Using formalin embalmed cadavers to teach fracture identification with ultrasound.

BACKGROUND: Ultrasound is being utilized more frequently to diagnose fractures in bone and track fracture reduction quickly, and without radiation exposure in the ED. Realistic and practical methods of teaching sonographic fracture identification to medical trainees are needed. The objective of this study is to determine the feasibility of using formalin-embalmed human cadavers in teaching medical trainees to use ultrasound to identify synthetic fractures in tibia, radius, and metacarpal bones. METHODS: First-year medical students attended an orientation presentation and a 15-min scanning workshop, to evaluate fractures in cadaver bones with an instructor. Next participants independently scanned bones to determine if a fracture was present. Questionnaires were given that assessed participant self-confidence and ability to evaluate still ultrasound images for fracture and differentiate between tissue layers before, after, and 5 months following training. RESULTS: Participants were collectively able to scan and differentiate between fractured and unfractured bone in 75% of 186 total bone scanning attempts (tibia: 81% correct, metacarpal: 68% correct, radius: 76% correct). When evaluating still ultrasound images for fracture, participants' scores rose significantly following training from an average score of 77.4 to 91.1% (p = 0.001). Five months post-training, scores fell slightly, to an average of 89.8% (p = 0.325). CONCLUSIONS: Ultrasound images of formalin-embalmed cadaveric fractures are of sufficient quality to use in teaching fracture identification to medical trainees. With only 15 minutes of scanning experience, medical trainees can learn to independently scan and significantly increase their ability to identify fractures in still ultrasound images.

Identification of Patients with Osteoporotic Vertebral Fractures via Simple Text Search of Routine Radiology Reports.

Secondary fracture prevention programs mostly identify patients with symptomatic non-vertebral fractures, whereas asymptomatic vertebral fractures are usually missed. We here describe the development and validation of a simple method to systematically identify patients with radiographic vertebral fractures using simple text-based searching of free-text radiology reports. The study consisted of two phases. In the development phase (DP), twelve search terms were used to identify vertebral fractures in all X-ray and CT reports issued over a period of 6 months. Positive reports were manually reviewed to confirm whether or not a vertebral fracture had in fact been reported. The three search terms most effective in detecting vertebral fractures during the DP were then applied during the implementation phase (IP) over several weeks to test their ability to identify patients with vertebral fractures. The search terms 'Loss of Height' (LoH), 'Compression Fracture' (CoF) and 'Crush Fracture' (CrF) identified the highest number of imaging reports with a confirmed vertebral fracture. These three search terms identified 581 of 689 (84%) of all true vertebral fractures with a positive predictive value of 76%. Using these three terms in the IP, 126 reports were identified of which 100 (79%) had a vertebral fracture confirmed on manual review. Amongst a sample of 587 reports in week 1 of the IP, 7 (1.2%) were false negatives. Many patients with vertebral fractures can be identified via a simple text-based search of electronic radiology reports. This method may be utilised by secondary fracture prevention programs to narrow the 'care gap' in osteoporosis management.

Natural language processing of radiology reports for identification of skeletal site-specific fractures.

BACKGROUND: Osteoporosis has become an important public health issue. Most of the population, particularly elderly people, are at some degree of risk of osteoporosis-related fractures. Accurate identification and surveillance of patient populations with fractures has a significant impact on reduction of cost of care by preventing future fractures and its corresponding complications. METHODS: In this study, we developed a rule-based natural language processing (NLP) algorithm for identification of twenty skeletal site-specific fractures from radiology reports. The rule-based NLP algorithm was based on regular expressions developed using MedTagger, an NLP tool of the Apache Unstructured Information Management Architecture (UIMA) pipeline to facilitate information extraction from clinical narratives. Radiology notes were retrieved from the Mayo Clinic electronic health records data warehouse. We developed rules for identifying each fracture type according to physicians' knowledge and experience, and refined these rules via verification with physicians. This study was approved by the institutional review board (IRB) for human subject research. RESULTS: We validated the NLP algorithm using the radiology reports of a community-based cohort at Mayo Clinic with the gold standard constructed by medical experts. The micro-averaged results of sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and F1-score of the proposed NLP algorithm are 0.930, 1.0, 1.0, 0.941, 0.961, respectively. The F1-score is 1.0 for 8 fractures, and above 0.9 for a total of 17 out of 20 fractures (85%). CONCLUSIONS: The results verified the effectiveness of the proposed rule-based NLP algorithm in automatic identification of osteoporosis-related skeletal site-specific fractures from radiology reports. The NLP algorithm could be utilized to accurately identify the patients with fractures and those who are also at high risk of future fractures due to osteoporosis. Appropriate care interventions to those patients, not only the most at-risk patients but also those with emerging risk, would significantly reduce future fractures.

