RESUMO
Reinforced cementoplasty with spindles is a recently introduced technique that is mainly used for pathological fractures or for bone metastases at risk of fracture in locations with shear stresses. The technique is less challenging to perform than percutaneous screw insertion and does not require equipment sterilization. No general anesthetic is required. A small trocar is all that is needed, and sutures are often unnecessary. Reinforced cementoplasty can therefore be considered as a technical evolution of cementoplasty with the simple addition of material within the trocar. This technique deserves more awareness so that it can be included in interventional radiologists' range of procedures.
Assuntos
Cimentos Ósseos , Cementoplastia , Humanos , Cimentos Ósseos/uso terapêutico , Resultado do Tratamento , Radiografia Intervencionista , Neoplasias Ósseas/cirurgia , Neoplasias Ósseas/secundário , Neoplasias Ósseas/diagnóstico por imagemRESUMO
PURPOSE: To assess the feasibility and technical accuracy of performing pedicular screw placement combined with vertebroplasty in the radiological setting. METHODS: Patients who underwent combined vertebroplasty and pedicle screw insertion under combined computed tomography and fluoroscopic guidance in 4 interventional radiology centers from 2018 to 2023 were retrospectively assessed. Patient demographics, vertebral lesion type, and procedural data were analyzed. Strict intra-pedicular screw positioning was considered as technical success. Pain score was assessed according to the Visual Analogue Scale before the procedure and in the 1-month follow-up consultation. RESULTS: Fifty-seven patients (38 men and 19 women) with a mean age of 72.8 (SD = 11.4) years underwent a vertebroplasty associated with pedicular screw insertion for the treatment of traumatic fractures (29 patients) and neoplastic disease (28 patients). Screw placement accuracy assessed by post-procedure CT scan was 95.7% (89/93 inserted screws). A total of 93 pedicle screw placements (36 bi-pedicular and 21 unipedicular) in 32 lumbar, 22 thoracic, and 3 cervical levels were analyzed. Mean reported procedure time was 48.8 (SD = 14.7) min and average injected cement volume was 4.4 (SD = 0.9) mL. A mean VAS score decrease of 5 points was observed at 1-month follow-up (7.7, SD = 1.3 versus 2.7, SD = 1.7), p < .001. CONCLUSION: Combining a vertebroplasty and pedicle screw insertion is technically viable in the radiological setting, with a high screw positioning accuracy of 95.7%.
Assuntos
Parafusos Pediculares , Fraturas da Coluna Vertebral , Vertebroplastia , Masculino , Humanos , Feminino , Idoso , Estudos Retrospectivos , Estudos de Viabilidade , Fraturas da Coluna Vertebral/diagnóstico por imagem , Fraturas da Coluna Vertebral/cirurgia , Vértebras Lombares/cirurgia , Vertebroplastia/métodosRESUMO
INTRODUCTION: Recent technological advances with dual-energy quantitative computed tomography (DEQCT) allow to combine two images of different level of energy to obtain simulated mono-energetic images at 60 keV (SIM60KeV-QCT) with improved image contrast in clinical practice. This study includes three topics: (1) compare bone mineral content (BMC), areal and volumetric bone mineral density (aBMD, vBMD) obtained with SIM60KeV-QCT, single-energy QCT (SEQCT), and dual X-ray absorptiometry (DXA); (2) compare ash density and weight with respective vBMD and BMC assessed on SIM60KeV-QCT, SEQCT, and DXA; and (3) compare the influence of reconstruction kernels on the accuracy of vBMD and BMC using ash density and ash weight as the reference values. METHODS: DXA, SEQCT, and DEQCT acquisitions were performed ex vivo on 42 human femurs. Standard kernel (SK) and bone kernel (BK) were applied to each stack of images. Ten diaphyses and 10 femoral necks were cut, scanned, and reconstructed using the techniques described above. Finally, the bone specimens were calcined to obtain the ash weight. RESULTS: QCT analysis (SEQCT, SIM60KeV-QCT) underestimated BMC value compared to DXA. For femoral necks, all QCT analyses provided an unbiased estimate of ash weight but underestimated ash density regardless of the kernel used. For femoral diaphysis, SEQCT BK, SIM60KeV-QCT BK, and SK underestimated ash weight but not ash density. CONCLUSION: BMC and vBMD quantifications with the SIM60KeV-QCT gave similar results as the SEQCT. Further studies are needed to optimize the use of SIM60KeV-QCT in clinical situations. SK should be used given the effect of kernels on QCT assessment.
