Accuracy of an artificial intelligence algorithm for detecting moderate-to-severe vertebral compression fractures on abdominal and thoracic computed tomography scans.

Objective: To describe the accuracy of HealthVCF, a software product that uses artificial intelligence, in the detection of incidental moderate-to-severe vertebral compression fractures (VCFs) on chest and abdominal computed tomography scans. Materials and Methods: We included a consecutive sample of 899 chest and abdominal computed tomography scans of patients 51-99 years of age. Scans were retrospectively evaluated by the software and by two specialists in musculoskeletal imaging for the presence of VCFs with vertebral body height loss > 25%. We compared the software analysis with that of a general radiologist, using the evaluation of the two specialists as the reference. Results: The software showed a diagnostic accuracy of 89.6% (95% CI: 87.4-91.5%) for moderate-to-severe VCFs, with a sensitivity of 73.8%, a specificity of 92.7%, and a negative predictive value of 94.8%. Among the 145 positive scans detected by the software, the general radiologist failed to report the fractures in 62 (42.8%), and the algorithm detected additional fractures in 38 of those scans. Conclusion: The software has good accuracy for the detection of moderate-to-severe VCFs, with high specificity, and can increase the opportunistic detection rate of VCFs by radiologists who do not specialize in musculoskeletal imaging.

Objetivo: Descrever a acurácia do software HealthVCF na detecção incidental de fraturas compressivas de corpos vertebrais moderadas a graves em exames de tomografia computadorizada do tórax e abdome. Materiais e Métodos: Foram incluídos 899 exames consecutivos de pacientes com idades entre 51 e 99 anos. As imagens foram retrospectivamente avaliadas pelo software e por dois radiologistas especializados em musculoesquelético que investigaram fraturas compressivas de corpos vertebrais com perda da altura somática > 25%. A análise comparativa foi realizada entre o software e um radiologista geral, usando a avaliação do especialista como referência. Resultados: O software apresentou uma acurácia de 89,6% (IC 95%: 87,4­91,5%) para fraturas compressivas moderadas a graves, com sensibilidade de 73,8%, especificidade de 92,7% e valor preditivo negativo de 94,8%. Entre as 145 tomografias positivas detectadas pelo software, o radiologista geral deixou de relatar as fraturas em 62 (42,8%) e o algoritmo detectou fraturas adicionais em 38 dessas tomografias. Conclusão: O software possui boa acurácia na detecção de fraturas compressivas moderadas a graves, com alta especificidade, podendo aumentar a taxa de detecção oportunística dessas fraturas por radiologistas não especializados em musculoesquelético.

Accuracy of an artificial intelligence algorithm for detecting moderate-to-severe vertebral compression fractures on abdominal and thoracic computed tomography scans / Acurácia de um algoritmo de inteligência artificial na detecção de fraturas compressivas moderadas a graves na tomografia computadorizada abdominal e torácica

Abstract Objective: To describe the accuracy of HealthVCF, a software product that uses artificial intelligence, in the detection of incidental moderate-to-severe vertebral compression fractures (VCFs) on chest and abdominal computed tomography scans. Materials and Methods: We included a consecutive sample of 899 chest and abdominal computed tomography scans of patients 51-99 years of age. Scans were retrospectively evaluated by the software and by two specialists in musculoskeletal imaging for the presence of VCFs with vertebral body height loss > 25%. We compared the software analysis with that of a general radiologist, using the evaluation of the two specialists as the reference. Results: The software showed a diagnostic accuracy of 89.6% (95% CI: 87.4-91.5%) for moderate-to-severe VCFs, with a sensitivity of 73.8%, a specificity of 92.7%, and a negative predictive value of 94.8%. Among the 145 positive scans detected by the software, the general radiologist failed to report the fractures in 62 (42.8%), and the algorithm detected additional fractures in 38 of those scans. Conclusion: The software has good accuracy for the detection of moderate-to-severe VCFs, with high specificity, and can increase the opportunistic detection rate of VCFs by radiologists who do not specialize in musculoskeletal imaging.

