Search

Purpose: The main objective of this study was to compare radiologists' performance without and with artificial intelligence (AI) assistance for the detection of bone fractures from trauma emergencies., Materials and Methods: Five hundred consecutive patients (232 women, 268 men) with a mean age of 37 ± 28 (SD) years (age range: 0.25-99 years) were retrospectively included. Three radiologists independently interpreted radiographs without then with AI assistance after a 1-month minimum washout period. The ground truth was determined by consensus reading between musculoskeletal radiologists and AI results. Patient-wise sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) for fracture detection and reading time were compared between unassisted and AI-assisted readings of radiologists. Their performances were also assessed by receiver operating characteristic (ROC) curves., Results: AI improved the patient-wise sensitivity of radiologists for fracture detection by 20% (95% confidence interval [CI]: 14-26), P< 0.001) and their specificity by 0.6% (95% CI: -0.9-1.5; P = 0.47). It increased the PPV by 2.9% (95% CI: 0.4-5.4; P = 0.08) and the NPV by 10% (95% CI: 6.8-13.3; P < 0.001). Thanks to AI, the area under the ROC curve for fracture detection of readers increased respectively by 10.6%, 10.2% and 9.9%. Their mean reading time per patient decreased by respectively 10, 16 and 12 s (P < 0.001)., Conclusions: AI-assisted radiologists work better and faster compared to unassisted radiologists. AI is of great aid to radiologists in daily trauma emergencies, and could reduce the cost of missed fractures., Competing Interests: Declaration of Competing Interests The authors declare no conflict of interests., (Copyright © 2022 Société française de radiologie. Published by Elsevier Masson SAS. All rights reserved.)

Purpose: The purpose of this study was to assess the long-term efficacy of CT-guided microwave ablation (MWA) in the treatment of osteoid osteoma. Secondary objectives were to assess early outcome and side-effects of MWA., Materials and Methods: Twenty-eight consecutive patients (18 men, 10 women) with a median age of 19.5 years (IQR: 16, 25.5) with a total of 28 non-spinal osteoid osteoma treated by CT-guided MWA were retrospectively included. The ablations were performed with a median power and duration of 60 Watt and 1 min 30 s, respectively. Pain referred to osteoid osteoma was assessed at predefined time points using a 0-10 numeric rating scale. At one month, contrast-enhanced follow-up MRI was performed to evaluate the nidus vascularization and the volume of necrosis induced by MWA. Clinical success was defined by the absence of osteoid osteoma-related pain, and technical success was defined by the presence of necrosis of the nidus on the one-month post-MWA MRI., Results: Long term success rate was 93% (26/28) after a follow-up of 55.5 months (IQR: 25.75, 74.5) and technical success rate was 96 % (25/26). One late failure was observed after a patient had been declared cured at one month but the formal proof of a late recurrence of osteoid osteoma could not be brought. Three minor complications were reported including mild reversible superficial radial nerve injury with a skin burn (grade 2) in one patient and moderate skin burn only in two patients., Conclusion: Our results suggest that CT-guided MWA is an effective option for a minimally-invasive treatment of osteoid osteoma with a low rate of complication and no late recurrence., Competing Interests: Declaration of Competing Interest The authors declare no conflict of interests., (Copyright © 2022 Société française de radiologie. Published by Elsevier Masson SAS. All rights reserved.)

Purpose: The purposes of this study were to estimate the prevalence of Rathke cleft cysts (RCC) in a pediatric population on brain MRI, to describe their appearance, and to estimate interobserver agreement in the detection of RCC., Materials and Methods: The brain MRI examinations of 460 children were retrospectively reviewed by two radiologists for the presence of RCC. There were 223 boys and 237 girls with a mean age of 8.8±4.3 (standard deviation [SD]) years (range: 0.1-14.9 years). When present, RCC were analyzed with respect to internal contain and further classified as serous RCC (i.e., high signal on T2-weighted sequences and iso or low signal on T1-weighted sequences) or mucosal RCC (i.e., low signal on T2-weighted sequences and high or iso signal on T1-weighted sequences). Cohen's Kappa coefficient was used to estimate interobserver agreement between the interpretations performed by the two radiologists for the presence of RCC., Results: A total of 14 children had a RCC present on brain MRI, yielding a prevalence of 3.04% (14/460); of these, 3/14 RCCs (21%) were of serous type and 11/14 (79%) were of mucosal type. Interobserver agreement for the presence of RCC was strong (Kappa=0.85; 95% CI: 0.70; 0.99)., Conclusion: The results of our study suggest that the prevalence of RCC in children is greater than previously described., (Copyright © 2019 Société française de radiologie. Published by Elsevier Masson SAS. All rights reserved.)