Assuntos
Densidade Óssea , Fêmur , Absorciometria de Fóton/métodos , Fêmur/diagnóstico por imagem , Colo do Fêmur/diagnóstico por imagem , Humanos , Minerais , Tomografia Computadorizada por Raios X/métodosRESUMO
OBJECTIVES: To assess the diagnostic performances of chest CT for triage of patients in multiple emergency departments during COVID-19 epidemic, in comparison with reverse transcription polymerase chain reaction (RT-PCR) test. METHOD: From March 3 to April 4, 2020, 694 consecutive patients from three emergency departments of a large university hospital, for which a hospitalization was planned whatever the reasons, i.e., COVID- or non-COVID-related, underwent a chest CT and one or several RT-PCR tests. Chest CTs were rated as "Surely COVID+," "Possible COVID+," or "COVID-" by experienced radiologists. Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated using the final RT-PCR test as standard of reference. The delays for CT reports and RT-PCR results were recorded and compared. RESULTS: Among the 694 patients, 287 were positive on the final RT-PCR exam. Concerning the 694 chest CT, 308 were rated as "Surely COVID+", 34 as "Possible COVID+," and 352 as "COVID-." When considering only the "Surely COVID+" CT as positive, accuracy, sensitivity, specificity, PPV, and NPV reached 88.9%, 90.2%, 88%, 84.1%, and 92.7%, respectively, with respect to final RT-PCR test. The mean delay for CT reports was three times shorter than for RT-PCR results (187 ± 148 min versus 573 ± 327 min, p < 0.0001). CONCLUSION: During COVID-19 epidemic phase, chest CT is a rapid and most probably an adequately reliable tool to refer patients requiring hospitalization to the COVID+ or COVID- hospital units, when response times for virological tests are too long. KEY POINTS: ⢠In a large university hospital in Lyon, France, the accuracy, sensitivity, specificity, PPV, and NPV of chest CT for COVID-19 reached 88.9%, 90.2%, 88%, 84.1%, and 92.7%, respectively, using RT-PCR as standard of reference. ⢠The mean delay for CT reports was three times shorter than for RT-PCR results (187 ± 148 min versus 573 ± 327 min, p < 0.0001). ⢠Due to high accuracy of chest CT for COVID-19 and shorter time for CT reports than RT-PCR results, chest CT can be used to orient patients suspected to be positive towards the COVID+ unit to decrease congestion in the emergency departments.
Assuntos
COVID-19/diagnóstico por imagem , Triagem , Idoso , Idoso de 80 Anos ou mais , COVID-19/epidemiologia , Serviço Hospitalar de Emergência , Epidemias , Feminino , França , Hospitais Universitários , Humanos , Masculino , Valor Preditivo dos Testes , SARS-CoV-2 , Fatores de Tempo , Tomografia Computadorizada por Raios XRESUMO
OBJECTIVES: To determine the prevalence of bone marrow oedema (BME) at the sacroiliac joint (SIJ) in early postpartum (EPP), nulliparous (NP) and late postpartum (LPP) women, and to identify factors associated with BME presence at the SIJ. METHODS: Three groups were obtained: NP (never given birth), EPP (given birth within 12 months) and LPP (given birth more than 24 months). The primary outcome was the presence of BME and/or structural lesions (erosions, osteophytes, ankylosis and sclerosis) at the SIJ MRI. RESULTS: BME prevalence was greater among EPP (33%) than NP (14%, p=0.001), but was not different to LPP (21%, p=0.071). The Assessment of SpondyloArthritis international Society (ASAS) MRI criteria for sacroiliitis were positive in 75%, 71% and 80%, respectively, of EPP, NP and LPP women with BME. EPP (38%) had similar prevalence of sclerosis than LPP (28%, p=0.135), but greater than NP (18%, p=0.001). Lastly, EPP (28%) had similar prevalence of osteophytes than LPP (42%) and NP (27%), although there was a difference between LPP and NP (p=0.006). CONCLUSIONS: EPP have higher BME prevalence at the SIJ than NP, EPP tend to have higher BME prevalence compared with LPP and BME presence decreases with time from delivery. Three-quarters of women with BME at the SIJ had a positive ASAS MRI criteria for sacroiliitis, indicating that BME presence as the main criterion for a positive diagnosis can lead to false-positive results. SIJ MRIs should not be interpreted in isolation, since age, time from delivery and other factors may outweigh the pertinence of MRI findings. Trial registration number NCT02956824.
