HAL Portal Chu Clermont-Ferrand
Not a member yet
7589 research outputs found
Sort by
Evaluation of a machine-learning model based on laboratory parameters for the prediction of acute leukaemia subtypes: a multicentre model development and validation study in France
No full textInternational audienceBackground: Acute leukaemias are life-threatening haematological cancers characterised by the infiltration of transformed immature haematopoietic cells in the blood and bone marrow. Prompt and accurate diagnosis of the three main acute leukaemia subtypes (ie acute lymphocytic leukaemia [ALL], acute myeloid leukaemia [AML], and acute promyelocytic leukaemia [APL]) is of utmost importance to guide initial treatment and prevent early mortality but requires cytological expertise that is not always available. We aimed to benchmark different machine-learning strategies using a custom variable selection algorithm to propose an extreme gradient boosting model to predict leukaemia subtypes on the basis of routine laboratory parameters.Methods: This multicentre model development and validation study was conducted with data from six independent French university hospital databases. Patients aged 18 years or older diagnosed with AML, APL, or ALL in any one of these six hospital databases between March 1, 2012, and Dec 31, 2021, were recruited. 22 routine parameters were collected at the time of initial disease evaluation; variables with more than 25% of missing values in two datasets were not used for model training, leading to the final inclusion of 19 parameters. The performances of the final model were evaluated on internal testing and external validation sets with area under the receiver operating characteristic curves (AUCs), and clinically relevant cutoffs were chosen to guide clinical decision making. The final tool, Artificial Intelligence Prediction of Acute Leukemia (AI-PAL), was developed from this model.Findings: 1410 patients diagnosed with AML, APL, or ALL were included. Data quality control showed few missing values for each cohort, with the exception of uric acid and lactate dehydrogenase for the cohort from Hôpital Cochin. 679 patients from Hôpital Lyon Sud and Centre Hospitalier Universitaire de Clermont-Ferrand were split into the training (n=477) and internal testing (n=202) sets. 731 patients from the four other cohorts were used for external validation. Overall AUCs across all validation cohorts were 0·97 (95% CI 0·95-0·99) for APL, 0·90 (0·83-0·97) for ALL, and 0·89 (0·82-0·95) for AML. Cutoffs were then established on the overall cohort of 1410 patients to guide clinical decisions. Confident cutoffs showed two (0·14%) wrong predictions for ALL, four (0·28%) wrong predictions for APL, and three (0·21%) wrong predictions for AML. Use of the overall cutoff greatly reduced the number of missing predictions; diagnosis was proposed for 1375 (97·5%) of 1410 patients for each category, with only a slight increase in wrong predictions. The final model evaluation across both the internal testing and external validation sets showed accuracy of 99·5% for ALL diagnosis, 98·8% for AML diagnosis, and 99·7% for APL diagnosis in the confident model and accuracy of 87·9% for ALL diagnosis, 86·3% for AML diagnosis, and 96·1% for APL diagnosis in the overall model.Interpretation: AI-PAL allowed for accurate diagnosis of the three main acute leukaemia subtypes. Based on ten simple laboratory parameters, its broad availability could help guide initial therapies in a context where cytological expertise is lacking, such as in low-income countries
Vécu de l’accouchement des femmes selon le type de poussée à l’accouchement : Etude EOLE
No full textInternational audienc
Patient Distress: A Study of the Feelings of Professionals in Palliative Care
No full textInternational audienc
Prevalence of Chronic Pain Among People with Dementia A Nationwide Study Using French Administrative Data
No full textInternational audienc
Chronic pain is a risk factor for incident Alzheimer’s disease a nationwide propensity matched cohort using administrative data
No full textInternational audienc
Impact of childhood leukemia on siblings: their long-term perception of family functioning and its relationship with their psychosocial characteristics using structural equation modeling
