Evaluating the impact of artificial intelligence-assisted image analysis on the diagnostic accuracy of front-line clinicians in detecting fractures on plain X-rays (FRACT-AI): protocol for a prospective observational study.

INTRODUCTION: Missed fractures are the most frequent diagnostic error attributed to clinicians in UK emergency departments and a significant cause of patient morbidity. Recently, advances in computer vision have led to artificial intelligence (AI)-enhanced model developments, which can support clinicians in the detection of fractures. Previous research has shown these models to have promising effects on diagnostic performance, but their impact on the diagnostic accuracy of clinicians in the National Health Service (NHS) setting has not yet been fully evaluated. METHODS AND ANALYSIS: A dataset of 500 plain radiographs derived from Oxford University Hospitals (OUH) NHS Foundation Trust will be collated to include all bones except the skull, facial bones and cervical spine. The dataset will be split evenly between radiographs showing one or more fractures and those without. The reference ground truth for each image will be established through independent review by two senior musculoskeletal radiologists. A third senior radiologist will resolve disagreements between two primary radiologists. The dataset will be analysed by a commercially available AI tool, BoneView (Gleamer, Paris, France), and its accuracy for detecting fractures will be determined with reference to the ground truth diagnosis. We will undertake a multiple case multiple reader study in which clinicians interpret all images without AI support, then repeat the process with access to AI algorithm output following a 4-week washout. 18 clinicians will be recruited as readers from four hospitals in England, from six distinct clinical groups, each with three levels of seniority (early-stage, mid-stage and later-stage career). Changes in the accuracy, confidence and speed of reporting will be compared with and without AI support. Readers will use a secure web-based DICOM (Digital Imaging and Communications in Medicine) viewer (www.raiqc.com), allowing radiograph viewing and abnormality identification. Pooled analyses will be reported for overall reader performance as well as for subgroups including clinical role, level of seniority, pathological finding and difficulty of image. ETHICS AND DISSEMINATION: The study has been approved by the UK Healthcare Research Authority (IRAS 310995, approved on 13 December 2022). The use of anonymised retrospective radiographs has been authorised by OUH NHS Foundation Trust. The results will be presented at relevant conferences and published in a peer-reviewed journal. TRIAL REGISTRATION NUMBERS: This study is registered with ISRCTN (ISRCTN19562541) and ClinicalTrials.gov (NCT06130397). The paper reports the results of a substudy of STEDI2 (Simulation Training for Emergency Department Imaging Phase 2).

Optimal measurement method for anterior instability on stress radiographs in anterior cruciate ligament tear: Considering the effect of static anterior tibial subluxation.

INTRODUCTION: Accurate assessment of anterior cruciate ligament (ACL) function is vital for guiding treatment. Nevertheless, the presence of tibial subluxation in the neutral position of a patient with an ACL injury may potentially introduce a confounding factor. This study aims to investigate whether tibial subluxation in the neutral position affects the diagnosis of anterior instability in patients with ACL injuries, potentially impacting the reliability and diagnostic accuracy of stress radiography. METHODS: This study included 88 patients: 30 with acute complete ACL tears (acute group), 28 with chronic complete ACL tears (chronic group), and 30 patients who underwent knee arthroscopic surgery other than ACL reconstruction (control group). Side-to-side differences (SSD) in stress radiography were measured using the Telos load status and the SSD of the gap between the Telos load and unload statuses. Diagnostic accuracy of the two methods was assessed using areas under the receiver operating characteristic curves (AUCs). RESULTS: The load SSD (5.92 ± 5.28 mm) was higher than the load-unload SSD (4.27 ± 5.99 mm) in the chronic group (P = 0.017). The load SSD demonstrated a significantly higher diagnostic value than that of the load-unload SSD in the combined group (AUC = 0.920 vs. 0.830; P = 0.012) and chronic group (AUC = 0.913 vs. 0.754; P = 0.002). After adjusting the symptoms for radiographic duration from 6 to 3 months in the chronic group, the load SSD exhibited a significantly higher diagnostic value (AUC = 0.902) than that of the load-unload SSD (AUC = 0.740; P < 0.001). CONCLUSION: The load SSD provides superior diagnostic accuracy compared to the load-unload SSD in ACL tear cases, where static anterior tibial subluxation may result in false negatives. Although load-unload SSD may have diagnostic value within the first 3 months post-injury, the load SSD method provides a reliable assessment of ACL function for patients beyond this timeframe.

