Commonly missed subtle skeletal injuries in children: a pictorial review

Diagnostic performance of an AI algorithm for the detection of appendicular bone fractures in pediatric patients

PURPOSE

To evaluate the diagnostic performance of an Artificial Intelligence (AI) algorithm, previously trained using both adult and pediatric patients, for the detection of acute appendicular fractures in the pediatric population on conventional X-ray radiography (CXR).

MATERIALS AND METHODS

In this retrospective study, anonymized extremities CXRs of pediatric patients (age <17 years), with or without fractures, were included. Six hundred CXRs (maintaining the positive-for-fracture and negative-for-fracture balance) were included, grouping them per body part (shoulder/clavicle, elbow/upper arm, hand/wrist, leg/knee, foot/ankle). Follow-up CXRs and/or second-level imaging were considered as reference standard. A deep learning algorithm interpreted CXRs for fracture detection on a per-patient, per-radiograph, and per-location level, and its diagnostic performance values were compared with the reference standard. AI diagnostic performance was computed by using cross-tables, and 95 % confidence intervals [CIs] were obtained by bootstrapping.

RESULTS

The final cohort included 312 male and 288 female with a mean age of 8.9±4.5 years. Three undred CXRs (50 %) were positive for fractures, according to the reference standard. For all fractures, the AI tool showed a per-patient 91.3% (95%CIs = 87.6-94.3) sensitivity, 76.7% (71.5-81.3) specificity, and 84% (82.1-86.0) accuracy. In the per-radiograph analysis the AI tool showed 85% (81.9-87.8) sensitivity, 88.5% (86.3-90.4) specificity, and 87.2% (85.7-89.6) accuracy. In the per-location analysis, the AI tool identified 606 bounding boxes: 472 (77.9 %) were correct, 110 (18.1 %) incorrect, and 24 (4.0 %) were not-overlapping.

CONCLUSION

The AI algorithm provides good overall diagnostic performance for detecting appendicular fractures in pediatric patients.

Artificial intelligence-based detection of paediatric appendicular skeletal fractures: performance and limitations for common fracture types and locations

BACKGROUND

Research into artificial intelligence (AI)-based fracture detection in children is scarce and has disregarded the detection of indirect fracture signs and dislocations.

OBJECTIVE

To assess the diagnostic accuracy of an existing AI-tool for the detection of fractures, indirect fracture signs, and dislocations.

MATERIALS AND METHODS

An AI software, BoneView (Gleamer, Paris, France), was assessed for diagnostic accuracy of fracture detection using paediatric radiology consensus diagnoses as reference. Radiographs from a single emergency department were enrolled retrospectively going back from December 2021, limited to 1,000 radiographs per body part. Enrolment criteria were as follows: suspected fractures of the forearm, lower leg, or elbow; age 0-18 years; and radiographs in at least two projections.

RESULTS

Lower leg radiographs showed 607 fractures. Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were high (87.5%, 87.5%, 98.3%, 98.3%, respectively). Detection rate was low for toddler's fractures, trampoline fractures, and proximal tibial Salter-Harris-II fractures. Forearm radiographs showed 1,137 fractures. Sensitivity, specificity, PPV, and NPV were high (92.9%, 98.1%, 98.4%, 91.7%, respectively). Radial and ulnar bowing fractures were not reliably detected (one out of 11 radial bowing fractures and zero out of seven ulnar bowing fractures were correctly detected). Detection rate was low for styloid process avulsions, proximal radial buckle, and complete olecranon fractures. Elbow radiographs showed 517 fractures. Sensitivity and NPV were moderate (80.5%, 84.7%, respectively). Specificity and PPV were high (94.9%, 93.3%, respectively). For joint effusion, sensitivity, specificity, PPV, and NPV were moderate (85.1%, 85.7%, 89.5%, 80%, respectively). For elbow dislocations, sensitivity and PPV were low (65.8%, 50%, respectively). Specificity and NPV were high (97.7%, 98.8%, respectively).

CONCLUSIONS

The diagnostic performance of BoneView is promising for forearm and lower leg fractures. However, improvement is mandatory before clinicians can rely solely on AI-based paediatric fracture detection using this software.

