HAL-HCL
Not a member yet
19000 research outputs found
Sort by
Apprentissage multi-séquences pour la segmentation des lésions de la sclérose en plaques dans l'IRM de la moelle épinière
No full textInternational audienceL'acquisition de plusieurs séquences d’IRM est recommandée pour visualiser les lésions de sclérose en plaques dans la moelle épinière. Cependant, les outils de segmentation automatique développés jusqu’à présent ne permettent pas d’analyser conjointement l’ensemble des séquences acquises pour un sujet donné, limitant la pertinence des analyses. Cette étude est la première à explorer une approche multi-séquence pour la détection de ces lésions. En particulier, nous proposons une méthode pour traiter les problèmes inhérents aux données multi-séquences de la moelle épinière, tels que les champs de vue différents, le décalage inter-acquisitions et l’absence de certaines séquences lors de l'entraînement et l'inférence. En particulier, pour gérer les séquences manquantes, nous examinons une méthode simple consistant à remplacer les features manquantes par la moyenne des features associées aux séquences disponibles. Fig. 1 donne un exemple d’une segmentation. Cette approche permet d'obtenir de meilleurs résultats de segmentation même dans le cas où une seule séquence est disponible à l’inférence, par rapport à un modèle entraîné directement sur cette séquence (cf. Table 1). Des expériences et analyses nous permettent de mieux comprendre ce résultat surprenant. En particulier, nous avons observé que l'encodeur et le décodeur bénéficient tous les deux de la variabilité introduite par le cadre multi-séquence. De plus, en proposant un schéma d'augmentation des features latentes pour reproduire cette variabilité dans le contexte mono-séquence, nous avons observé des améliorations par rapport à un apprentissage classique. Nous avons présenté ce travail à MICCAI en octobre 2024 (https://papers.miccai.org/miccai-2024/paper/3549_paper.pdf)
Longitudinal imaging with multicolour CT provides a surrogate marker of response to treatment with a novel iodine-labelled hydrogel in a murine model of osteoarthritis
No full textInternational audienc
Optimizing the patellofemoral compartment in total knee arthroplasty: Is it time for dynamic assessment?
No full textInternational audienceAbstract Despite improvements in implant design, surgical techniques and assistive technologies for total knee arthroplasty (TKA), anterior knee pain (AKP) remains frequently reported, even by satisfied patients. This persistent problem calls for better understanding and management of the patellofemoral or anterior compartment during surgery, just as the techniques and strategies deployed to optimize the flexion and extension spaces through personalized alignment, bone cuts and ligament balancing. Assistive technologies such as navigation and robotics provide new tools to manage this ‘third space’ through precise pre‐operative planning and dynamic intra‐operative assessment. Such endeavors must start with clear definitions of the ‘third space’, how it should be measured, what constitutes its ‘safe zone’, and how it affects outcomes. There are yet no established methods to evaluate the patellofemoral compartment, and no clear thresholds to define over‐ or under‐stuffing. Static assessment using lateral radiographs provides a limited understanding and depends considerably on flexion angle, while dynamic evaluation at multiple flexion angles or using intra‐operative computer or robotic‐assistance enables a broader perspective and solutions to manage patellar tracking and anterior offset. Future studies should investigate the impact of variations in anterior offset in TKA, define its safe zone, and understand the effects of of thresholds for over‐ or under‐stuffing. Experimental methods such as in‐vivo motion analysis and force sensors could elucidate the influence of anterior offset on flexion and extension biomechanics
Osteotomies for genu varum: should we always correct at the tibia? A multicenter analysis of practices in France
No full textInternational audienc
Does the use of shortened stems reduce early femoral complications in total hip arthroplasty using the direct anterior approach?
No full textInternational audienceAbstract Introduction The choice of femoral stem design during total hip arthroplasty (THA) through the Direct Anterior Approach (DAA) is critical. Shortened stems offer potential benefits such as bone preservation and reduced stress shielding. This study aimed to compare early complications at one year of follow-up between shortened and standard stems in DAA THA. Methods A retrospective monocentric case–control study included patients undergoing DAA THA from 2013 to 2023. Two cohorts were analyzed: 537 THA with standard stems and 346 THA with shortened stems. Three hundred forty-three patients in each group were matched (1:1) based on age, sex, and Body Mass Index (BMI). Two independent observers assessed femoral complications at one year. Femoral stem positioning was measured. Results The mean follow-up was 12 ± 0.5 months. The mean age was 64.1 ± 11.7 years. The mean BMI was 26.4 ± 4.4 kg/m 2 . Shortened stems showed a significantly lower rate of femoral complications (1.4% vs. 5.5%, P = 0.005), particularly for the GT fractures ( P = 0.006). In the shortened group, stem alignment was neutral in 69% of cases, varus in 27%, and valgus in 4%. Conclusion Shortened stems in DAA THA were associated with a lower rate of femoral complications, particularly fewer GT fractures. Although shortened stems were more often positioned in varus, this did not impact short-term complication rates. Trial Registration The Advisory Committee on Research Information Processing in the Field of Health (CCTIRS) approved this study on June 4, 2015 (Study ID 15–430)
Orthoplastic approach for the management of open tibial fractures: Should Orthopedic surgeons be trained to perform pedicled flaps?