Steroid-responsive anemia with bony dysplasias: What lurks behind!

Ghosal hematodiaphyseal dysplasia (GHDD) is an autosomal recessive disorder characterized by diaphyseal dysplasia of long bones, bone marrow fibrosis, and steroid-responsive anemia. Patients with this disease have a mutation in the thromboxane-AS1 (TBXAS1) gene located on chromosome 7q33.34. They present with short stature, varying grades of myelofibrosis, and, hence cytopenias. Patients with the above presentation were evaluated through clinical presentation, X-ray of long bones, bone marrow examinations, and confirmed by genetic testing. In this article, we present two cases: The first case is a 3-year-old boy who presented with progressive pallor and ecchymotic patches for a year. On investigation, he had bicytopenia and bone marrow fibrosis. His anemia was steroid responsive and was finally diagnosed as GHDD. The second case is a 20-month-old girl who presented with blood in stools, developmental delay, anemia, and increased intensity of long bones on X-ray. Since other investigations were normal, suspicion of GHDD was raised, and a genetic workup was conducted which suggested mutation in TBXAS1 gene, confirming the diagnosis of GHDD. Children with refractory anemia and cortical thickening on skeletogram should always be evaluated for dysplasias. Timely treatment with steroids reduces transfusion requirements and halts bone damage, thus leading to better growth and improved quality of life.

Musculoskeletal Issues in Children and Adolescents: Adolescent Idiopathic Scoliosis.

Adolescent idiopathic scoliosis (AIS) is the most common type of scoliosis in children older than 10 years in the United States. AIS is defined as a lateral spine curvature of 10° or more in the coronal plane, without congenital or neuromuscular comorbidities. The U.S. Preventive Services Task Force (USPSTF) and American Academy of Family Physicians (AAFP) do not recommend for or against AIS screening in asymptomatic patients. Physical examination includes the forward bend test with or without scoliometer, wherein scoliometer rotation between 5° and 7° warrants further evaluation with x-rays. Definitive diagnosis with x-rays allows for measurement of the Cobb angle. For Cobb angles less than 20°, watchful waiting and/or referral for physical therapy are indicated. Referral to a spine specialist for bracing is reasonable for curves between 20° and 26° and is recommended for curves between 26° and 45°. Surgical intervention is considered for initial Cobb angles greater than 40° and recommended for Cobb angles greater than 50°.

An unusual Ogden type V tibial tubercle avulsion fracture: a case report and literature review.

BACKGROUND: Ogden type V tibial tubercle avulsion fracture is an unusual type of physial injury. Thus, little is known about its mechanism of injury and treatment. The type of osteosynthesis is variable and depends on the experience of the surgeon. We commonly used cancellous screws fixation combined with tension band wiring for displaced fracture of the anterior tibial tuberosity. CASE PRESENTATION: The present manuscript describes a case of a Han nationality 13-year-old boy who presented with severe pain of the left knee, which began after landing following a high jump. He had no significant past medical history apart from a high body mass index of 30.3. Radiographs revealed that he had an unusual Ogden type V tibial tubercle avulsion fracture. He was treated by open reduction and combined fixation with cannulated screws and tension-band wiring. After 3 months, the fracture healed without any complications or knee symptoms with full range of motion. He underwent reoperation for symptomatic hardware, which was removed at 5 months after initial surgery, and returned to his prior level of sporting activity at 1 year follow-up. CONCLUSION: Our case suggests that excellent functional outcome could be achievable by open reduction with the combination of internal fixation and tension-band wiring for Ogden type V tibial tubercle avulsion fracture. This type of osteosynthesis could not only achieve anatomical reduction and stable fixation for such fractures, but also avoid further damage to the proximal tibial epiphysis, which prevents serious complications, such difference in leg length.

Pin fixation is an effective method for fixation of bunion osteotomy with various procedures: a retrospective cohort study.