Assuntos
Período Pós-Parto , Sacroileíte/diagnóstico por imagem , Espondilartrite/diagnóstico por imagem , Dor nas Costas/etiologia , Medula Óssea/diagnóstico por imagem , Medula Óssea/patologia , Edema/diagnóstico por imagem , Edema/patologia , Feminino , Humanos , Inflamação/diagnóstico por imagem , Inflamação/etiologia , Inflamação/patologia , Imageamento por Ressonância Magnética , Paridade , Gravidez , Prevalência , Articulação Sacroilíaca/patologia , Sacroileíte/etiologia , Sacroileíte/patologia , Espondilartrite/patologiaRESUMO
INTRODUCTION: Haemophilic ankle arthropathy due to repeated joint bleeds often leads to altered gait in adult patients with haemophilia. AIM: To investigate the association between clinical gait features and blood-induced ankle joint damage scored using MRI findings in patients with haemophilic ankle arthropathy. METHODS: This observational study investigated 48 ankles of 24 patients with severe haemophilia (median age of 33 years). Blood-induced ankle joint damage was scored by an experienced radiologist using the International Prophylaxis Study Group (IPSG-)MRI score which evaluates the presence or absence of effusion, synovial hypertrophy, haemosiderin, surface erosions, subchondral cysts and cartilage degeneration. Using 3D gait analysis, peak ankle joint power generation and absorption (W/kg) were measured for each ankle since these are surrogate measures for joint loading during walking. Associations between MRI findings and these two clinical gait features were calculated using Spearman's ρ correlation with an α-level correction (α = 0.01) for multiple tests. RESULTS: We found large negative associations between ankle joint peak power generation and IPSG-MRI score (ρ = -0.631; P = <.001), IPSG-MRI osteochondral subscore (ρ = -0.701; P = <.001), severity of synovial hypertrophy (ρ = -0.507; P = <.001) and haemosiderin (ρ = -0.400; P = .005). Associations were also found for ankle joint peak power absorption and IPSG-MRI score (ρ = -0.425; P = .003) and IPSG-MRI osteochondral subscore (ρ = -0.556; P = <.001). CONCLUSION: Severe blood-induced ankle joint damage relates to a lowered tolerance towards ankle joint mechanical loading during walking in patients with haemophilia.
Assuntos
Articulação do Tornozelo/diagnóstico por imagem , Marcha/fisiologia , Hemartrose/complicações , Hemofilia A/complicações , Imageamento por Ressonância Magnética/métodos , Adulto , Articulação do Tornozelo/patologia , Feminino , Humanos , MasculinoRESUMO
OBJECTIVE: The aim of this study was to reevaluate dural ectasia criteria in Marfan syndrome patients fulfilling the revised Ghent criteria. METHODS: Lumbar computed tomography scans of 19 Marfan patients and 30 matched control subjects were retrospectively assessed. Dural sac ratio (DSR), nerve root sleeve diameter, pedicle width, and a scalloping or meningocele presence were each assessed by 2 readers blinded from the diagnosis. Mann-Whitney-Wilcoxon tests compared the patient and control groups. Receiver operating characteristic curve analysis and multivariate models determined the optimal cutoff value. RESULTS: A DSR value greater than 0.69 at L5 (DSR-L5) such as L4 scalloping of more than 2.65 mm (scall-L4) and 6 or more vertebrae showing a scalloping of more than 3 mm (6-scall) were found very specific but with limited sensitivity. Multivariate model combining DSR-L5 + scall-L4 showed good positive predictive value, whereas model combining DSR-L5 + 6-scall showed good negative predictive value. CONCLUSIONS: Assessment of DSR and vertebral scalloping allows valuable depiction of dural ectasia in Marfan syndrome patients.