No full textInternational audienceIntroductionThe management of open tibial fractures has greatly benefited from the development of the orthoplastic approach. Despite advances in soft tissue reconstruction techniques, the use of simple pedicled flaps performed by orthopedic surgeons remains a relevant option.HypothesisIn the context of orthoplastic reconstruction of tibial fractures, the use of pedicled flaps performed by orthopedic surgeons can optimize soft tissue coverage strategies.Materials and methodsA retrospective study was conducted on 63 patients treated for open tibial fractures requiring soft tissue reconstruction between July 1, 2012, and July 1, 2021. Two groups were compared: Group 1 included patients operated on before July 1, 2018, when flaps were performed exclusively by plastic surgeons; Group 2 included patients operated on after July 1, 2018, when flaps were performed by either plastic or orthopedic surgeons. Demographic and injury-related data, type of flaps used, and the 1-month flap failure rate were analyzed. Time to coverage, time to soft tissue healing, and time to secondary internal fixation were recorded and compared between groups.ResultsIn Group 1, 28 out of 45 patients (62%) underwent soft tissue coverage using flaps, of which 30 out of 31 flaps (97%) were performed by plastic surgeons. The free-to-pedicled flap ratio was 16:15. In Group 2, all 18 patients (100%) underwent flap coverage, of which 18 out of 20 flaps (90%) were performed by orthopedic surgeons. The free-to-pedicled flap ratio in this group was 2:18. The median time to coverage was significantly reduced in Group 2, from 28 days to 7 days (p < 0.05). Healing times were shortened. Flap failure rates were not significantly different between groups.ConclusionWhen the injury pattern allows, the use of pedicled flaps by orthopedic surgeons appears to reduce the time to coverage in open tibial fractures, thereby optimizing the orthoplastic reconstruction strategy. Training orthopedic surgeons in these simple, reliable, and reproducible reconstruction techniques is therefore recommended.Level of evidenceIV
Exploring the link between personality dimensions and non-motor fluctuations in Parkinson's disease
No full textInternational audienceBackground: Parkinson's disease (PD) patients on dopaminergic drugs may experience non-motor fluctuations (NMFs) which are often heterogeneous and respond variably to treatments.Objective: We evaluated if personality was associated to NMFs and could modulate the NMFs responsiveness to dopaminergic medication and deep brain stimulation of the sub-thalamic nucleus (STN-DBS).Methods: From the PREDISTIM cohort, personality dimensions of 235 PD patients were assessed by the Temperament and Character Inventory (TCI) before STN-DBS (V0). NMFs were evaluated using the NMFs Severity Scale at V0 and one year after STN-DBS (V1). Linear regression models were performed between TCI dimensions and NMFs at V0; and logistic regression models were done between TCI dimensions and 1) groups of dopa-sensitive patients (responders to ON medication at V0) versus non-dopa-sensitive ones, and 2) responders versus non-responders to STN-DBS at V1. Odds ratios (OR) were also calculated.Results: Significant associations were found between two TCI personality dimensions (“Harm Avoidance” and “Self-Directedness”) and severity of NMFs in OFF medication at V0: PD patients with higher Harm Avoidance and lower Self-Directedness scores having more NMFs. TCI personality dimensions were not associated with the dopa-sensitivity while Novelty Seeking was significantly associated with the STN-DBS-responder group for the psychiatric (OR = 1.09 [1.02–1.17]) and for the dysautonomic NMFs (OR = 1.11 [1.04–1.18]).Conclusions: Certain personality dimensions (Harm Avoidance and Self-Directedness) are associated with NMFs severity at baseline, and PD patients with high Novelty Seeking seem to be better candidates for NMFs improvement after STN-DBS.Plain Language SummaryNon-motor fluctuations are highly invalidating for patients with Parkinson's disease and their response to treatments is variable. This new research examined personality dimensions impact on non-motor fluctuations and their responsiveness to dopaminergic medication and to deep brain stimulation. Two hundred and thirty-five patients with Parkinson's disease participated in the study and filled out several questionnaires to score their personality dimensions and their non-motor fluctuations severity before and after one year of deep brain stimulation. Comparisons were done between different groups of patients according to their response to dopaminergic treatments and to deep brain stimulation. As