Artificial Intelligence for Detecting Acute Fractures in Patients Admitted to an Emergency Department: Real-Life Performance of Three Commercial Algorithms.

RATIONALE AND OBJECTIVES: Interpreting radiographs in emergency settings is stressful and a burden for radiologists. The main objective was to assess the performance of three commercially available artificial intelligence (AI) algorithms for detecting acute peripheral fractures on radiographs in daily emergency practice. MATERIALS AND METHODS: Radiographs were collected from consecutive patients admitted for skeletal trauma at our emergency department over a period of 2 months. Three AI algorithms-SmartUrgence, Rayvolve, and BoneView-were used to analyze 13 body regions. Four musculoskeletal radiologists determined the ground truth from radiographs. The diagnostic performance of the three AI algorithms was calculated at the level of the radiography set. Accuracies, sensitivities, and specificities for each algorithm and two-by-two comparisons between algorithms were obtained. Analyses were performed for the whole population and for subgroups of interest (sex, age, body region). RESULTS: A total of 1210 patients were included (mean age 41.3 ± 18.5 years; 742 [61.3%] men), corresponding to 1500 radiography sets. The fracture prevalence among the radiography sets was 23.7% (356/1500). Accuracy was 90.1%, 71.0%, and 88.8% for SmartUrgence, Rayvolve, and BoneView, respectively; sensitivity 90.2%, 92.6%, and 91.3%, with specificity 92.5%, 70.4%, and 90.5%. Accuracy and specificity were significantly higher for SmartUrgence and BoneView than Rayvolve for the whole population (P < .0001) and for subgroups. The three algorithms did not differ in sensitivity (P = .27). For SmartUrgence, subgroups did not significantly differ in accuracy, specificity, or sensitivity. For Rayvolve, accuracy and specificity were significantly higher with age 27-36 than ≥53 years (P = .0029 and P = .0019). Specificity was higher for the subgroup knee than foot (P = .0149). For BoneView, accuracy was significantly higher for the subgroups knee than foot (P = .0006) and knee than wrist/hand (P = .0228). Specificity was significantly higher for the subgroups knee than foot (P = .0003) and ankle than foot (P = .0195). CONCLUSION: The performance of AI detection of acute peripheral fractures in daily radiological practice in an emergency department was good to high and was related to the AI algorithm, patient age, and body region examined.

Application of artificial intelligence to imaging interpretations in the musculoskeletal area: Where are we? Where are we going?

The interest of researchers, clinicians and radiologists, in artificial intelligence (AI) continues to grow. Deep learning is a subset of machine learning, in which the computer algorithm itself can determine the optimal imaging features to answer a clinical question. Convolutional neural networks are the most common architecture for performing deep learning on medical images. The various musculoskeletal applications of deep learning are the detection of abnormalities on X-rays or cross-sectional images (CT, MRI), for example the detection of fractures, meniscal tears, anterior cruciate ligament tears, degenerative lesions of the spine, bone metastases, classification of e.g., dural sac stenosis, degeneration of intervertebral discs, assessment of skeletal age, and segmentation, for example of cartilage. Software developments are already impacting the daily practice of orthopedic imaging by automatically detecting fractures on radiographs. Improving image acquisition protocols, improving the quality of low-dose CT images, reducing acquisition times in MRI, or improving MR image resolution is possible through deep learning. Deep learning offers an automated way to offload time-consuming manual processes and improve practitioner performance. This article reviews the current state of AI in musculoskeletal imaging.

Artificial Intelligence for Classification of Soft-Tissue Masses at US.

PURPOSE: To train convolutional neural network (CNN) models to classify benign and malignant soft-tissue masses at US and to differentiate three commonly observed benign masses. MATERIALS AND METHODS: In this retrospective study, US images obtained between May 2010 and June 2019 from 419 patients (mean age, 52 years ± 18 [standard deviation]; 250 women) with histologic diagnosis confirmed at biopsy or surgical excision (n = 227) or masses that demonstrated imaging characteristics of lipoma, benign peripheral nerve sheath tumor, and vascular malformation (n = 192) were included. Images in patients with a histologic diagnosis (n = 227) were used to train and evaluate a CNN model to distinguish malignant and benign lesions. Twenty percent of cases were withheld as a test dataset, and the remaining cases were used to train the model with a 75%-25% training-validation split and fourfold cross-validation. Performance of the model was compared with retrospective interpretation of the same dataset by two experienced musculoskeletal radiologists, blinded to clinical history. A second group of US images from 275 of the 419 patients containing the three common benign masses was used to train and evaluate a separate model to differentiate between the masses. The models were trained on the Keras machine learning platform (version 2.3.1), with a modified pretrained VGG16 network. Performance metrics of the model and of the radiologists were compared by using the McNemar test, and 95% CIs for performance metrics were estimated by using the Clopper-Pearson method (accuracy, recall, specificity, and precision) and the DeLong method (area under the receiver operating characteristic curve). RESULTS: The model trained to classify malignant and benign masses demonstrated an accuracy of 79% (95% CI: 68, 88) on the test data, with an area under the receiver operating characteristic curve of 0.91 (95% CI: 0.84, 0.98), matching the performance of two expert readers. Performance of the model distinguishing three benign masses was lower, with an accuracy of 71% (95% CI: 61, 80) on the test data. CONCLUSION: The trained CNN was capable of differentiating between benign and malignant soft-tissue masses depicted on US images, with performance matching that of two experienced musculoskeletal radiologists.© RSNA, 2020.