BACKGROUND: Various fixation devices are available for bunion osteotomy. In this study, we evaluated the radiographic outcomes, postoperative complications, and recurrence rate in a series of hallux valgus deformities treated with various osteotomy procedures using a pin for the fixation of the osteotomy. METHODS: Two-hundred forty-seven patients with hallux valgus deformity managed with a Simple, Effective, Rapid and Inexpensive (SERI) osteotomy, distal chevron osteotomy, or proximal crescentic osteotomy and K-wire fixation were included. The mean follow-up of the patients was 53.9 ± 8.9 months. Radiographic evaluations included the assessment of the Hallux valgus angle (HVA), intermetatarsal angle (IMA), and union. Clinical evaluations included the assessment of the range of motion, pain in the first metatarsophalangeal joint, and patient satisfaction. RESULTS: In the last visit, the mean improvement of HVA was 23.9 ± 9.1º (P < 0.001). The mean IMA improvement was 6.1 ± 6º (P < 0.001). The mean metatarsophalangeal flexion and extension were 33 ± 10.7º and 34.6 ± 9.2º, respectively. Postoperative complications included pin tract infection in eight (3.2%) patients, deep infection in five (2%) patients, and early pin complication in four (1.6%) patients. Recurrence was observed in five (2%) patients. Twenty-three (9.3%) patients had slight pain in the last follow-up. The mean surgical time was smaller in the SERI osteotomy (P < 0.001). The mean hospitalization period was longer in the proximal osteotomy group (P = 0.039). The mean metatarsophalangeal flexion and extension were significantly smaller in the distal chevron osteotomy (P = 0.046 and P = 0.037, respectively). 90% of patients were satisfied or very satisfied with the surgical outcomes. CONCLUSION: K-wire fixation is a safe and effective device for the fixation of bunion osteotomy, and this effectiveness is even higher with SERI and proximal crescentic osteotomy.

Nineteen years of radiographic screening: Impact of sepsis and evolution of osteochondrosis dissecans prevalence in Walloon sport horses born between 2004 and 2022.

Osteochondrosis dissecans is a particularly common developmental orthopaedic disorder in equines. Among its causes, the hereditary component is no longer a matter of debate, and, for several decades, the various studbooks for sport horses have been selecting stallions in order to limit the prevalence of this condition in the population. However, to our knowledge, no study has evaluated changes in the prevalence of osteochondrosis dissecans over time through longitudinal monitoring of a population of sport horses. The study presented here is part of a large screening programme for developmental orthopaedic pathologies in Wallonia (Belgium) and assessed the presence of these lesions over a period of 19 years in the Walloon sport horse population according to constant and standardised sampling and diagnostic criteria. The effects of breeding conditions and infection in foals were also assessed by means of questionnaires. The results showed no significant change in the prevalence of osteochondrosis dissecans in a population of 1099 individuals born between 2004 and 2022. Furthermore, individuals who had suffered from sepsis during their growing period were very significantly predisposed (p < 0.001) to the development of osteochondrosis dissecans compared to a control group, with respectively 14/21 (67%) and 103/364 (28%) of individuals affected. This study suggests that the selection programmes applied to the population studied are not sufficiently strong or adapted to reduce the prevalence of osteochondrosis dissecans in the population over a period of 19 years. Moreover, this study confirms that foals with sepsis and concurrent osteochondrosis dissecans lesions should not necessarily be excluded from breeding programmes on this basis.

Factors Associated with Patient Satisfaction After Periacetabular Osteotomy.

BACKGROUND: The periacetabular osteotomy (PAO) was developed to treat acetabular dysplasia and avoid the progression of osteoarthritis. Prior research has correlated demographic and radiographic variables with postoperative patient outcomes but with limited focus on patient expectations and correlation with satisfaction. The purpose of this study was to determine whether any such associations exist with the PAO. METHODS: An anonymous, de-identified satisfaction survey was applied to patients undergoing a PAO between April 2017 and April 2019. Forty patients (26 females) who underwent PAOs were included in the final analysis. The average age of the cohort was 22.0 ± 5.1 years of age. All patients had a minimum of 12 months of follow-up from the date of surgery. Patient demographics, diagnosis, and complications were recorded. Radiographs were then analyzed to determine pre- and postoperative lateral center edge angle (LCEA) and Tönnis roof angle, and correlations between satisfaction and radiographic analyses were performed. Statistical analysis included non-parametric Spearman's correlation and receiver operator characteristic. Statistical significance was set at p < 0.05. RESULTS: Thirty (75%) patients were satisfied with their outcome. There were no statistically significant associations between patient age or sex and postoperative satisfaction (p > 0.05). The average LCEA was 10.9° ± 6.9° preoperatively and 26.0° ± 4.2° postoperatively with an average change of 15.1° ± 5.6°. The average Tönnis angle was 18.8° ± 3.3° preoperatively, decreasing to 11.6° ± 3.2° postoperatively with an average change of -7.2° ± 3.2°. Interobserver reliability was high, ranging from r = 0.782, p < 0.001 for postoperative Tönnis angle to r = 0.958, p < 0.001 for preoperative LCEA. Preoperative LCEA correlated positively with satisfaction, r = 0.351, p = 0.027. Logistic regression demonstrated that for every increasing degree of preoperative LCEA, odds of postoperative satisfaction increased by 1.13 (95% CI: 1.01 to 1.27), p = 0.034. CONCLUSION: This study suggests that postoperative patient satisfaction after PAO may be associated with preoperative patient demographics such as LCEA. It also suggests that more dysplastic hips may have lower rates of postoperative satisfaction than patients with less severe deformity. These associations warrant further study, which may yield prognostic value for future surgery.