Assuntos
Dura-Máter/diagnóstico por imagem , Síndrome de Marfan/diagnóstico , Tomografia Computadorizada por Raios X/métodos , Adolescente , Adulto , Feminino , Humanos , Processamento de Imagem Assistida por Computador/métodos , Masculino , Pessoa de Meia-Idade , Estudos Retrospectivos , Coluna Vertebral/diagnóstico por imagem , Adulto JovemRESUMO
OBJECTIVES: The aims of this study were to evaluate the agreement between cone-beam computed tomography (CBCT) and multislice computed tomography for the characterization of extremity fractures and to compare image quality, radiation dose, and patient tolerance. METHODS: Thirty-six patients with suspected fracture affecting distal extremities or who required preoperative fracture assessment were enrolled prospectively. Each patient underwent CBCT and multislice computed tomography the same day. Both examinations were evaluated independently twice by 2 trained radiologists using the Müller AO classification for fracture characterization. RESULTS: Cohen κ coefficient for agreement between the imaging techniques was almost perfect for fracture characterization, κ = 0.94 [95% confidence interval, 0.91-0.98]. There was substantial to almost perfect agreement for secondary findings. Cone-beam computed tomography was well tolerated and significantly less irradiant and had better subjective image. CONCLUSIONS: An excellent agreement between both imaging techniques was found. This confirms the ability of CBCT to assess fractures and its potential in the management of patients with distal limb trauma.
Assuntos
Tomografia Computadorizada de Feixe Cônico/métodos , Fraturas Ósseas/diagnóstico por imagem , Tomografia Computadorizada Multidetectores/métodos , Cooperação do Paciente/psicologia , Ossos da Extremidade Inferior/diagnóstico por imagem , Ossos da Extremidade Inferior/lesões , Ossos da Extremidade Superior/diagnóstico por imagem , Ossos da Extremidade Superior/lesões , Tomografia Computadorizada de Feixe Cônico/psicologia , Feminino , Humanos , Masculino , Tomografia Computadorizada Multidetectores/psicologia , Estudos Prospectivos , Doses de Radiação , Intensificação de Imagem Radiográfica/métodos , Radiologistas , Sensibilidade e EspecificidadeRESUMO
OBJECTIVE: To determine whether the diagnostic value of cone-beam computed tomography (CBCT) is equivalent or superior to that of multi-detector computed tomography (MDCT) for the radiological assessment of conductive hearing loss with intact tympanic membrane in adults. Evaluation of inter- and intraobserver variability and measurement of the radiation dosimetry associated with each technique were secondary objectives. STUDY DESIGN: Prospective, single-center study. METHODS: Ten adults were included from April to June 2013. All patients underwent MDCT and CBCT with reconstruction of temporal bones. Two radiologists with ENT experience reviewed the results twice. Diagnostic agreement between MDCT and CBCT and inter- and intraobserver agreement was evaluated with the kappa statistic. Comparisons of dosimetry were evaluated by calculating the ratio of the CT dose index (CTDI) between MDCT and CBCT. RESULTS: Diagnostic agreement between MDCT and CBCT was satisfactory (kappa = 0.69). Inter- and intraobserver agreement was also acceptable, and the average ratio of the CTDI of MDCT and CBCT was 4.01. CONCLUSION: CBCT is a reliable method that uses a low dose of radiation to investigate conductive hearing loss with intact tympanic membrane in adults. Its relevance and potential superiority to MDCT in diagnosing middle ear pathologies such as otosclerosis remain to be demonstrated, but the preliminary data are promising.
Assuntos
Tomografia Computadorizada de Feixe Cônico , Perda Auditiva Condutiva/diagnóstico , Tomografia Computadorizada Multidetectores , Osso Temporal/diagnóstico por imagem , Adulto , Feminino , Humanos , Masculino , Pessoa de Meia-Idade , Estudos Prospectivos , Doses de Radiação , RadiometriaRESUMO
PURPOSE: To assess the T1 ρ and T2 values in the hip cartilage of healthy volunteers and to evaluate the reproducibility of these measurements. MATERIALS AND METHODS: The right hip joint of 30 asymptomatic volunteers was explored with 3T magnetic resonance imaging (MRI). Quantitative 3D T1 ρ- and T2 -maps sequences were repeated twice with a 30-minute delay (immediate reproducibility). The same protocol was repeated 14 days later (short-term reproducibility). Immediate and short-term reproducibility were estimated using coefficients of variation and correlation concordance coefficients (CCC). The precisions of the measurements were estimated by the ratio of the standard deviations. A mixed linear model was used to analyze the effect of patient's characteristics on T1 ρ and T2 values. RESULTS: Immediate reproducibility was significantly better than short-term reproducibility for T1 ρ (CCC of 0.75 versus 0.55; P = 0.007) and T2 (CCC 0.65 versus 0.32; P < 0.001). The precisions of the measurements were estimated between 5.5% and 9.1%. Median T1 ρ values were 6.0 msec higher in women than in men (P = 0.006), with no significant influence of age, body mass index (BMI), or sports activity. Median T2 values were not significantly different between men and women (0.4 msec lower in women; P = 0.76). There was no significant influence of age, BMI, or sports activity. T1 ρ and T2 values were lower in lateral regions than in medial regions (4.9 msec and 2.5 msec lower respectively; P < 0.0001). CONCLUSION: Immediate reproducibility of T1 ρ and T2 values is better than short-term, with limited effect of 30 minutes decubitus. T1 ρ values are significantly higher in women. LEVEL OF EVIDENCE: 2 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2018;47:1022-1033.