Purpose: The first aim was to evaluate feasibility and reproducibility of 2-dimensional ultrasound (2D) shear wave elastography (SWE) of human fetal lungs and liver between 24 and 34weeks of gestation. The second aim was to model fetal lung-to-liver elastography ratio (LLE ratio) and to assess its variations according to gestational age and maternal administration of corticosteroids., Material and Methods: 2D-SWE examinations were prospectively performed in fetuses of women with an uncomplicated pregnancy (group 1) and fetuses of women with a threatened preterm labor requiring administration of corticosteroids (group 2). Two 2D-SWE examinations were performed at "day 0" and "day 2" in group 1; before and 24hours after a course of corticosteroid in group 2. Three operators performed 2 cycles of 3 measurements on the lung (regions A1, A2, A3) and the liver (regions IV, V, VI). Repeatability and reproducibility of measurements were calculated. The fetal LLE ratio was modeled from the most reproducible regions., Results: Fifty-five women were enrolled in group 1 and 48 in group 2. For the lung, 8.6% of measurements were considered invalid and 6.9% for the liver. The most reproducible region for the lung was A3 [ICC between 0.70 (95% CI: 0.42-0.85) and 0.78 (95% CI: 0.48-0.90)] and region VI for the liver [ICC between 0.70 (95% CI: 0.40-0.85) and 0.84 (95% CI: 0.60-0.94)]. According to gestational age, a moderate positive linear correlation was found for stiffness values of A3 (R=0.56), V (R=0.46) and VI (R=0.44). LLE ratio values at "day 0" were not different between the two groups but decreased at "day 2" in group 2 (0.2; 95% CI: 0.07-0.34; P<0.001)., Conclusion: Quantitative fetal lung and liver stiffness measurements are possible with 2D-SWE with acceptable reproducibility., (Copyright © 2019 Société française de radiologie. Published by Elsevier Masson SAS. All rights reserved.)

Purpose: To assess the normal values of the antero-posterior (AP) diameter of the posterior interosseous nerve (PIN) of the elbow as it passes beneath the arcade of Frohse and to search for PIN-diameter differences between the upstream, entry point and downstream of the arcade., Material and Methods: Thirty asymptomatic patients prospectively underwent bilateral B-mode ultrasound of the PIN of the elbow. There were 15 men and 15 women with a mean age of 30.2±5.31 (SD) years (range: 26-43 years). Of these, 23 patients were right-handers (23/30; 77%) and 7 were left handers (7/30; 23%). AP diameter of the PIN was measured in long axis at three different locations including the entry point of the arcade, 5-mm upstream and 5-mm downstream the arcade. A comparison between the three measurements was performed using paired t-test., Results: The mean AP diameters of the PIN were 0.83±0.21 (SD) mm (range: 0.43-1.31mm), 0.6±0.17 (SD) mm (range: 0.29-1.16mm) and 0.49±0.13 (SD) mm (range: 0.26-0.86mm) at 5-mm upstream, entry point of the arcade and 5-mm downstream the arcade of Frohse, respectively. Significant drops in PIN diameter were found between upstream and the arcade (-0.23mm; 27%; P<0.001), between the arcade and downstream (-0.11mm; 17%; P<0.001), and between upstream and downstream the arcade (-0.34mm; 40%; P<0.001)., Conclusion: Disparity in AP diameter of the PIN of the elbow in the arcade of Frohse is a normal finding and should not be erroneously interpreted as entrapment when present alone., (Copyright © 2019 Société française de radiologie. Published by Elsevier Masson SAS. All rights reserved.)

Purpose: The purpose of this study was twofold. First, to compare the contrast between spinal multiple myeloma (MM) focal lesions and surrounding bone marrow obtained on T2-weighted Dixon fat-only MR images to that obtained on T1-weighted spin-echo images. Second, to search for correlation between bone marrow fat fraction assessed by T2-weighted Dixon sequence and International Myeloma Working Group myeloma defining events., Materials and Methods: A total of 39 patients with 112 focal MM lesions were included. There were 25 men and 14 women with a mean age of 68.8±9.8 [SD] years (range: 49-88 years). Contrast between focal MM lesions and surrounding bone marrow was calculated on T1-weighted spin-echo and T2-weighted Dixon (including water-only and fat-only) images. Contrast between focal MM lesions and bone marrow was compared using ANOVA and post-hoc Tukey tests. Correlation between bone marrow fat fraction and myeloma defining events was assessed using Spearman's correlation test., Results: MM lesion contrast was greater on T2-weighted Dixon (F (2;93)=35.10) than on T1-weighted images (P<0.0001). Greatest MM lesion contrast was achieved with T2-weighted Dixon fat-only (0.63±0.21 [SD]; range: 0.06-0.91) compared to T2-weighted Dixon water-only (0.45±0.20 [SD]; range: 0.07-0.8) (P=0.0003) and T1-weighted (0.23±0.19 [SD]; range: 0.04-0.87) (P<0.0001) images. There were no significant correlations between myeloma defining events and fat fraction., Conclusion: T2-weighted Dixon fat-only images provide greater contrast between MM lesions and adjacent bone marrow than T1-weighted images. The usefulness of a T1-weighted sequence associated to a T2-weighted Dixon sequence has to be determined., (Copyright © 2019 Société française de radiologie. Published by Elsevier Masson SAS. All rights reserved.)

