MétaCan
Menu
Back to cohort
Record W4414041568 · doi:10.1055/s-0045-1810854

Malignant transformation from IPMN to invasive IPMN and PDAC is characterized by distinct shifts in body composition – an AI-based body composition analysis

2025· article· en· W4414041568 on OpenAlexaff
A.V. Chernysheva, Malte Maria Sieren, Judith Schütte, Sam Mogadas, Lennart Berkel, H Graßhoff, Thomas J. Sauer, Kim C. Honselmann, Ulrich F. Wellner, Eva Dazert, N C von Bubnoff, R. Deck, Cassandra Lill, Lynn Wagner, Tobias Keck, René Hosch, Felix Nensa, Timo Gemoll, R Klöckner, Louisa Bolm

Bibliographic record

VenueZeitschrift für Gastroenterologie · 2025
Typearticle
Languageen
FieldMedicine
TopicRadiomics and Machine Learning in Medical Imaging
Canadian institutionsArtificial Intelligence in Medicine (Canada)
Fundersnot available
KeywordsComposition (language)Malignant transformationTransformation (genetics)Cancer

Abstract

fetched live from OpenAlex

Introduction: Intraductal papillary mucinous neoplasms (IPMN) are cystic lesions of the pancreas that may undergo malignant transformation. A comprehensive characterization of body composition has not been performed in patients with non-invasive IPMN as compared to invasive IPMN or pancreatic ductal adenocarcinoma (PDAC) yet. Patients and Methods: Patients with IPMN, invasive IPMN and PDAC were identified from our prospectively maintained institutional database. Analyzing patients’ routine CT scans at the time of diagnosis, body compartments were automatically segmented with a validated AI-based body composition algorithm (BCA), and body composition parameters including adipose tissue compartments, muscle and bone were quantified. Body composition measures were compared between patients with IPMN, invasive IPMN and PDAC. Results: A total of 181 patients were identified, 53 (29.3%) had IPMN, 16 (8.8%) had invasive IPMN, and 112 (61.9%) had PDAC. Median age was 68 (range 39-87) and 51.9% (n=94) of the patients were female. Mean BMI in all patients was 25.3 kg/m2, BMI values were comparable for patients with IPMN (24.6 kg/m2), invasive IPMN (24.7 kg/m2), and PDAC (25.5kg/m2). Comparing patients with IPMN and invasive IPMN, there was a trend for more pronounced visceral obesity (mean 0.31 vs. 0.33, p=0.055). In contrast, subcutaneous obesity was more common among IPMN patients (0.59 vs. 0.58, p=0.045). There was no difference regarding sarcopenia measures between the two groups. Patients with PDAC as opposed to IPMN had considerably higher rates of visceral obesity (0.68 vs. 0.59, p=0.015). In contrast, IPMN patients displayed higher rates of subcutaneous obesity than PDAC (0.58 vs. 0.56, p=0.013). There was a trend for more pronounced sarcopenia in PDAC patients as compared to IPMN (1.63 vs. 1.71, p=0.083). Comparing body composition parameters between PDAC and invasive IPMN, no statistically significant differences were detected for the adipose tissue, muscle and bone compartments. Conclusion: We performed the first analysis systematically by comparing objectively derived body composition measures between IPMN, invasive IPMN, and PDAC. IPMN as compared to malignant lesions was characterized by distinct body composition profiles. Invasive IPMN and even more so PDAC was associated with cancer body composition markers visceral obesity and sarcopenia. Body composition parameters may therefore be an important tool for early detection of malignancy in IPMN. Publication History Article published online: 04 September 2025 © 2025. Thieme. All rights reserved. Georg Thieme Verlag KG Oswald-Hesse-Straße 50, 70469 Stuttgart, Germany

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame machine prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.

metaresearch head score (Codex)0.000
metaresearch head score (Gemma)0.001
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Observational · Consensus signal: Observational
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.001
Threshold uncertainty score0.003

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0000.001
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0010.001
Science and technology studies0.0000.000
Scholarly communication0.0010.000
Open science0.0000.001
Research integrity0.0000.001
Insufficient payload (model declined to judge)0.0010.000

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.009
GPT teacher head0.297
Teacher spread0.288 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designObservational
Domainnot available
GenreEmpirical

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations0
Published2025
Admission routes1
Has abstractno

Explore more

Same venueZeitschrift für GastroenterologieSame topicRadiomics and Machine Learning in Medical ImagingFrench-language works237,207