Assuntos
Cartilagem Articular/anatomia & histologia , Cartilagem Articular/fisiologia , Articulação do Quadril/anatomia & histologia , Articulação do Quadril/fisiologia , Imageamento Tridimensional/métodos , Imageamento por Ressonância Magnética/métodos , Adolescente , Adulto , Feminino , Humanos , Masculino , Pessoa de Meia-Idade , Valores de Referência , Reprodutibilidade dos Testes , Fatores Sexuais , Adulto JovemRESUMO
OBJECTIVES: To assess the prevalence of spinal inflammation on MRI in patients with chronic back pain (CBP) of maximally 3 years duration and to evaluate the yield of adding a positive MRI-spine as imaging criterion to the Assessment of SpondyloArthritis international Society (ASAS) classification criteria for axial spondyloarthritis (axSpA). METHODS: Baseline imaging of the sacroiliac joints (X-SI), MRI of the sacroiliac joints (MRI-SI) and MRI-spine were scored by ≥2 experienced central readers per modality in the SPondyloArthritis Caught Early (SPACE) and DEvenir des Spondylarthropathies Indifférenciées Récentes (DESIR) cohorts. Inflammation suggestive of axSpA was assessed in the entire spine. A positive MRI-spine was defined by the presence of ≥5 inflammatory lesions. Alternative less strict definitions were also tested. RESULTS: In this study, 541 and 650 patients with CBP from the SPACE and DESIR cohorts were included. Sacroiliitis on X-SI and MRI-SI was found in 40/541 (7%) and 76/541 (14%) patients in SPACE, and in DESIR in 134/650 (21%) and 231/650 (36%) patients, respectively. In SPACE and DESIR, a positive MRI-spine was seen in 4/541 (1%) and 48/650 (7%) patients. Of the patients without sacroiliitis on imaging, 3/447 (1%) (SPACE) and 8/382 (2%) (DESIR) patients had a positive MRI-spine. Adding positive MRI-spine as imaging criterion led to new classification in only one patient in each cohort, as the other patients already fulfilled the clinical arm. Other definitions of a positive MRI-spine yielded similar results. CONCLUSION: In two cohorts of patients with CBP with a maximum symptom duration of 3 years, a positive MRI-spine was rare in patients without sacroiliitis on MRI-SI and X-SI. Addition of MRI-spine as imaging criterion to the ASAS axSpA criteria had a low yield of newly classified patients and is therefore not recommended.
