.811; P = .003), and KL grading (AUC, 0.898; P < .001) were significantly associated with GAE outcome. To predict GAE responders, KL grade ≤ 2 yielded a sensitivity of 87.5% and a specificity of 60.9%, BML grade ≤ 1 yielded a sensitivity of 75.0% and a specificity of 69.6%, and meniscal injury grade ≤ 2 yielded a sensitivity of 83.3% and a specificity of 72.7%.ConclusionsLarge BMLs and severe meniscal injuries on MR imaging, as well as high KL grades, indicated poor responses to GAE.