SMSébastien Molière

Sébastien Molière

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Sébastien Molière has a total of 251 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Structured MRI reporting for deep pelvic endometriosis: objective report quality and surgeon-perceived clinical utility in a randomized comparative study

    by Louis Burgart, Aline Host, Victor Gabriele, Valentin Ance, Valentin Bernard, Pierre Blonde, Mariem Eleuch, Rodolphe Jochem, Baptiste Labetowiez, Gaspard Ludes, et al.

    Objectives To compare structured and free-text MRI reporting in terms of objective report quality and surgeon ratings of perceived clinical utility in suspected deep pelvic endometriosis. Materials and methods In this single-center randomized reader study, 20 radiology residents were assigned to free-text or structured reporting. Five residents became unavailable before report generation and provided no outcome data. The remaining 15 readers completed the same 22 pelvic MRI examinations, generating 330 reports: 154 free-text and 176 structured reports. Reports were compared with an imaging reference standard established by two pelvic radiologists through independent review and consensus adjudication. The primary endpoint was clinically weighted completeness. Secondary endpoints included unweighted completeness, explicit negative discrepancies relative to the expert imaging reference, positive findings unsupported by the expert imaging reference, and reporting time. Objective outcomes were analyzed using mixed-effects models or Poisson generalized estimating equations. In an exploratory paired assessment, two expert gynecologic surgeons evaluated 44 reports, comprising one free-text and one structured report for each MRI examination. Results Structured reporting was associated with an 8.1%-point increase in completeness compared with free-text reporting (95% CI, 4.1–12.1; p < 0.001) and a 5.6%-point increase in clinically weighted completeness (95% CI, 1.3–10.0; p = 0.011). Reporting time was 4.52 min longer with structured reporting (95% CI, 3.05–5.99; p < 0.001). Structured reporting was also associated with more explicit negative discrepancies relative to the expert imaging reference (rate ratio, 2.94; 95% CI, 1.73–4.98; p < 0.001). These discrepancies were mainly explicit compartment-based negative statements. In the exploratory surgeon assessment, structured reports received higher ratings for perceived completeness, usefulness for clinical decision-making, and adequacy to clinical needs, whereas clarity ratings were similar between formats. Conclusion Structured MRI reporting was associated with higher objective completeness and, in an exploratory assessment, higher surgeon ratings for several dimensions of perceived clinical utility. These findings should be balanced against longer reporting time and more explicit negative discrepancies relative to the expert imaging reference.

  • Modulation du micro-environnement tumoral par la MMP-11

    by Sébastien Molière

    Contexte: Le microenvironnement tumoral joue un rôle crucial dans l’évolution du cancer, notamment dans le cancer du sein. Bien que les analyses transcriptomiques et génomiques offrent une vue d'ensemble, la compréhension des interactions tumeur-stroma reste incomplète. Parmi les molécules d'intérêt, la métalloprotéase matricielle-11 se distingue par son rôle dans la progression du cancer du sein, et son expression stromale liée à un mauvais pronostic. Axes de recherche: a) Rôle pronostique de la MMP-11 dans une cohorte de cancer du sein, b) Rôle de la MMP-11 dans la progression tumorale et le métabolisme à travers un modèle murin, c) Identification et caractérisation des substrats de la MMP-11, notamment les protéines IGFBP. Résultats: Les résultats ont montré que l'expression stromale de MMP-11 a un rôle pronostique dans les stades précoces du cancer du sein. Les expériences ont révélé une action antiadipogénique autocrine de la MMP-11 adipocytaire, mais l’effet protumoral paracrine attendu n’a pas pu être confirmé. Enfin, il a été découvert que la MMP-11 peut cliver la majorité des IGFBPs humaines, suggérant un mécanisme potentiellement nouveau par lequel la MMP- 11 pourrait influencer le cancer, y compris par des actions intracellulaires uniques.