Does Magnetic Resonance Imaging After Diagnostic Ultrasound for Soft Tissue Masses Change Clinical Management?

OBJECTIVES: To evaluate whether a follow-up magnetic resonance imaging (MRI) scan performed after initial ultrasound (US) to evaluate soft tissue mass (STM) lesions of the musculoskeletal system provides additional radiologic diagnostic information and alters clinical management. METHODS: A retrospective chart review was performed of patients undergoing initial US evaluations of STMs of the axial or appendicular skeleton between November 2012 and March 2019. Patients who underwent US examinations followed by MRI for the evaluation of STM lesions were identified. For inclusion, the subsequent pathologic correlation was required from either a surgical or image-guided biopsy. Imaging studies with pathologic correlations were then reviewed by 3 musculoskeletal radiologists, who were blinded to the pathologic diagnoses. The diagnostic utility of MRI was then assessed on the basis of a 5-point grading scale, and inter-reader agreements were determined by the Fleiss κ statistic. RESULTS: Ninety-two patients underwent MRI after US for STM evaluations. Final pathologic results were available in 42 cases. Samples were obtained by surgical excision or open biopsy (n = 34) or US-guided core biopsy (n = 8). The most common pathologic diagnoses were nerve sheath tumors (n = 9), lipomas (n = 5), and leiomyomas (n = 5). Imaging review showed that the subsequent MRI did not change the working diagnosis in 73% of cases, and the subsequent MRI was not considered to narrow the differential diagnosis in 68% of cases. There was slight inter-reader agreement for the diagnostic utility of MRI among individual cases (κ = 0.10) between the 3 readers. CONCLUSIONS: The recommendation of MRI to further evaluate STM lesions seen with US frequently fails to change the working diagnosis or provide significant diagnostic utility.

CT-guided radiofrequency ablation for osteoid osteomas: a systematic review.

OBJECTIVES: CT-guided radiofrequency ablation (CT-RFA) is considered to be the gold standard for treatment of osteoid osteoma (OO) yet treatment failures (TFs) continue to be reported. This systematic review was conducted to evaluate factors associated with TF, such as ablation time, lesion location, and patient age as well as evaluating how TF has trended over time. METHODS: Original studies reporting on patients undergoing CT-RFA of OO published between 2002 and 2019 were identified. TF was defined as patients with (1) recurrent or persistent pain +/- (2) imaging evidence of persistent OO. TFs were subdivided into those occurring after the index procedure (primary TF) or those occurring after repeat RFA (secondary TF). Subgroup analysis was performed for TF based on the study date (2002-2010 or 2010-2019), time duration of ablation at 90 °C (6 min or > 6 min), patient age, and tumor location (spinal vs. appendicular). RESULTS: Sixty-nine studies were included for a total of 3023 patients. The global primary TF rate was 8.3% whereas the secondary TF rate was 3.1%. The TF rate reported in studies published after 2011(7%) was about half that during the earlier time period 2002-2010 (14%). There was no statistical difference in TF corrected for age, OO location, or duration of ablation (respectively p = 0.39, 0.13, and 0.23). The global complication rate was 3%, the most frequent being skin burns (n = 24; 0.7%). CONCLUSIONS: A decrease in TF observed between 2011-2019 compared to 2002-2010 may reflect improvements in operator technique or advancements in equipment. Duration of ablation, patient age, or location of OO failed to significantly correlate with TF. KEY POINTS: • CT-guided radiofrequency ablation of osteoid osteomas is a safe technique with a low rate of treatment failure (8.3% failure rate after the primary radiofrequency reducing to 3.1% following a secondary treatment). • The treatment failure rate has decreased over time, possibly due to an improved understanding of the disease process, better technique, and advances in equipment. • Duration of ablation, patient age, or lesion location did not significantly correlate with treatment failure.

Hepatic involvement in HELLP syndrome: an update with emphasis on imaging features.

HELLP syndrome, which consists of hemolysis, elevated liver enzymes, and low platelet count is an unusual complication of pregnancy that is observed in only 10% to 15% of women with preeclampsia. Hepatic involvement in HELLP syndrome may present with various imaging features depending on the specific condition that includes nonspecific abnormalities such as perihepatic free fluid, hepatic steatosis, liver enlargement, and periportal halo that may precede more severe conditions such as hepatic hematoma and hepatic rupture with hemoperitoneum. Maternal clinical symptoms may be nonspecific and easily mistaken for a variety of other conditions that should be recognized. Because hepatic hematoma occurring in association with preeclampsia and HELLP syndrome is a potentially life-threatening complication, prompt depiction is critical and may help reduce morbidity and mortality. This review provides an update on demographics, risk factors, pathophysiology, and clinical features of hepatic complications due to HELLP syndrome along with a special emphasis on the imaging features of these uncommon conditions.