Nahal Mansouri
Clinician in pulmonology and scientist in Lausanne. Last author of a 2025 deep-learning study that detects bronchiolitis obliterans syndrome on chest CT.
As of . Primary source: ORCID record.
Summary
Nahal Mansouri is listed in the ORCID record as a Cheffe de Clinique in the Division of Pulmonology at the Centre Hospitalier Universitaire Vaudois and as a senior scientist at the Ecole Polytechnique Federale de Lausanne. Nahal Mansouri is included in this category because an opened abstract describes a deep neural network for bronchiolitis obliterans syndrome, a form of CLAD, in a paper with Nahal Mansouri as last author.
Work summary
Nahal Mansouri is the last author of a 2025 study that trained a deep neural network, a layered image-analysis model, to detect bronchiolitis obliterans syndrome (BOS, the airway-centred form of chronic lung allograft dysfunction) on chest CT scans [DOI 10.1038/s43856-025-00732-x]. The model used a co-training method designed for small datasets and was tested on scans from 75 patients after transplant, 26 of whom had BOS, reaching a ROC-AUC (an accuracy score from 0 to 1) of 0.90, and 0.87 in at-risk patients whose lung function was only slightly reduced. Earlier work, also with Nahal Mansouri as last author, applied machine learning to COVID-19 data: one model predicted symptoms and mortality from age, sex and medical history in 23,749 hospital patients [DOI 10.3389/frai.2021.673527], and another predicted mortality on the day of intensive care admission in 797 patients [DOI 10.3389/fdgth.2021.681608]. The COVID-19 studies lie outside lung transplantation and show the methods background rather than CLAD work.
Based on 3 or more opened abstracts. Written only from abstracts that were opened (PubMed or Europe PMC); full texts were not read. Plain-language explanations are added by the editors and are not from the papers.
Role and field
- Role
- Cheffe de Clinique, Division of Pulmonology, Centre Hospitalier Universitaire Vaudois; Senior scientist, Ecole Polytechnique Federale de Lausanne
- Role status
- verified, self-asserted in ORCID employment entries
- Role basis
- ORCID employment entries (current, no end date), read on 2026-10-10.
- Field
- pulmonology; machine learning for chest imaging and clinical prediction
- Field basis
- Unit names in the ORCID employment entries and the source abstracts.
Only what a cited source supports is stated. Entries marked self-asserted come from the person's own ORCID record. The field is taken from the name of the unit, not inferred from the person's name.
Details
- affiliation
- Centre Hospitalier Universitaire Vaudois; Ecole Polytechnique Federale de Lausanne (ORCID, self-asserted)
- ai evidence
- Abstracts of the three source papers, all describing machine-learning or deep-learning models.
Sources and links
- ORCID record (primary)
- OpenAlex author profile
- Deep learning to detect bronchiolitis obliterans syndrome from chest CT (Communications Medicine 2025), DOI 10.1038/s43856-025-00732-x
- Symptom prediction and mortality risk calculation for COVID-19 using machine learning (Frontiers in Artificial Intelligence 2021), DOI 10.3389/frai.2021.673527
- Machine learning to predict mortality for COVID-19 patients on day 0 in the ICU (Frontiers in Digital Health 2022), DOI 10.3389/fdgth.2021.681608
Known gaps in this record
- Only one source paper concerns CLAD; the other two concern COVID-19.
- The role is self-asserted in ORCID and may be out of date.
- Funding and conflicts of interest were not recorded.
- Full texts were not read.
Information resource only. Not medical advice. Not a substitute for the care of the patient's transplant team.