Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
December 8, 2025Blood

Performance of different large language models (LLMs) as decision support tools across various hematologic malignancies

View Full Paper
Ask AI
Bookmark
Share

Authors

JAJeremy S. AbramsonANAjay K. NookaDSDavid M. Schuster

Discussion

Loading...

Member takes

Overview

Analysis reveals LLMs align well with treatment recommendations in hematologic malignancies, suggesting valuable decision support tools.

Key Points

  • Examining the performance of large language models as decision support tools in hematologic malignancies.
  • Analyzed 38 complex malignant hematology cases previously reviewed by MDR panels.
  • Compared 3 large language models: ChatGPT 4.5, Claude Opus 4, and Gemini Ultra.
  • Scored recommendations on categories including completeness, reasoning, and relevance.
  • Competence scores: ChatGPT 4.5 (849), Claude Opus 4 (932), Gemini Ultra (964).
  • LLMs performed best in non-Hodgkin lymphoma cases but struggled with multiple myeloma.
  • All LLMs showed good concordance with MDR recommendations, with minor discrepancies.

Cite This Study

Abramson et al. (2025) studied this question.

synapsesocial.com/papers/69362f6e4fa91c937236e13fhttps://doi.org/10.1182/blood-2025-4359
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. 1Evaluating large language models in real-world hematologic clinical decision-making: Performance, limitations, and clinical implications2025
  2. 2Clinical fidelity of large language models in chronic myeloid leukemia: A multimodel comparative study.2026
  3. 3Benchmarking large language models in breast cancer care: agreement with radiology-led multidisciplinary tumor board decisions2026 · 1 citations
  4. 4Large Language Models in Multidisciplinary Decision-Making for Hepatopancreatobiliary Oncology: Retrospective Comparative Feasibility Study2026
  5. 5Evolution of publicly available large language models for complex decision-making in breast cancer care2024 · 34 citations