Bilateral tuberculous dactylitis of both hand and feet in a female toddler: a case report on a rare presentation of skeletal tuberculosis in children.

BACKGROUND: Even though tuberculosis is a common disease among children in developing countries, tuberculous dactylitis is an uncommon form of Skeletal tuberculosis specially with involvement of both the hands and feet. CASE PRESENTATION: A one-and-a-half-year-old previously healthy female Ethiopian toddler presented to our pediatric outpatient clinic with a history of two-month duration of painful multiple swellings over both her hands and feet. The swelling involved the proximal phalanx of the left index finger, dorsum of the right hand, and dorsum of both feet over the first metatarsal bone. Physical examination, radiologic findings, and histopathology suggested tuberculous dactylitis. The patient was treated with anti-tuberculosis drugs for one year and she showed clinical and radiologic improvement and recovery. CONCLUSION: Tubercular dactylitis should be considered in the differential diagnosis of children from endemic areas presenting with bone and joint pain or swelling. Our experience of a twelve-month course of antitubercular treatment, which is in line with WHO recommendations, for skeletal tuberculosis, showed excellent outcomes.

Automated Association for Osteosynthesis Foundation and Orthopedic Trauma Association classification of pelvic fractures on pelvic radiographs using deep learning.

High-energy impacts, like vehicle crashes or falls, can lead to pelvic ring injuries. Rapid diagnosis and treatment are crucial due to the risks of severe bleeding and organ damage. Pelvic radiography promptly assesses fracture extent and location, but struggles to diagnose bleeding. The AO/OTA classification system grades pelvic instability, but its complexity limits its use in emergency settings. This study develops and evaluates a deep learning algorithm to classify pelvic fractures on radiographs per the AO/OTA system. Pelvic radiographs of 773 patients with pelvic fractures and 167 patients without pelvic fractures were retrospectively analyzed at a single center. Pelvic fractures were classified into types A, B, and C using medical records categorized by an orthopedic surgeon according to the AO/OTA classification system. Accuracy, Dice Similarity Coefficient (DSC), and F1 score were measured to evaluate the diagnostic performance of the deep learning algorithms. The segmentation model showed high performance with 0.98 accuracy and 0.96-0.97 DSC. The AO/OTA classification model demonstrated effective performance with a 0.47-0.80 F1 score and 0.69-0.88 accuracy. Additionally, the classification model had a macro average of 0.77-0.94. Performance evaluation of the models showed relatively favorable results, which can aid in early classification of pelvic fractures.

[Radiographic measurements of acetabular component orientation with Widmer´s method in total hip arthroplasty. Descriptive case series]. / Mediciones radiográficas de la orientación del componente acetabular con el método de Widmer en la artroplastía total de cadera. Serie de casos descriptiva.

INTRODUCTION: total hip arthroplasty (THA) is one of the most performed surgeries worldwide, with high satisfaction rates. The orientation of the acetabular component has a direct impact on the risk of dislocation, recently with the support of robotic surgery the margin of error in implant placement has decreased; however, the conventional technique even without fluoroscopic support continues to have satisfactory results within the safety zone. MATERIAL AND METHODS: retrospective, cross-sectional, descriptive case series of patients treated with THA at Hospital General Xoco between 2022 and 2024. Degrees of anteversion and inclination were measured with Widmer's method on postoperative radiographs. RESULTS: the radiographs of 113 patients were studied, 80 female and 33 male, with a mean age of 63.2 ± 13.01 years (95% CI: 60.6-65.4), a mean inclination of 42.2° ± 8.1° (95% CI: 40.7-43.2) and anteversion of 14.3° ± 8.5° (95% CI: 12.5-15.4); 76% of the population was within Lewinnek safe zone; by etiology: osteoarthrosis 74%, sequelae of dysplasia 68% and intracapsular fracture 82%; difference between the values of the affected side: left 65%, right 83%, of 3.9° and 4.7°/6.4o and 9° in relation to the overall values of the population. CONCLUSION: in our population undergoing THA, without the use of robotic technique or support of imaging studies, anteversion and inclination figures were recorded within the Lewinnek safety parameters with a conventional method.