Purpose: To evaluate the reliability of pelvimetric measurements performed using stereoradiographic imaging (SRI), and to assess maternal and fetal radiation doses compared to low-dose computer tomography (CT) pelvimetry., Materials and Methods: Thirty-five pregnant women (mean age, 29.6±5.5 [SD] years; range: 20-41 years) were prospectively included. All women underwent simultaneous frontal and lateral low-dose SRI and low-dose CT examination of the pelvis. Pelvimetry measurements were obtained from both examinations and radiation doses obtained with the two techniques were compared., Results: SRI-CT correlation (Pearson coefficient correlation [r]; mean bias [mb]) was strong for transverse inlet diameter (r=0.92; mb=-0.09cm), anteroposterior diameter of the pelvic inlet (r=0.92; mb = 0.47cm), maximal transverse diameter (r=0.9; mb=0.21cm), sacrum length (r=0.9; mb=0.09cm). Correlation was good. Correlation was good for the sacrum depth (r=0.75; mb=0.06cm) and Magnin's index (r=0.7; mb=0.5cm). Correlation was moderate for anteroposterior diameter of pelvic outlet (r=0.6; mb=0.52cm). The fetal dose was 13.1 times lower using SRI (87±26μGy) than CT (1140±220μGy, P<0.0001). The effective maternal dose was 3.1 times lower using SRI (97±21μSv) than CT (310±60μSv; P<0.0001)., Conclusion: Pelvic inlet measurements using SRI are reliable. Compared to CT pelvimetry, SRI leads to a significant decrease in fetal and maternal radiation doses. These findings should prompt physicians to use SRI as the first-line approach for pelvimetry., (Copyright © 2018 Soci showét showé françaises de radiologie. Published by Elsevier Masson SAS. All rights reserved.)

Imaging of the sacroiliac joints is the key point in diagnosing and classifying spondyloarthritis. Since the integration of MRI criteria to the Assessment of Spondyloarhtitis Society (ASAS) classification in 2009, the attention was focused on the presence of bone marrow edema to characterize sacroiliitis. However, returning to basics and analysing structural signs is of utmost importance to avoid overdiagnosis of spondyloarthritis., (Copyright © 2016 Editions françaises de radiologie. Published by Elsevier Masson SAS. All rights reserved.)

Introduction: 3-dimensional models (3D) appeared in the medical field 20 years ago. The recent development of consumer 3D printers explains the renewed interest in this technology. We describe the technical and practical modalities of this surgical tool, illustrated by concrete examples., Technical Note: The OsiriX(®) software (version 5.8.5, Geneva, Switzerland) was used for 3D surface reconstruction of the area of interest, the generation and export of ".stl" file. The NetFabb(®) software (Basic version 5.1.1, Lupburg, Germany) provided the preparation of ".stl" file. The 3D-printer was an Up plus 2 Easy 120(®) (PP3DP, Beijing Technology Co. TierTime Ltd., Chine). The printer used fused deposition modeling. The softwar Up!(®) allowed the 3d impression as required., Results: The first case illustrated the value of 3D printing in the upper (frontal sinus and orbital roof). The second case concerned the preconfiguration of the osteosynthesis material for a complex fracture of the midface through the "mirroring" system. The third case showed the conformation of a prereconstruction for segmental mandibulectomy., Discussion: Current 3D-printers are easy to use and represent a promising solution for medical prototyping. The 3D printing will quickly become undeniable because of its advantages: information sharing, simulation, surgical guides, pedagogy., (Copyright © 2015 Elsevier Masson SAS. All rights reserved.)