  • Open Access

    Human–AI interaction in a cancer-enriched double-reading breast screening cohort: diagnostic accuracy and second-reader behavior

    by Eloïse Sossavi, Mickaël Tardy, Florie Hurstel, Jean Schwartz, Antoine Wackenthaler, Claire Harter, Julien Uttner, Mélanie Mollion, Marie-Françoise Bretz, Sébastien Molière

    Rationale and objectives: To evaluate the impact of deploying AI as the first reader (R1) in a double-reading breast-screening workflow and to characterize second-reader (R2) behavior—including the effect of disclosing whether R1 was AI or human. Materials and methods: This retrospective study used a cancer-enriched cohort of 220 women (95 cancers), with prevalence-weighted analyses performed to approximate population screening metrics. Five radiologists and one commercially available AI (Breast-SlimView ® , Hera-MI) each served as R1; four radiologists served as R2. For each R2, cases were randomized 1:1 to AI-first versus human-first and, independently, to disclosure versus concealment of R1 identity. R2 could validate, dismiss, or add annotations. The primary endpoint was final decision correctness by breast. We used GEE logistic regression to estimate the overall effect of using AI as the first reader and to isolate second-reader behavior independently of first-reader accuracy. Results: At the prespecified R1 operating point, AI had sensitivity/specificity/accuracy of 85.2%/79.5%/80.8% versus 84.3%/84.5%/85.0% for human R1s; crude final accuracy was lower for AI-first. At 0.6% prevalence, AI-first yielded higher recalls (20.8% vs. 16.8%) with slightly lower PPV (2.7% vs. 3.0%). Conditioning on R1 correctness, R2s were approximately twice more likely to overturn an incorrect AI-initiated opinion than an incorrect human-initiated one (OR ≈ 2.05, p < 0.001). Disclosure that R1 was AI increased R2 corrections (from 13.6% to 19.1%, p = 0.029). Thirteen AI-true-positive cues were dismissed by R2. Conclusions: At this operating point, AI-first reduced crude accuracy due to lower specificity, yet reader-behavior analyses indicate greater scrutiny of AI-initiated opinions. Protocol, threshold, and user-interface choices may raise specificity while preserving beneficial human–AI dynamics.

  • Open Access

    Periportal halo on early postoperative CT predicts biliary complications following liver transplantation

    by Hadrien Lasolle, François Faitot, Sébastien Molière

  • Open Access

    Is There a Therapeutic Benefit of Axillary Surgery in Non-Metastatic Breast Cancer? A SEER Cohort Database Study

    by Jonathan Sabah, Alexis Marouk, Sébastien Molière, Massimo Lodi

    Background. Axillary lymph node biopsy (ALND) has traditionally been considered the gold standard for axillary staging and treatment in clinically node-positive breast cancer patients. However, in patients with nodal disease, the therapeutic benefit of ALND is uncertain. This study, based on a large cohort, aims to evaluate breast cancer-specific survival depending on the extent of axillary surgery in non-metastatic breast cancer using real-world data from the Surveillance, Epidemiology, and End Results (SEER) database. Methods. This retrospective cohort study comprised 825,240 patients diagnosed with breast cancer between 2000 and 2020. Results. ALND was associated with a worse survival outcome in pN0 and pN1 populations (respectively, hazard ratio [HR] 1.16; 95% confidence interval [CI] 1.12–1.2; p < 0.001 and HR 1.38; 95%CI 1.3–1.46; p < 0.001). In pN2 and pN3 populations, there was ~4.3% relative reduction in the hazard of breast cancer-related death for each additional node removed; and higher positive-to-removed lymph node ratio was associated with worse prognosis (HR 3.450; 95%CI 2.99–3.98; p < 0.001). Conclusions. SLNB is associated with significantly better specific survival compared to ALND in negative/low axillary involvement, in higher axillary involvement categories extensive axillary surgery was associated with better prognosis.


  • Une ecchymose (qui) dure

    by Sébastien Molière

  • Open Access

    Un fibroadénome, vraiment ?

    by Sébastien Molière

  • Open Access

    A Systematic Review of the Diagnostic Accuracy of Deep Learning Models for the Automatic Detection, Localization, and Characterization of Clinically Significant Prostate Cancer on Magnetic Resonance Imaging

    by Sébastien Molière, Dimitri Hamzaoui, Guillaume Ploussard, Romain Mathieu, Gaelle Fiard, Michael Baboudjian, Benjamin Granger, Morgan Roupret, Hervé Delingette, Raphaele Renard-Penna