INTRODUCCIÓN: la artroplastía total de cadera (ATC) es una de las cirugías más realizadas a nivel mundial, con altos porcentajes de satisfacción. La orientación del componente acetabular tiene impacto directo en el riesgo de luxación; recientemente, con el apoyo de la cirugía robótica, el margen de error en la colocación de los implantes ha disminuido; sin embargo, la técnica convencional, incluso sin apoyo fluoroscópico, continúa teniendo resultados satisfactorios dentro de la zona de seguridad. MATERIAL Y MÉTODOS: serie de casos retrospectiva, transversal y descriptiva, de pacientes tratados con ATC en Hospital General Xoco entre 2022 y 2024. Se midieron los grados de anteversión e inclinación con el método de Widmer en las radiografías postoperatorias. RESULTADOS: se estudiaron las radiografías de 113 pacientes, 80 mujeres y 33 hombres, con edad media de 63.2 ± 13.01 años (IC95%: 60.6-65.4), se obtuvo una inclinación media de 42.2° ± 8.1° (IC95%: 40.7-43.2) y anteversión de 14.3° ± 8.5° (IC95% 12.5-15.4); 76% de la población se encontraba dentro de la zona segura de Lewinnek; por etiología: osteoartrosis 74%, secuelas de displasia 68% y fractura intracapsular 82%; diferencia entre los valores del lado afectado: izquierdo 65%, derecho 83%, de 3.9° y 4.7°/6.4° y 9° en relación con los valores globales de la población. CONCLUSIÓN: en nuestra población sometida a ATC, sin uso de técnica robótica o apoyo de estudios de imagen, se registraron cifras de anteversión e inclinación dentro de los parámetros de seguridad de Lewinnek con un método convencional.

[Study on automatic and rapid diagnosis of distal radius fracture by X-ray].

This article aims to combine deep learning with image analysis technology and propose an effective classification method for distal radius fracture types. Firstly, an extended U-Net three-layer cascaded segmentation network was used to accurately segment the most important joint surface and non joint surface areas for identifying fractures. Then, the images of the joint surface area and non joint surface area separately were classified and trained to distinguish fractures. Finally, based on the classification results of the two images, the normal or ABC fracture classification results could be comprehensively determined. The accuracy rates of normal, A-type, B-type, and C-type fracture on the test set were 0.99, 0.92, 0.91, and 0.82, respectively. For orthopedic medical experts, the average recognition accuracy rates were 0.98, 0.90, 0.87, and 0.81, respectively. The proposed automatic recognition method is generally better than experts, and can be used for preliminary auxiliary diagnosis of distal radius fractures in scenarios without expert participation.

Association between folate intake and radiographic progression, pain function scores in subjects with radiographic knee osteoarthritis: Data from the osteoarthritis initiative.

BACKGROUND: Folate has an important role in the functioning of the musculoskeletal system, including modulation of inflammation, immunity, cartilage regeneration, prevention of osteoporosis, and maintenance of muscle strength, but evidence on the association between folate intake and knee pain, functional scores, and radiographic progression in patients with knee osteoarthritis (OA) is still limited. METHODOLOGY: Our population-based cohort was extracted from the osteoarthritis initiative (OAI), focusing on individuals with prevalent radiographic knee OA (with a Kellgren-Lawrence score ≥2). Folate consumption was determined using the food frequency questionnaire. Data regarding the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) scores and radiographic readings were collected over 48 months. We analyzed the compiled data using generalized additive mixed models. RESULTS: Our cohort consisted of 1472 OA patients (626 men and 846 women, mean [SD] age 62.35 [8.92]). At the 48-month follow-up, we observed a significant correlation between higher folate intake and a slower progression of knee pain and functional scores, as evidenced by a statistically significant decrease in the WOMAC total score, WOMAC pain subscale score, and WOMAC function/disability subscale score (p < .05). The fully adjusted models estimated a reduction of -0.028 points per 50 µg/1000 kcal of daily folate intake on the WOMAC pain subscale, -0.117 points on the WOMAC function subscale, and -0.160 points on the total WOMAC scale. Furthermore, our nonparametric fit analysis suggested that a higher intake of folate might decelerate the radiographic progression of OA. Stratified analyses indicated that an increase in folate consumption might particularly benefit men, older adults, overweight and obese individuals, and those with a higher dietary fiber intake. CONCLUSION: Higher folate intake is correlated with improved knee function and reduced pain in patients with knee OA and might deter the radiographic progression of OA. The benefits appear to be more pronounced in men, older adults, overweight and obese individuals, and those with a higher dietary fiber intake.