Assuntos
Imageamento por Ressonância Magnética , Articulação Sacroilíaca/diagnóstico por imagem , Sacroileíte/diagnóstico por imagem , Coluna Vertebral/diagnóstico por imagem , Espondiloartropatias/diagnóstico por imagem , Adulto , Dor nas Costas/diagnóstico por imagem , Dor nas Costas/etiologia , Dor Crônica/diagnóstico por imagem , Dor Crônica/etiologia , Estudos de Coortes , Feminino , Humanos , Masculino , Variações Dependentes do Observador , Radiografia , Espondiloartropatias/complicações , Adulto JovemAssuntos
Colesterol , Doenças do Nervo Facial/etiologia , Paralisia Facial/etiologia , Granuloma de Corpo Estranho/complicações , Síndromes de Compressão Nervosa/etiologia , Osso Petroso/diagnóstico por imagem , Adulto , Doenças do Nervo Facial/diagnóstico por imagem , Doenças do Nervo Facial/cirurgia , Feminino , Granuloma de Corpo Estranho/diagnóstico por imagem , Granuloma de Corpo Estranho/cirurgia , Humanos , Imageamento por Ressonância Magnética , Síndromes de Compressão Nervosa/diagnóstico por imagem , Síndromes de Compressão Nervosa/cirurgia , Osso Petroso/cirurgia , Tomografia Computadorizada por Raios XAssuntos
Mucocele/diagnóstico por imagem , Seio Esfenoidal/diagnóstico por imagem , Doença Crônica , Descompressão Cirúrgica , Drenagem , Hemianopsia/etiologia , Hemianopsia/fisiopatologia , Humanos , Imageamento por Ressonância Magnética , Masculino , Mucocele/complicações , Mucocele/fisiopatologia , Mucocele/cirurgia , Doenças do Nervo Óptico/etiologia , Doenças do Nervo Óptico/fisiopatologia , Doenças Orbitárias/etiologia , Doenças Orbitárias/fisiopatologia , Dor/etiologia , Dor/fisiopatologia , Seio Esfenoidal/cirurgia , Resultado do Tratamento , Transtornos da Visão/etiologia , Transtornos da Visão/fisiopatologia , Acuidade Visual , Adulto JovemRESUMO
Elastography assesses the biomechanical and structural properties of tissues by measuring their stiffness. Despite promising results, elastography has not yet earned its place in the daily practice of musculoskeletal radiologists. The purpose of this article is to present and examine the data available to date on ultrasound elastography of the tendons through a review of the literature to provide musculoskeletal radiologists with an overview that may help them better understand and use elastography routinely. The most common techniques in ultrasound elastography are described. Then, the aspects of the physiologic and pathologic tendon are presented and discussed. One must make this technique one's own to better apprehend its contribution to the musculoskeletal imaging field, while bearing in mind that further research will be required before admitting elastography as a reliable and validated tool able to optimize our daily clinical practice.
Assuntos
Técnicas de Imagem por Elasticidade , Tendinopatia/diagnóstico por imagem , Tendões/diagnóstico por imagem , Fenômenos Biomecânicos , Módulo de Elasticidade , HumanosRESUMO
OBJECTIVES: Surgery for congenital malformation of the hand is complex and protocols are not available. Simulation could help optimize results. The objective of the present study was to design, produce and assess a 3D-printed anatomical support, to improve success in rare and complex surgeries of the hand. MATERIAL AND METHODS: We acquired MRI imaging of the right hand of a 30 year-old subject, then analyzed and split the various skin layers for segmentation. Thus we created the prototype of a healthy hand, using 3D multi-material and silicone printing devices, and drew up a printing protocol suitable for all patients. We printed a base comprising bones, muscles and tendons, with a multi-material 3D printer, then used a 3D silicone printer for skin and subcutaneous fatty cell tissues in a glove-like shape. To evaluate the characteristics of the prototype, we performed a series of dissections on the synthetic hand and on a cadaveric hand in the anatomy lab, comparing realism, ease of handling and the final result of the two supports, and evaluated their respective advantages in surgical and training contexts. A grading form was given to each surgeon to establish a global score. RESULTS: This evaluation highlighted the positive and negative features of the model. The model avoided intrinsic problems of cadavers, such as muscle rigidity or tissue fragility and atrophy, and enables the anatomy of a specific patient to be rigorously respected. On the other hand, vascular and nervous networks, with their potential anatomical variants, are lacking. This preliminary phase highlighted the advantages and inconveniences of the prototype, to optimize the design and printing of future models. It is an indispensable prerequisite before performing studies in eligible pediatric patients with congenital hand malformation. CONCLUSION: The validation of 3D-printed anatomical model of a human hand opens a large field of applications in the area of preoperative surgical planning. The postoperative esthetic and functional benefit of such pre-intervention supports in complex surgery needs assessing.