Detecting pediatric wrist fractures using deep-learning-based object detection

BACKGROUND

Missed fractures are the leading cause of diagnostic error in the emergency department, and fractures of pediatric bones, particularly subtle wrist fractures, can be misidentified because of their varying characteristics and responses to injury.

OBJECTIVE

This study evaluated the utility of an object detection deep learning framework for classifying pediatric wrist fractures as positive or negative for fracture, including subtle buckle fractures of the distal radius, and evaluated the performance of this algorithm as augmentation to trainee radiograph interpretation.

MATERIALS AND METHODS

We obtained 395 posteroanterior wrist radiographs from unique pediatric patients (65% positive for fracture, 30% positive for distal radial buckle fracture) and divided them into train (n = 229), tune (n = 41) and test (n = 125) sets. We trained a Faster R-CNN (region-based convolutional neural network) deep learning object-detection model. Two pediatric and two radiology residents evaluated radiographs initially without the artificial intelligence (AI) assistance, and then subsequently with access to the bounding box generated by the Faster R-CNN model.

RESULTS

The Faster R-CNN model demonstrated an area under the curve (AUC) of 0.92 (95% confidence interval [CI] 0.87-0.97), accuracy of 88% (n = 110/125; 95% CI 81-93%), sensitivity of 88% (n = 70/80; 95% CI 78-94%) and specificity of 89% (n = 40/45, 95% CI 76-96%) in identifying any fracture and identified 90% of buckle fractures (n = 35/39, 95% CI 76-97%). Access to Faster R-CNN model predictions significantly improved average resident accuracy from 80 to 93% in detecting any fracture (P < 0.001) and from 69 to 92% in detecting buckle fracture (P < 0.001). After accessing AI predictions, residents significantly outperformed AI in cases of disagreement (73% resident correct vs. 27% AI, P = 0.002).

CONCLUSION

An object-detection-based deep learning approach trained with only a few hundred examples identified radiographs containing pediatric wrist fractures with high accuracy. Access to model predictions significantly improved resident accuracy in diagnosing these fractures.

Australian children's foot, ankle and leg problems in primary care: a secondary analysis of the Bettering the Evaluation and Care of Health (BEACH) data

OBJECTIVES

To explore children's foot, ankle and leg consultation patterns and management practices in Australian primary care.

DESIGN

Cross-sectional, retrospective study.

SETTING

Australia Bettering the Evaluation and Care of Health program dataset.

PARTICIPANTS

Data were extracted for general practitioners (GPs) and patients <18 years from April 2000 to March 2016 inclusive.

MAIN OUTCOME MEASURES

Demographic characteristics: sex, GP age groups (ie, <45, 45-54, 55+ years), GP country of training, patient age grouping (0-4, 5-9, 10-14, 15-18 years), postcode, concession card status, indigenous status, up to three patient encounter reasons, up to four encounter problems/diagnoses and the clinical management actioned by the GP.

RESULTS

Children's foot, ankle or leg problems were managed at a rate of 2.05 (95% CI 1.99 to 2.11) per 100 encounters during 229 137 GP encounters with children. There was a significant increase in the rate of foot, ankle and leg problems managed per 100 children in the population, from 6.1 (95% CI 5.3 to 6.8) in 2005-2006 to 9.0 (95% CI 7.9 to 10.1) in 2015-2016. Management of children's foot, ankle and leg problems were independently associated with male patients (30% more than female), older children (15-18 years were 7.1 times more than <1 years), male GPs (13% more) and younger GPs (<45 years of age 13% more than 55+). The top four most frequently managed problems were injuries (755.9 per 100 000 encounters), infections (458.2), dermatological conditions (299.4) and unspecified pain (176.3). The most frequently managed problems differed according to age grouping.

CONCLUSIONS

Children commonly present to GPs for foot, ankle and leg problems. Presentation frequencies varied according to age. Unexpectedly, conditions presenting commonly in adults, but rarely in children, were also frequently recorded. This data highlights the importance of initiatives supporting contemporary primary care knowledge of diagnoses and management of paediatric lower limb problems to minimise childhood burden of disease.