Orthoses Effect on Radiographic Measurements of Hallux Abducto Valgus: A Systematic Review.

BACKGROUND: Hallux abducto valgus (HAV), commonly referred to as a bunion, is a prevalent foot deformity with multifactorial etiology, including genetic predisposition, biomechanical abnormalities, and footwear choices. HAV is often managed conservatively with orthotic devices aimed at reducing deformity progression and alleviating associated pain. However, the effectiveness of orthotics in altering radiographic measurements of HAV remains debated. This systematic review aims to evaluate the impact of orthotic interventions on radiographic parameters. METHODS: A comprehensive search of five electronic databases-PubMed, Cochrane Library, CINAHL, Medline, and EMBASE-was conducted, covering the period from inception to March 2021. The search included terms related to HAV and orthotic interventions. Studies were included if they provided pre- and post-treatment radiographic measurements of halux abductus angle (HAA), hallux valgus angle (HVA), or intermetatarsal angle (IMA) and involved the use of orthotics. A total of 523 references were initially identified, with five studies meeting the inclusion criteria for review. Data extraction focused on study characteristics, orthotic type, radiographic measurements, and follow-up duration. RESULTS: The systematic review found insufficient high-quality evidence to support the effectiveness of orthotics in slowing the progression of HAV deformity. Of the five studies included, results were mixed: two studies reported changes in radiographic measurements following orthotic use, though one of these involved toe spacers rather than traditional orthotics. The remaining studies found no significant impact of orthotics on HAA, HVA, or IMA. The heterogeneity in study design, population, and orthotic types limited the ability to draw robust conclusions. CONCLUSIONS: Current evidence does not conclusively support the use of orthotics for altering the radiographic progression of HAV. The reviewed studies highlight significant variability in outcomes and suggest that while orthotics may provide symptomatic relief, their role in deformity correction remains unclear. Whether and to what extent orthotics affect common radiographic measures of HAV deformity should be proven with further studies and investigation.

A few-shot learning framework for the diagnosis of osteopenia and osteoporosis using knee X-ray images.

OBJECTIVE: We developed a few-shot learning (FSL) framework for the diagnosis of osteopenia and osteoporosis in knee X-ray images. METHODS: Computer vision models containing deep convolutional neural networks were fine-tuned to enable generalization from natural images (ImageNet) to chest X-ray images (normal vs. pneumonia, base images). Then, a series of automated machine learning classifiers based on the Euclidean distances of base images were developed to make predictions for novel images (normal vs. osteopenia vs. osteoporosis). The performance of the FSL framework was compared with that of junior and senior radiologists. In addition, the gradient-weighted class activation mapping algorithm was used for visual interpretation. RESULTS: In Cohort #1, the mean accuracy (0.728) and sensitivity (0.774) of the FSL models were higher than those of the radiologists (0.512 and 0.448). A diagnostic pipeline of FSL model (first)-radiologists (second) achieved better performance (0.653 accuracy, 0.582 sensitivity, and 0.816 specificity) than radiologists alone. In Cohort #2, the diagnostic pipeline also showed improved performance. CONCLUSIONS: The FSL framework yielded practical performance with respect to the diagnosis of osteopenia and osteoporosis in comparison with radiologists. This retrospective study supports the use of promising FSL methods in computer-aided diagnosis tasks involving limited samples.

Detecting pediatric appendicular fractures using artificial intelligence.