Assuntos
Estudos de Viabilidade , Mãos , Modelos Anatômicos , Impressão Tridimensional , Humanos , Mãos/cirurgia , Mãos/diagnóstico por imagem , Adulto , Imageamento por Ressonância Magnética , CadáverRESUMO
The gold standard for otosclerosis diagnosis, aside from surgery, is high-resolution temporal bone computed tomography (TBCT), but it can be compromised by the small size of the lesions. Many artificial intelligence (AI) algorithms exist, but they are not yet used in daily practice for otosclerosis diagnosis. The aim was to evaluate the diagnostic performance of AI in the detection of otosclerosis. This case-control study included patients with otosclerosis surgically confirmed (2010-2020) and control patients who underwent TBCT and for whom radiological data were available. The AI algorithm interpreted the TBCT to assign a positive or negative diagnosis of otosclerosis. A double-blind reading was then performed by two trained radiologists, and the diagnostic performances were compared according to the best combination of sensitivity and specificity (Youden index). A total of 274 TBCT were included (174 TBCT cases and 100 TBCT controls). For the AI algorithm, the best combination of sensitivity and specificity was 79% and 98%, with an ideal diagnostic probability value estimated by the Youden index at 59%. For radiological analysis, sensitivity was 84% and specificity 98%. The diagnostic performance of the AI algorithm was comparable to that of a trained radiologist, although the sensitivity at the estimated ideal threshold was lower.
RESUMO
INTRODUCTION: Metastases increase the risk of fracture when affecting the femur. Consequently, clinicians need to know if the patient's femur can withstand the stress of daily activities. The current tools used in clinics are not sufficiently precise. A new method, the CT-scan-based finite element analysis, gives good predictive results. However, none of the existing models were tested for reproducibility. This is a critical issue to address in order to apply the technique on a large cohort around the world to help evaluate bone metastatic fracture risk in patients. The aim of this study is then to evaluate 1) the reproducibility 2) the transposition of the reproduced model to another dataset and 3) the global sensitivity of one of the most promising models of the literature (original model). METHODS: The model was reproduced based on the paper describing it and discussion with authors to avoid reproduction errors. The reproducibility was evaluated by comparing the results given in the original model by the original first team (Leuven, Belgium) and the reproduced model made by another team (Lyon, France) on the same dataset of CT-scans of ex vivo femurs. The transposition of the model was evaluated by comparing the results of the reproduced model on two different datasets. The global sensitivity analysis was done by using the Morris method and evaluates the influence of the density calibration coefficient, the segmentation, the orientations and the length of the femur. RESULTS: The original and reproduced models are highly correlated (r2 = 0.95), even though the reproduced model gives systematically higher failure loads. When using the reproduced model on another dataset, predictions are less accurate (r2 with the experimental failure load decreases, errors increase). The global sensitivity analysis showed high influence of the density calibration coefficient (mean variation of failure load of 84 %) and non-negligible influence of the segmentation, orientation and length of the femur (mean variation of failure load between 7 and 10 %). CONCLUSION: This study showed that, although being validated, the reproduced model underperformed when using another dataset. The difference in performance depending on the dataset is commonly the cause of overfitting when creating the model. However, the dataset used in the original paper (Sas et al., 2020a) and the Leuven's dataset gave similar performance, which indicates a lesser probability for the overfitting cause. Also, the model is highly sensitive to density parameters and automation of measurement may minimize the uncertainty on failure load. An uncertainty propagation analysis would give the actual precision of such model and improve our understanding of its behavior and is part of future work.
RESUMO
Bone segmentation is an important step to perform biomechanical failure load simulations on in-vivo CT data of patients with bone metastasis, as it is a mandatory operation to obtain meshes needed for numerical simulations. Segmentation can be a tedious and time consuming task when done manually, and expert segmentations are subject to intra- and inter-operator variability. Deep learning methods are increasingly employed to automatically carry out image segmentation tasks. These networks usually need to be trained on a large image dataset along with the manual segmentations to maximize generalization to new images, but it is not always possible to have access to a multitude of CT-scans with the associated ground truth. It then becomes necessary to use training techniques to make the best use of the limited available data. In this paper, we propose a dedicated pipeline of preprocessing, deep learning based segmentation method and post-processing for in-vivo human femurs and vertebrae segmentation from CT-scans volumes. We experimented with three U-Net architectures and showed that out-of-the-box models enable automatic and high-quality volume segmentation if carefully trained. We compared the failure load simulation results obtained on femurs and vertebrae using either automatic or manual segmentations and studied the sensitivity of the simulations on small variations of the automatic segmentation. The failure loads obtained using automatic segmentations were comparable to those obtained using manual expert segmentations for all the femurs and vertebrae tested, demonstrating the effectiveness of the automated segmentation approach for failure load simulations.