Ultrasonography of the bone surface in children: normal and pathological findings in the bone cortex and periosteum

Ultrasound (US) is widely used in pediatric musculoskeletal pathology at all ages. Although the focus is often on soft tissues, joints and cartilage, the examiner might be confronted with changes in the underlying bone surface that are important to understand and integrate in the diagnosis. This article illustrates the normal US aspects of the cortical bone surface and periosteum, as well as the most common US anomalies seen in infections, trauma and bone tumors in children.

A Case of Acute Plastic Deformation of the Forearm in a Medieval Hispano-Mudejar Skeleton (13-14th Centuries AD)

INTRODUCTION

Acute plastic deformation refers to a traumatic bending or bowing without a detectable cortical defect.

CASE PRESENTATION AND DISCUSSION

We describe a rare case from an individual that was exhumed from the Hispano-Mudejar necropolis in Uceda (Guadalajara, Spain) dated between the 13th and 14th centuries AD. The case corresponds to an adult woman, with a bowing involvement of the left ulna and radius. After making the differential diagnosis with various pathologies likely to present with this alteration, we reached the diagnosis of acute plastic deformation of the forearm through external and radiological examination and comparison with the healthy contralateral forearm.

CONCLUSIONS

Acute plastic deformation is a rare traumatic injury, not described until the last century and only rarely described in palaeopathological contexts. We contribute a new case, the first being sufficiently documented, contributing to the knowledge and diagnosis of this type of trauma in the ancient bone, while deepening the knowledge of the living conditions of the medieval Mudejar population of Uceda.

Pediatric radius torus fractures in x-rays-how computer vision could render lateral projections obsolete

It is an indisputable dogma in extremity radiography to acquire x-ray studies in at least two complementary projections, which is also true for distal radius fractures in children. However, there is cautious hope that computer vision could enable breaking with this tradition in minor injuries, clinically lacking malalignment. We trained three different state-of-the-art convolutional neural networks (CNNs) on a dataset of 2,474 images: 1,237 images were posteroanterior (PA) pediatric wrist radiographs containing isolated distal radius torus fractures, and 1,237 images were normal controls without fractures. The task was to classify images into fractured and non-fractured. In total, 200 previously unseen images (100 per class) served as test set. CNN predictions reached area under the curves (AUCs) up to 98% [95% confidence interval (CI) 96.6%-99.5%], consistently exceeding human expert ratings (mean AUC 93.5%, 95% CI 89.9%-97.2%). Following training on larger data sets CNNs might be able to effectively rule out the presence of a distal radius fracture, enabling to consider foregoing the yet inevitable lateral projection in children. Built into the radiography workflow, such an algorithm could contribute to radiation hygiene and patient comfort.

Radiologic Evaluation of the Child with a Limp

A child presenting with a limp can present a diagnostic challenge to pediatricians. Clinical presentation, age, and history all contribute to the initial differential diagnosis; however, imaging plays a key role in the ultimate diagnosis, and the correct imaging study is essential to save time and health care expenses. This article will present a few of the more common causes of a limp and the recently updated imaging recommendations from the American College of Radiology to aid in final diagnosis. [Pediatr Ann. 2020;49(9):e395-e402.].

Frequently Missed Fractures in Pediatric Trauma: A Pictorial Review of Plain Film Radiography

Missed fractures are common in pediatric trauma patients. Pediatric bone differs from adult bone in its composition and response to injury, leading to fracture patterns that may be subtle, radiographically unfamiliar, and challenging to distinguish from normal variation. Familiarity with the unique fracture types of the pediatric skeleton and site-specific injury patterns is critical, because prompt diagnosis can significantly alter clinical management and outcome. This article examines the unique features of pediatric bone contributing to missed fractures, the incidence of missed fractures, common injury types of the pediatric skeleton, and frequently missed site-specific fracture patterns, highlighting problem-solving techniques for challenging cases.