OBJECTIVE: The primary objective was to assess the diagnostic accuracy of a deep learning-based artificial intelligence model for the detection of acute appendicular fractures in pediatric patients presenting with a recent history of trauma to the emergency department. The secondary goal was to examine the effect of assistive support on the emergency doctor's ability to detect fractures. METHODS: The dataset was 5,150 radiographs of which 850 showed fractures, while 4,300 radiographs did not show any fractures. The process utilized 4,532 (88%) radiographs, inclusive of both fractured and non-fractured radiographs, in the training phase. Subsequently, 412 (8%) radiographs were appraised during validation, and 206 (4%) were set apart for the testing phase. With and without artificial intelligence assistance, the emergency doctor reviewed another set of 2,000 radiographs (400 fractures and 600 non-fractures each) for labeling in the second test. RESULTS: The artificial intelligence model showed a mean average precision 50 of 89%, a specificity of 92%, a sensitivity of 90%, and an F1 score of 90%. The confusion matrix revealed that the model trained with artificial intelligence achieved accuracies of 93 and 95% in detecting fractures, respectively. Artificial intelligence assistance improved the reading sensitivity from 93.7% (without assistance) to 97.0% (with assistance) and the reading accuracy from 88% (without assistance) to 94.9% (with assistance). CONCLUSION: A deep learning-based artificial intelligence model has proven to be highly effective in detecting fractures in pediatric patients, enhancing the diagnostic capabilities of emergency doctors through assistive support.

Sacroiliac and spine imaging in spondyloarthritis: Does phenotype or sex matter?

BACKGROUND: Spondyloarthritis (SpA) encompasses a spectrum of immune-mediated inflammatory conditions primarily affecting the axial skeleton, including sacroiliitis and spondylitis, each with distinct features. This study aimed to investigate imaging disparities, focusing on sacroiliac magnetic resonance and spine radiography, across phenotypes and between males and females in axial SpA. METHOD: A cross-sectional study was conducted to assess clinical data, laboratory findings, magnetic resonance imaging (MRI) scores of sacroiliac joints using the Spondyloarthritis Research Consortium of Canada (SPARCC) and Sacroiliac Joint Structural Score (SSS), and cervical and lumbar spine radiographs utilizing the Modified Stoke Ankylosing Spondylitis Spine Score (mSASSS). The study aimed to compare these parameters between two groups: axial spondyloarthritis (axSpA, radiographic and non-radiographic) and axial psoriatic arthritis (axPsA), as well as between males and females. RESULTS: Ninety-four patients were included, with 62 patients in the axSpA group and 32 patients in the axPsA group. There were no differences in disease activity, mobility, radiographic damage in the spine (Modified Stoke Ankylosing Spondylitis Spine Score- mSASSS), or sacroiliac magnetic resonance imaging (MRI) scores (Spondyloarthritis Research Consortium of Canada Magnetic Resonance Imaging Index - SPARCC and Sacroiliac Joint Structural Score - SSS) between the two phenotypes. Regarding sex, in imaging exams, men had higher mSASSS (p = 0.008), SSS (p = 0.001), and fat metaplasia (MG) score based on SSS (p = 0.001), while women had significantly higher SPARCC scores (p = 0.039). In the male group, the presence of HLA-B27 allele had an impact on more structural lesions on MRI (SSS), p = 0.013. CONCLUSION: In this study, imaging of sacroiliac joints and spine in patients with axial SpA did not show differences in phenotypes but did reveal differences based on sex, which may have an impact on future diagnostic recommendations. Further studies are needed to confirm these findings.

Age Prediction Using Pixel Value Sum from Radiographic Proximal Femur in a Thai Population / Predicción de la Edad Utilizando la Suma de Valores de Pixeles del Fémur Proximal Radiográfico en una Población Tailandesa

SUMMARY: The present study aimed to investigate the utility of the proximal femur in the forensic age estimation by assessing changes in bone densities through radiographs. Using Otsu's threshold, bone density was quantified by counting all white pixel values within selected regions of interest, which include femoral head (FH), femoral neck (FN), Ward's triangle (WT), and greater trochanter (GT) from 354 left femora of Northern Thai descent. The pixel width of medullary cavity (MC) was also estimated. Furthermore, the study evaluated the performance of linear regression (LR) models for age estimation from radiographic images of proximal femora. Negative correlations were observed between FH, FN, WT, and GT pixel intensity with the age-at-death of the samples, with females exhibiting stronger correlations than males. Moreover, a positive correlation was found between age and MC width in female samples, while male MC widths did not show any relationship with increasing age. The results showed a slight difference between the LR model applied to both sexes, which integrated all variables, and the alternative configuration that only utilized relevant attributes. Both models exhibited similar performance, with a narrow range of root mean square error (RMSE) values, ranging from 12.67 to 12.71 years, and a correlation coefficient range of 0.51 to 0.52. For females, the LR model with FN and WT as selected attributes (RMSE = 11.85 years, correlation coefficient = 0.65) performed decently, while for males, the LR model with all variables showed RMSE of 12.52 years and correlation coefficient of 0.46. This study showcased the potential application of pixel intensity in predicting age.