Resumo Objetivo: Descrever a acurácia do software HealthVCF na detecção incidental de fraturas compressivas de corpos vertebrais moderadas a graves em exames de tomografia computadorizada do tórax e abdome. Materiais e Métodos: Foram incluídos 899 exames consecutivos de pacientes com idades entre 51 e 99 anos. As imagens foram retrospectivamente avaliadas pelo software e por dois radiologistas especializados em musculoesquelético que investigaram fraturas compressivas de corpos vertebrais com perda da altura somática > 25%. A análise comparativa foi realizada entre o software e um radiologista geral, usando a avaliação do especialista como referência. Resultados: O software apresentou uma acurácia de 89,6% (IC 95%: 87,4-91,5%) para fraturas compressivas moderadas a graves, com sensibilidade de 73,8%, especificidade de 92,7% e valor preditivo negativo de 94,8%. Entre as 145 tomografias positivas detectadas pelo software, o radiologista geral deixou de relatar as fraturas em 62 (42,8%) e o algoritmo detectou fraturas adicionais em 38 dessas tomografias. Conclusão: O software possui boa acurácia na detecção de fraturas compressivas moderadas a graves, com alta especificidade, podendo aumentar a taxa de detecção oportunística dessas fraturas por radiologistas não especializados em musculoesquelético.

Unsuspected cerebrospinal fluid leak following a minor sacral fracture: A case report.

We describe a unique case of a 38-year-old female who presented at the emergency room due to a minor sacral fracture and local pain complicated by an initially undiagnosed cerebrospinal fluid (CSF) leak. This case emphasizes the importance of radiologists being attentive to subtle imaging findings of CSF leaks on computed tomography (CT) examinations, even in cases of low-energy traumas, as well as the complementary role of a dedicated magnetic resonance (MR) imaging protocol for such diagnosis.

Iliotibial band reconstruction with allograft fasciae latae tissue: Imaging aspects of a novel surgical technique.

We describe a unique case of 43-year-old male who presented with a persistent lateral knee pain caused by impingement between a femoral surgical screw and the iliotibial band, which was treated with surgical resection of the screw debris. The patient had reincidence of the symptoms and a magnetic resonance showed a wide and unrepairable tear of the iliotibial band, which was treated with interposition of a folded fasciae latae allograf. After the procedure, the patient had excellent clinical results and imaging evaluation showed progressive allograft integration. This case highlights the imaging findings and surgical aspects of an iliotibial band reconstruction, a novel surgical procedure that could be considered in patients with an unrepairable iliotibial band injury.

Magnetic resonance imaging findings in patients with polymyalgia rheumatica.

Objective: To describe the prevalence of magnetic resonance imaging (MRI) findings in patients with the clinical diagnosis of polymyalgia rheumatica (PMR). Materials and Methods: Sixteen consecutive patients with untreated PMR, meeting the American College of Rheumatology criteria, underwent MRI examinations of the shoulder(s), hip(s), or both, depending on clinical complaints. Six patients also underwent MRI of the spine. Results: We evaluated 24 shoulders, among which we identified subacromial-subdeltoid bursitis in 21 (87.5%), glenohumeral joint effusion in 17 (70.8%), and fluid distention of the long head of the biceps tendon sheath in 15 (62.5%). Peritendinitis and capsular edema were observed in 21 (87.5%) and 17 (70.8%) shoulders, respectively. We also evaluated 17 hips, identifying hip joint effusion in 12 (70.6%), trochanteric bursitis in 11 (64.7%), peritendinitis in 17 (100%), and capsular edema in 14 (82.4%). All six of the patients who underwent MRI of the spine were found to have interspinous bursitis. Conclusion: Subacromial-subdeltoid bursitis, glenohumeral joint effusion, and hip joint effusion are common findings in patients with PMR. In addition, such patients appear to be highly susceptible to peritendinitis and capsular edema. There is a need for case-control studies to validate our data and to determine the real impact that these findings have on the diagnosis of PMR by MRI.