The different distribution of enzymatic collagen cross-links found in adult and children bone result in different mechanical behavior of collagen

Enzymatic collagen cross-linking has been shown to play an important role in the macroscopic elastic and plastic deformation of bone across ages. However, its direct contribution to collagen fibril deformation is unknown. The aim of this study is to determine how covalent intermolecular connections from enzymatic collagen cross-links contribute to collagen fibril elastic and plastic deformation of adults and children's bone matrix. We used ex vivo data previously obtained from biochemical analysis of children and adults bone samples (n = 14; n = 8, respectively) to create 22 sample-specific computational models of cross-linked collagen fibrils. By simulating a tensile test for each fibril, we computed the modulus of elasticity (E), ultimate tensile and yield stress (σ_u and σ_y), and elastic, plastic and total work (W_e, W_p and W_tot) for each collagen fibril. We present a novel difference between children and adult bone in the deformation of the collagen phase and suggest a link between collagen fibril scale and macroscale for elastic behavior in children bone under the influence of immature enzymatic cross-links. We show a parametric linear correlation between W_e and immature enzymatic collagen cross-links at the collagen fibril scale in the children population that is similar to the one we found at the macroscale in our previous study. Finally, we suggest the key role of covalent intermolecular connections to stiffness parameters (e.g. elastic modulus and W_e) in children's collagen fibril and to toughness parameters in adult's collagen fibril, respectively.

Computer-Aided Detection (CADx) for Plastic Deformation Fractures in Pediatric Forearm

Bowing fractures are incomplete fractures of tubular long bones, often observed in pediatric patients, where plain radiographic film is the non-invasive imaging modality of choice in routine radiological workflow. Due to weak association between bent bone and distinct cortex disruption, bowing fractures may not be diagnosed properly while reading plain radiography. Missed fractures and dislocations are common in accidents and emergency practice, particularly in children. These missed injuries can result in more complicated treatment or even long-term disability. The most common reason for missed fractures is that junior radiologists or physicians lack expertise in pediatric skeletal injury diagnosis. Not only is additional radiation exposure inevitable in the case of misdiagnosis, but other consequences include the patient's prolonged uncomfortableness and possible unnecessary surgical procedures. Therefore, a computerized image analysis system, which would be secondary to the radiologists' interpretations, may reduce adverse effects and improve the diagnostic rates of bowing fracture (detection and quantification). This system would be highly desirable and particularly useful in emergency rooms. To address this need, we investigated and developed a new Computer Aided Detection (CADx) system for pediatric bowing fractures. The proposed system has been tested on 226 cases of pediatric forearms with bowing fractures with respect to normal controls. Receiver operation characteristic (ROC) curves show that the sensitivity and selectivity of the developed CADx system are satisfactory and promising. A clinically feasible graphical user interface (GUI) was developed to serve the practical needs in the emergency room as a diagnostic reference. The developed CADx system also has strong potential to train radiology residents for diagnosing pediatric forearm bowing fractures.

Learning from diagnostic errors: a good way to improve education in radiology

PURPOSE

To evaluate the causes and the main categories of diagnostic errors in radiology as a method for improving education in radiology.

MATERIAL AND METHODS

A Medline search was performed using PubMed (National Library of Medicine, Bethesda, MD) for original research publications discussing errors in diagnosis with specific reference to radiology. The search strategy employed different combinations of the following terms: (1) diagnostic radiology, (2) radiological error and (3) medical negligence. This review was limited to human studies and to English-language literature. Two authors reviewed all the titles and subsequently the abstracts of 491 articles that appeared pertinent. Additional articles were identified by reviewing the reference lists of relevant papers. Finally, the full text of 75 selected articles was reviewed.

RESULTS

Several studies show that the etiology of radiological error is multi-factorial. The main category of claims against radiologists includes the misdiagnoses. Radiologic "misses" typically are one of two types: either missed fractures or missed diagnosis of cancer. The most commonly missed fractures include those in the femur, the navicular bone, and the cervical spine. The second type of "miss" is failure to diagnose cancer. Lack of appreciation of lung nodules on chest radiographs and breast lesions on mammograms are the predominant problems.

CONCLUSION

Diagnostic errors should be considered not as signs of failure, but as learning opportunities.