No full textInternational audienceAbstract Background Neurofibromatosis type 1 (NF1) is characterized by the highly variable and unpredictable development of benign peripheral nerve sheath tumours: cutaneous (cNFs), subcutaneous (scNFs) and plexiform (pNFs) neurofibromas. Objectives To identify neurofibroma modifier genes, in order to develop a database of patients with NF1. Methods All patients were phenotypically evaluated by a medical practitioner using a standardized questionnaire and the causal NF1 variant identified. We enrolled 1333 patients with NF1 who were genotyped for > 7 million common variants. Results A genome-wide association case-only study identified a significant association with 9q21.33 in the pNF phenotype in the discovery cohort. Twelve, three and four regions suggestive of association at the P ≤ 1 × 10–6 threshold were identified for pNFs, cNFs and scNFs, respectively. Evidence of replication was observed for 4, 2 and 6 loci, including 168 candidate modifier protein-coding genes. Among the candidate modifier genes, some were implicated in the RAS–mitogen-activated protein kinase pathway, cell-cycle control and myelination. Using an original CRISPR/Cas9-based functional assay, we confirmed GAS1 and SPRED2 as pNF and scNF candidate modifiers, as their inactivation specifically affected NF1-mutant Schwann cell growth. Conclusions Our study may shed new light on the pathogenesis of NF1-associated neurofibromas and will, hopefully, contribute to the development of personalized care for patients with this deleterious and life-threatening condition
Assessment of pain prevalence in cancer patients undergoing anticancer treatments and in cancer survivors after completion of anticancer treatments: A French nationwide cross‐sectional study
No full textInternational audiencePain is a common and disabling symptom of cancer and its treatment. This study aimed to provide an update on the prevalence, characteristics, and impact of pain on quality of life (QoL) in cancer patients and survivors in France. Data were collected using self‐assessment questionnaires as part of a nationwide web‐based survey conducted between January and March 2023. Pain was reported by 44.7% of the study population ( n = 1029), including by 49.2% (95% CI [44.8; 53.6]) of cancer patients ( n = 255/518) and 40.1% (95% CI [35.8; 44.5]) of cancer survivors ( n = 205/511). Chronic pain was more prevalent among survivors (99.0%) than patients (87%), but no between‐group differences in the prevalence of neuropathic pain (66.8% vs. 67.5%, respectively) or other pain characteristics (pain intensity, location, etc.) were observed. Pain had a negative impact on QoL in both groups, but the impact on global health status, functioning, symptom severity, and depression was greater among cancer patients. Analgesic use was also more frequent among patients than survivors. Breast cancer, being overweight or obese, and having a poorer global health status were identified as main factors increasing the likelihood of pain. Pain therefore remains a common symptom among cancer patients and survivors in France. Further improvements to management are needed, including strategies to target chronic and neuropathic pain, and the high frequency of pain associated with breast cancer. Multimodal interventions to improve global health status, help individuals maintain a healthy weight, and reduce the impact of cancer pain on QoL could also be evaluated
Frequency of irritable bowel syndrome in spondyloarthritis: a multicentric cross-sectional study and meta-analysis
No full textInternational audienceObjective: To evaluate the prevalence of symptoms and factors associated with irritable bowel syndrome (IBS) in axial spondyloarthritis (ax-SpA). Methods: In a cross-sectional multicentric study, consecutive patients with ax-SpA treated with biologics in five rheumatology departments were asked for IBS Rome IV criteria. Demographic data, lifestyle behaviours and disease characteristics were recorded. Second, a systematic literature review and meta-analysis were performed following Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. Results Of the 500 patients with ax-SpA included, 124 reported IBS symptoms (25%). Female gender, unemployment, higher Bath Ankylosing Spondylitis Disease Activity Index (BASDAI) and worse Bath Ankylosing Spondylitis Functional Index scores, multiple lines of biologics, fibromyalgia, anxiety, depression and lower physical activity were associated with IBS symptoms. In multivariate model, the risk of IBS was associated with anxiety and physical inactivity. From the literature review, the prevalence of IBS in patients with SpA was 15.4% (8.8% to 23.3%). Meta-analysis of the five studies comparing the presence of IBS in patients with SpA (323/7292) and healthy controls (484/35587) showed a significant increase of IBS in patients with SpA (OR=1.59 (1.05 to 2.40)). Conclusion: The prevalence of IBS symptoms was high in the ax-SpA population and should therefore be considered in the presence of gastrointestinal disorders. The presence of IBS symptoms was associated with anxiety and low physical activity in multivariate analysis. Patients with IBS symptoms tended to have more difficult to manage disease characterised by higher activity, worse functional score and multiple lines of treatment in univariate analysis