El presente estudio tuvo como objetivo investigar la utilidad del fémur proximal en la estimación forense de la edad mediante la evaluación de cambios en las densidades óseas a través de radiografías. Utilizando el umbral de Otsu, la densidad ósea se cuantificó contando todos los valores de pixeles blancos dentro de regiones de interés seleccionadas, que incluyen la cabeza femoral (CF), el cuello femoral (CF), el triángulo de Ward (WT) y el trocánter mayor (TM) de 354 fémures izquierdos de ascendencia del norte de Tailandia. También se estimó el ancho de pixeles de la cavidad medular (CM). Además, el estudio evaluó el rendimiento de modelos de regresión lineal (RL) para la estimación de la edad a partir de imágenes radiográficas de fémur proximal. Se observaron correlaciones negativas entre la intensidad de los pixeles CF, CF, WT y TM con la edad de muerte, y las mujeres exhibieron correlaciones más fuertes que los hombres. Además, se encontró una correlación positiva entre la edad y el ancho del CM en muestras de mujeres, mientras que el ancho del CM del hombre no mostró ninguna relación con el aumento de la edad. Los resultados mostraron una ligera diferencia entre el modelo RL aplicado a ambos sexos, que integraba todas las variables, y la configuración alternativa que sólo utilizaba atributos relevantes. Ambos modelos mostraron un rendimiento similar, con un rango estrecho de valores del error cuadrático medio (RMSE), que oscilaba entre 12,67 y 12,71 años, y un rango de coeficiente de correlación de 0,51 a 0,52. Para las mujeres, el modelo RL con CF y WT como atributos seleccionados (RMSE = 11,85 años, coeficiente de correlación = 0,65) tuvo un desempeño satisfactorio, mientras que para los hombres, el modelo RL con todas las variables mostró un RMSE de 12,52 años y un coeficiente de correlación de 0,46. Este estudio mostró la posible aplicación de la intensidad de los pixeles en la predicción de la edad.

Significant overlap of inflammatory and degenerative features on imaging among patients with degenerative disc disease, diffuse idiopathic skeletal hyperostosis and axial spondyloarthritis: a real-life cohort study.

BACKGROUND: Differentiating between degenerative disc disease (DDD), diffuse idiopathic skeletal hyperostosis (DISH), and axial spondyloarthritis (axSpA) represents a diagnostic challenge in patients with low back pain (LBP). We aimed to evaluate the distribution of inflammatory and degenerative imaging features in a real-life cohort of LBP patients referred to a tertiary university rheumatology center. METHODS: In a retrospective cross-sectional analysis of patients referred for LBP, demographics, symptom information, and available imaging were collected. SpA-like changes were considered in the spine in the presence of one of the following lesions typically related to SpA: erosions, sclerosis, squaring, and syndesmophytes on conventional radiographs (CR) and bone marrow oedema (BMO), erosions, sclerosis, and fat lesions (FL) on MRI. SIJ CR were graded per New York criteria; on MRIs, SIJs were evaluated by quadrant for BMO, erosions, FL, sclerosis and ankylosis, similar to the approach used by the Berlin SIJ MRI scoring system. The final diagnosis made by the rheumatologist was the gold standard. Data were presented descriptively, by patient and by quadrant, and compared among the three diagnosis groups. RESULTS: Among 136 referred patients, 71 had DDD, 38 DISH, and 27 axSpA; median age 62 years [IQR55-73], 63% males. On CR, SpA-like changes were significantly higher in axSpA in the lumbar (50%, vs. DDD 23%, DISH 22%), in DISH in the thoracic (28%, vs. DDD 8%, axSpA 12%), and in DDD in the cervical spine (67% vs. DISH 0%, axSpA 33%). On MRI, BMO was significantly higher in DISH in the thoracic (37%, vs. DDD 22%, axSpA 5%) and equally distributed in the lumbar spine (35-42%). FL were significantly more frequently identified in DISH and axSpA in the thoracic (56% and 52%) and DDD and axSpA in the lumbar spine (65% and 74%, respectively). Degenerative changes were frequent in the three groups. Sacroiliitis (NY criteria) was identified in 49% (axSpA 76%, DDD 48%, DISH 29%). CONCLUSION: A significant overlap was found among DDD, DISH, and axSpA for inflammatory and degenerative imaging features. Particularly, SpA-like spine CR features were found in one-fourth of patients with DISH, and MRI BMO was found in one-third of those patients.