Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
December 8, 2025BloodOpen Access

A large language model-based framework for automated phenotypic characterization in myeloproliferative neoplasms

View Full Paper
Ask AI
Bookmark
Share

Authors

JOJ. K. OberoiAMA. S. MuhammadUAUmair Ayub

Discussion

Loading...

Member takes

Overview

Automated framework improves diagnosis and risk stratification in myeloproliferative neoplasms, indicating a need for further validation.

Key Points

  • Automated data extraction achieved 95% accuracy for pathology reports and notes across three MPN types, enhancing diagnosis.
  • F1 scores for extracted data were 0.95 for PV and ET, and 0.98 for MF, showcasing strong performance in diagnostic characterization.
  • A rule-based algorithm and large language model were utilized for accurate extraction from unstructured clinical records.
  • This framework may enable faster clinical decisions, but prospective validation is needed for broader implementation.

Cite This Study

Oberoi et al. (2025) studied this question.

synapsesocial.com/papers/69362f694fa91c937236df7bhttps://doi.org/10.1182/blood-2025-5605
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1A Proteomics-Driven Machine Learning Tool for Distinguishing ET from pre-PMF2025
  2. 2Multi-endpoint AI morphology model (MEAM) enhances risk prediction for vascular events and disease progression in MPNs2025
  3. 3Diagnostic conformity to who criteria and use of cytoreductive therapy in classical myeloproliferative neoplasms: Experience from a resource-limited country2025
  4. 4Peripheral blood-based single-cell and genomic profiling of hematopoietic stem and progenitor cells enables accurate diagnosis and risk stratification in myeloproliferative neoplasms2025
  5. 5Integrated machine learning-based prognostic models for patients with Myelodysplastic Syndromes/neoplasms (MDS)2025