results, patients with personality traits of high “Harm Avoidance” (more anxious individuals) and low “Self-Directedness” (less goal-oriented individuals) had globally more non-motor fluctuations, and the personality dimension of “Novelty Seeking” (impulsive individuals) was related to a better non-motor fluctuations responsiveness to deep brain stimulation. Hence, non-motor fluctuations are associated with some specific personality dimensions on top of early onset of Parkinson's disease, long disease-duration, high doses of levodopa and female gender. Moreover, Parkinson's disease patients with impulsive-like personality seem to be better candidates for non-motor fluctuations improvement after deep brain stimulation
Metrics for evaluation of automatic epileptogenic zone localization in intracranial electrophysiology
No full textInternational audiencePrecise localization of the epileptogenic zone is critical for successful epilepsy surgery. However, imbalanced datasets in terms of epileptic vs. normal electrode contacts and a lack of standardized evaluation guidelines hinder the consistent evaluation of automatic machine learning localization models. Methods: This study addresses these challenges by analyzing class imbalance in clinical datasets and evaluating common assessment metrics. Data from 139 drug-resistant epilepsy patients across two Institutions were analyzed. Metric behaviors were examined using clinical and simulated data. Results: Complementary use of Area Under the Receiver Operating Characteristic (AUROC) and Area Under the Precision-Recall Curve (AUPRC) provides an optimal evaluation approach. This must be paired with an analysis of class imbalance and its impact due to significant variations found in clinical datasets. Conclusions: The proposed framework offers a comprehensive and reliable method for evaluating machine learning models in epileptogenic zone localization, improving their precision and clinical relevance. Significance: Adopting this framework will improve the comparability and multicenter testing of machine learning models in epileptogenic zone localization, enhancing their reliability and ultimately leading to better surgical outcomes for epilepsy patients.</div
Uterine volume assay after gonadotoxic therapies in childhood, adolescence, and young adulthood: A systematic review and Bayesian network meta‐analysis
No full textInternational audienceIntroduction: Uterine damage after pelvic radiotherapy or total-body irradiation is well described, with decreased uterine volume and high obstetrical morbidity. Some recent studies have reported a smaller uterus in child, adolescent, and young adult cancer survivors treated with chemotherapy only. This systematic review investigated the long-term effects of gonadotoxic therapy on uterine volume during childhood, adolescence, and young adulthood.Material and Methods: Data sources were Medline, Embase, and the Cochrane Library databases from 1990 to April 2023 searched using the following search terms: cancer survivors, bone marrow transplantation, chemotherapy, radiotherapy, and uterine volume. Study selection and synthesis: Only comparative studies reporting uterine volume in adult women who had received chemotherapy and/or radiotherapy during childhood, adolescence, or young adulthood (<25 years) were included. Two independent reviewers performed study selection, bias assessment using the ROBINS-I tool, and data extraction. The main outcome was uterine volume (mL). A Bayesian network meta-analysis with meta-regression for parity and serious risk of bias was performed using a random-effects model.Results: After reviewing 2847 abstracts, four studies were selected for the meta-analysis. Uterine volume data were available for 225 women after chemotherapy, 153 women after chemoradiotherapy, and 257 control women without cancer. Uterine volume was significantly lower in the chemoradiotherapy group than in the control group (−29.2 mL [−49.1, −12.5]). Uterine volume was significantly decreased in the chemoradiotherapy group compared to the chemotherapy group (−20.9 mL [−39.1, −0.3]). The difference in the mean uterine volume between the control and chemotherapy groups was 8.2 mL [−11.8, 34.2] and was not significant.Conclusions: Our meta-analysis confirms the well-known data on chemoradiotherapy-induced uterine damage. Although some studies have suggested the potential impact of high doses of chemotherapy on uterine volume, this meta-analysis did not find any significant decrease in uterine volume after chemotherapy. This result could help counsel age-reproductive women and physicians who perform assisted reproductive technologies in long-term CAYA Cancer survivors