3D-printed model is a useful addition in orthopedic resident education for the understanding of tibial plateau fractures.

This study aimed to explore the role of the three-dimension (3D) printed models in orthopedic resident training of tibial plateau fractures. A total of 41 residents from our institution were divided into two groups. The intervention group, consisting of 20 residents, had access to 3D-printed models illustrating thirteen tibial plateau fractures. In contrast, the control group, comprising 21 residents, received digital images of thirteen identical tibial plateau fractures. Evaluation of learning outcomes included the accurate identification of tibial plateau fracture patterns, deduction of traumatic mechanisms, preoperative plan, assessment time, and subjective questionnaire responses. The participants with 3D printed models scored significantly higher in both the Schatzker classification and Luo three-column classification compared to those without 3D printed models. Residents in the intervention group performed better in accuracy in deducing traumatic mechanisms compared to the control group. In addition, the sum score of preoperative plan in the intervention group was significantly higher than that in the control group. Specifically, participants with 3D printed models scored higher in surgical approach choice and implants placement than these in the control group. Residents exposed to 3D printed models also spent less time to complete the assessment than those with access only to digital imaging. Subjective assessments indicated that 3D-printed models boosted confidence in fracture identification, improved preoperative plan for fracture management and enhanced the understanding in injury mechanism of tibial plateau fractures. Furthermore, residents agreed that the use of 3D-printed models heightened their interest in learning tibial plateau fractures. Therefore, the addition of 3D printed models significantly contributed to a comprehensive understanding of tibial plateau fractures, the improvement in fracture identification, inferring injury mechanisms and preoperative plan.

Convolutional Neural Network to Classify Infrared Thermal Images of Fractured Wrists in Pediatrics.

Convolutional neural network (CNN) models were devised and evaluated to classify infrared thermal (IRT) images of pediatric wrist fractures. The images were recorded from 19 participants with a wrist fracture and 21 without a fracture (sprain). The injury diagnosis was by X-ray radiography. For each participant, 299 IRT images of their wrists were recorded. These generated 11,960 images (40 participants × 299 images). For each image, the wrist region of interest (ROI) was selected and fast Fourier transformed (FFT) to obtain a magnitude frequency spectrum. The spectrum was resized to 100 × 100 pixels from its center as this region represented the main frequency components. Image augmentations of rotation, translation and shearing were applied to the 11,960 magnitude frequency spectra to assist with the CNN generalization during training. The CNN had 34 layers associated with convolution, batch normalization, rectified linear unit, maximum pooling and SoftMax and classification. The ratio of images for the training and test was 70:30, respectively. The effects of augmentation and dropout on CNN performance were explored. Wrist fracture identification sensitivity and accuracy of 88% and 76%, respectively, were achieved. The CNN model was able to identify wrist fractures; however, a larger sample size would improve accuracy.

Inducing a Closed Ankle Fracture in a Cadaveric Model to Create Interdisciplinary Teaching Models for Preclinical Medical Students.

This paper describes a cost-effective fracture simulation that aids in preclinical learning of fracture identification. This project establishes feasibility of inducing closed fractures in a donated cadaver. The study team made an identified Weber C ankle fracture in this donor, maintaining the soft tissue envelope surrounding the fracture, using minimal materials in a stepwise process that will be reproducible across educational programs regardless of access to highly specialized equipment. The resulting fracture can be used to help students identify fractures during their preclinical education and has demonstrated educational potential for future use and adaptation.

Natural language processing of radiology reports for the identification of patients with fracture.

Text-search software can be used to identify people at risk of re-fracture. The software studied identified a threefold higher number of people with fractures compared with conventional case finding. Automated software could assist fracture liaison services to identify more people at risk than traditional case finding. PURPOSE: Fracture liaison services address the post-fracture treatment gap in osteoporosis (OP). Natural language processing (NLP) is able to identify previously unrecognized patients by screening large volumes of radiology reports. The aim of this study was to compare an NLP software tool, XRAIT (X-Ray Artificial Intelligence Tool), with a traditional fracture liaison service at its development site (Prince of Wales Hospital [POWH], Sydney) and externally validate it in an adjudicated cohort from the Dubbo Osteoporosis Epidemiology Study (DOES). METHODS: XRAIT searches radiology reports for fracture-related terms. At the development site (POWH), XRAIT and a blinded fracture liaison clinician (FLC) reviewed 5,089 reports and 224 presentations, respectively, of people 50 years or over during a simultaneous 3-month period. In the external cohort of DOES, XRAIT was used without modification to analyse digitally readable radiology reports (n = 327) to calculate its sensitivity and specificity. RESULTS: XRAIT flagged 433 fractures after searching 5,089 reports (421 true fractures, positive predictive value of 97%). It identified more than a threefold higher number of fractures (421 fractures/339 individuals) compared with manual case finding (98 individuals). Unadjusted for the local reporting style in an external cohort (DOES), XRAIT had a sensitivity of 70% and specificity of 92%. CONCLUSION: XRAIT identifies significantly more clinically significant fractures than manual case finding. High specificity in an untrained cohort suggests that it could be used at other sites. Automated methods of fracture identification may assist fracture liaison services so that limited resources can be spent on treatment rather than case finding.