Objetivo: Descrever os achados de ressonância magnética (RM) mais prevalentes em pacientes com diagnóstico clínico de polimialgia reumática (PMR). Materiais e Métodos: Dezesseis pacientes com PMR não tratada, classificados pelos critérios do American College of Rheumatology, foram submetidos a RM do ombro e/ou quadril, segundo suas queixas clínicas. Seis pacientes também foram submetidos a RM da coluna. Resultados: Foram avaliados 24 ombros, identificando-se bursite subacromial-subdeltoide em 21 (87,5%), sinovite glenoumeral em 17 (70,8%) e distensão líquida da bainha do tendão da cabeça longa do bíceps em 15 (62,5%). Peritendinite e edema capsular foram observados em 21 (87,5%) e 17 (70,8%) ombros, respectivamente. Dezessete quadris foram analisados, identificando-se sinovite em 12 (70,6%), bursite trocantérica em 11 (64,7%), peritendinite em 17 (100%) e edema capsular em 14 (82,4%). Os seis pacientes que realizaram RM da coluna apresentavam bursite interespinhosa. Conclusão: Bursite subacromial-subdeltoide, sinovite glenoumeral e do quadril são achados de imagem prevalentes em pacientes com PMR. Além disso, achados como peritendinite e edema capsular tiveram alta prevalência nesses pacientes. Estudos de casocontrole devem ser realizados para validar esses dados e estabelecer o real impacto desses achados no diagnóstico de PMR.

Magnetic resonance imaging findings in patients with polymyalgia rheumatica / Achados de ressonância magnética em pacientes com polimialgia reumática

Abstract Objective: To describe the prevalence of magnetic resonance imaging (MRI) findings in patients with the clinical diagnosis of polymyalgia rheumatica (PMR). Materials and Methods: Sixteen consecutive patients with untreated PMR, meeting the American College of Rheumatology criteria, underwent MRI examinations of the shoulder(s), hip(s), or both, depending on clinical complaints. Six patients also underwent MRI of the spine. Results: We evaluated 24 shoulders, among which we identified subacromial-subdeltoid bursitis in 21 (87.5%), glenohumeral joint effusion in 17 (70.8%), and fluid distention of the long head of the biceps tendon sheath in 15 (62.5%). Peritendinitis and capsular edema were observed in 21 (87.5%) and 17 (70.8%) shoulders, respectively. We also evaluated 17 hips, identifying hip joint effusion in 12 (70.6%), trochanteric bursitis in 11 (64.7%), peritendinitis in 17 (100%), and capsular edema in 14 (82.4%). All six of the patients who underwent MRI of the spine were found to have interspinous bursitis. Conclusion: Subacromial-subdeltoid bursitis, glenohumeral joint effusion, and hip joint effusion are common findings in patients with PMR. In addition, such patients appear to be highly susceptible to peritendinitis and capsular edema. There is a need for case-control studies to validate our data and to determine the real impact that these findings have on the diagnosis of PMR by MRI.

Resumo Objetivo: Descrever os achados de ressonância magnética (RM) mais prevalentes em pacientes com diagnóstico clínico de polimialgia reumática (PMR). Materiais e Métodos: Dezesseis pacientes com PMR não tratada, classificados pelos critérios do American College of Rheumatology, foram submetidos a RM do ombro e/ou quadril, segundo suas queixas clínicas. Seis pacientes também foram submetidos a RM da coluna. Resultados: Foram avaliados 24 ombros, identificando-se bursite subacromial-subdeltoide em 21 (87,5%), sinovite glenoumeral em 17 (70,8%) e distensão líquida da bainha do tendão da cabeça longa do bíceps em 15 (62,5%). Peritendinite e edema capsular foram observados em 21 (87,5%) e 17 (70,8%) ombros, respectivamente. Dezessete quadris foram analisados, identificando-se sinovite em 12 (70,6%), bursite trocantérica em 11 (64,7%), peritendinite em 17 (100%) e edema capsular em 14 (82,4%). Os seis pacientes que realizaram RM da coluna apresentavam bursite interespinhosa. Conclusão: Bursite subacromial-subdeltoide, sinovite glenoumeral e do quadril são achados de imagem prevalentes em pacientes com PMR. Além disso, achados como peritendinite e edema capsular tiveram alta prevalência nesses pacientes. Estudos de caso- controle devem ser realizados para validar esses dados e estabelecer o real impacto desses achados no diagnóstico de PMR.

Acute Ligament Injuries of the Elbow.

Ulnar collateral, radial collateral, lateral ulnar collateral, and annular ligaments can be injured in an acute trauma, such as valgus stress in athletes and elbow dislocation. Recognizing normal anatomy in magnetic resonance imaging and ultrasonography studies is important to identify ligamentous abnormalities in these imaging modalities.