Purpose: Botulinum toxin (BT) injection is a new treatment for piriformis syndrome (PS). The main purpose of our study was to use MRI to evaluate changes in piriformis muscle morphology after treatment with BT injections., Patients and Methods: Twenty patients presenting with PS who had undergone an MRI were included retrospectively: 12 patients treated with BT injections and eight untreated patients. The following parameters were assessed and compared to a normal contralateral muscle: maximum thickness, volume, and Goutallier's classification grade of fatty infiltration of the piriformis and internal obturator muscles. Pain was assessed through a visual analogue scale (VAS)., Results: The untreated patients had no significant difference in the volume (P=1.0) or thickness of the piriformis muscle (P=0.61). The treated patients showed a significant reduction in the thickness (-4.2mm; P<0.001) and volume (-74.4mm(3); P<0.001) and an increase in the fatty infiltration (P<0.001) of the piriformis muscle treated by BT injection. Muscular atrophy was correlated with the number of BT injections and with the time until an MRI was performed. There was also significant pain relief after BT treatment., Conclusion: BT leads to atrophy and fatty degeneration of the piriformis muscle that can be quantified by MRI and these factors explain why BT injections are effective in the treatment of PS., (Copyright © 2014 Éditions françaises de radiologie. Published by Elsevier Masson SAS. All rights reserved.)

Objectives: To demonstrate the reliability of the EOS imaging system in measuring the internal diameters of the bony pelvis., Materials and Methods: A prospective study comparing the results of the pelvimetry of 18 dry pelvises carried out on the EOS imaging system to measurements taken manually and using the two current gold standard CT methods. Pelvimetric measurements of each pelvic bone were obtained using four methods and compared: direct manual measurements, spiral and sequential CT pelvimetry, and 2D-3D low-dose biplanar X-rays. The various obstetric diameters were measured to the millimetre and compared., Results: There was no significant difference in the different diameters assessed, with the exception of the interspinous diameter. There was a highly significant correlation (P < 0.001) between the values measured manually and by EOS for the Magnin index (Pearson = 0.98), the obstetric conjugate diameter (Pearson = 0.99), and the median transverse diameter (Pearson = 0.87)., Conclusion: The EOS imaging system allows for an ex vivo determination of the obstetrical diameters that is reliable enough to estimate obstetric prognosis, producing comparable measurements to CT. In view of concerns about protection from radiation, this low-dose imaging technique could become, after in vivo prospective validation, the new gold standard for pelvimetry and therefore a good alternative to CT., (Copyright © 2014 Éditions françaises de radiologie. Published by Elsevier Masson SAS. All rights reserved.)

Purpose: Mandibular reconstruction with fibula free flap harvest is currently the reference technique. Various preoperative processes have been developed to optimize this reconstruction. We report our experience with a simple, inexpensive, preoperative technique requiring a 3D printer, a device for maintaining mandibular reduction, a paper-cutting guide., Technical Note: Stereomodels of the mandible were obtained from computed tomography scan data and printed 3D in ABS. It allowed planning mandibular osteotomies, determine the angle between two bone fragments, and preoperatively modeling the osteosynthesis plate. A paper-cutting guide, and a simple device for maintaining mandibular reduction were also built. Two patients were operated on with this technique, with follow-up at 6 and 8 months. Reconstructions were successful with good clinical outcome in terms of mandibular contour and reconstructed segments positions., Discussion: Preoperative planning of reconstruction may be used for mandibular osteotomies, fibular osteotomies, maintaining mandibular reduction, osteosynthesis, or placing implants for dental rehabilitation. The most complex procedures can virtually plan all these steps, but they are expensive and long to implement. Nevertheless, such procedures are quite expansive and require time not always compatible with carcinoma. Using a mandibular stereomodel is fast, easy, and cheap., (Copyright © 2013 Elsevier Masson SAS. All rights reserved.)

Objectives: To describe morphologic abnormalities and signs of patellar lateral femoral friction syndrome (PLFFS) detected by magnetic resonance imaging (MRI)., Materials and Methods: Prospective study of 56 knees (21 patients and 30 controls) studied by 3Tesla MRI. Comparative analysis of clinical data, quantitative and qualitative imaging criteria in a population of patients with anterior knee pain associated with an abnormal MRI signal along the lateral alar folds of the infrapatellar fat pad, a characteristic sign of PLFFS, and a control population with no anterior knee pain or abnormal signal from the infrapatellar fat pad., Results: Patients with PLFFS have anterior and/or lateral knee pain. Their knee has anatomical predispositions for instability, primarily with patella alta (P<0.0001), patellar tilt more than 13.5° (P<0.0001), a patellar nose length less than 9 mm (P=0.0037), a patellar nose ratio less than 0.25 (P<0.0001), a TT-TG distance more than 10 mm (P<0.0001), and a trochlear prominence more than 4 mm (P=0.0056). In 35% of patients, patellar chondropathy is visible, and 48% of patients have patellar or trochlear subchondral abnormalities., Conclusion: Anterior, lateral, and medial knee pain may be related to PLFFS. Anatomical predispositions contributing to instability are found in these patients. There may be associated chondropathies and osteochondropathies., (Copyright © 2012. Published by Elsevier Masson SAS.)