The Accuracy of Prevalent Vertebral Fracture Detection in Children Using Targeted Case-Finding Approaches.

Due to concerns about cumulative radiation exposure in the pediatric population, it is not standard practice to perform spine radiographs in most conditions that predispose to vertebral fracture (VF). In this study we examined the accuracy of two clinical predictors, back pain and lumbar spine bone mineral density (LS BMD), to derive four case-finding paradigms for detection of prevalent VF (PVF). Subjects were 400 children at risk for PVF (leukemia 186, rheumatic disorders 135, nephrotic syndrome 79). Back pain was assessed by patient report, LS BMD was measured by dual-energy X-ray absorptiometry, and PVF were quantified on spine radiographs using the modified Genant semiquantitative method. Forty-four patients (11.0%) had PVF. Logistic regression analysis between LS BMD and PVF produced an odds ratio (OR) of 1.9 (95% confidence interval [CI], 1.5 to 2.5) per reduction in Z-score unit, an area under the receiver operating characteristic curve of 0.70 (95% CI, 0.60 to 0.79), and an optimal BMD Z-score cutoff of -1.6. Case identification using either low BMD alone (Z-score < -1.6) or back pain alone gave similar results for sensitivity (55%, 52%, respectively), specificity (78%, 81%, respectively), positive predictive value (PPV; 24%, 25%, respectively), and negative predictive value (NPV; 93%, 93%, respectively). The paradigm using low BMD plus back pain produced lower sensitivity (32%), higher specificity (96%), higher PPV (47%), and similar NPV (92%). The approach using low BMD or back pain had the highest sensitivity (75%), lowest specificity (64%), lowest PPV (20%), and highest NPV (95%). All paradigms had increased sensitivities for higher fracture grades. Our results show that BMD and back pain history can be used to identify children with the highest risk of PVF so that radiography can be used judiciously. The specific paradigm to be applied will depend on the expected PVF rate and the clinical approach to the use of radiography. © 2019 American Society for Bone and Mineral Research.

Adding Lateral Spine Imaging for Vertebral Fractures to Densitometric Screening: Improving Ascertainment of Patients at High Risk of Incident Osteoporotic Fractures.

The current diagnosis of osteoporosis is limited to a T-score ≤-2.5. However, asymptomatic vertebral fractures (VF) are known to predict a high risk of subsequent fractures and pharmaceutical intervention is known to reduce future fracture risk in these individuals. In a prospective, population-based cohort of ambulant older women, we sought to evaluate the role of VF detection by screening densitometric lateral spine imaging (LSI) for VF at time of bone density testing to the effect on the magnitude of fracture risk. A total of 1084 women (mean age 75 years ± SD 3 years) had baseline LSI that identified 100 (9%) women with VFs and 89 (8%) with femoral neck (FN) T-score osteoporosis ≤-2.5. Follow-up identified incident clinical spine fracture in 73 (7%), 305 (28%) with any fracture-related hospitalization, and 121 (11%) with a hip fracture-related hospitalization. Compared with those without baseline VF, in those with baseline VF, relative risk (RR) for incident clinical spine, hip, and any fracture were 3.46 (95% confidence interval [CI] 2.14-5.60, p < 0.001); 1.72 (95% CI 1.09-2.71, p = 0.02), and 1.4 (95% CI 1.07-1.84, p = 0.02), respectively. In 675 (62%) of women with femoral neck osteopenia (T-score <-1 to >-2.5), 61 (9%) also had a VF. Compared with those without baseline VF, RR for any incident fragility fractures and fractures at spine and hip in those with baseline VF were 1.6 (95% CI 1.2-2.1, p < 0.01), 3.9 (95% CI 2.2-6.9, p < 0.01), and 1.6 (95% CI 0.9-2.8, p = 0.10), respectively. On basis of the prognosis, older women with LSI VF with osteopenia should be diagnosed with osteoporosis and should be considered for pharmaceutical intervention. © 2018 American Society for Bone and Mineral Research.