MedCalc-Bench: Evaluating Large Language Models for Medical Calculations

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Khandekar, Nikhil, Jin, Qiao, Xiong, Guangzhi, Dunn, Soren, Applebaum, Serina S, Anwar, Zain, Sarfo-Gyamfi, Maame, Safranek, Conrad W, Anwar, Abid A, Zhang, Andrew, Gilson, Aidan, Singer, Maxwell B, Dave, Amisha, Taylor, Andrew, Zhang, Aidong, Chen, Qingyu, Lu, Zhiyong
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910507389157376
author Khandekar, Nikhil
Jin, Qiao
Xiong, Guangzhi
Dunn, Soren
Applebaum, Serina S
Anwar, Zain
Sarfo-Gyamfi, Maame
Safranek, Conrad W
Anwar, Abid A
Zhang, Andrew
Gilson, Aidan
Singer, Maxwell B
Dave, Amisha
Taylor, Andrew
Zhang, Aidong
Chen, Qingyu
Lu, Zhiyong
author_facet Khandekar, Nikhil
Jin, Qiao
Xiong, Guangzhi
Dunn, Soren
Applebaum, Serina S
Anwar, Zain
Sarfo-Gyamfi, Maame
Safranek, Conrad W
Anwar, Abid A
Zhang, Andrew
Gilson, Aidan
Singer, Maxwell B
Dave, Amisha
Taylor, Andrew
Zhang, Aidong
Chen, Qingyu
Lu, Zhiyong
contents As opposed to evaluating computation and logic-based reasoning, current benchmarks for evaluating large language models (LLMs) in medicine are primarily focused on question-answering involving domain knowledge and descriptive reasoning. While such qualitative capabilities are vital to medical diagnosis, in real-world scenarios, doctors frequently use clinical calculators that follow quantitative equations and rule-based reasoning paradigms for evidence-based decision support. To this end, we propose MedCalc-Bench, a first-of-its-kind dataset focused on evaluating the medical calculation capability of LLMs. MedCalc-Bench contains an evaluation set of over 1000 manually reviewed instances from 55 different medical calculation tasks. Each instance in MedCalc-Bench consists of a patient note, a question requesting to compute a specific medical value, a ground truth answer, and a step-by-step explanation showing how the answer is obtained. While our evaluation results show the potential of LLMs in this area, none of them are effective enough for clinical settings. Common issues include extracting the incorrect entities, not using the correct equation or rules for a calculation task, or incorrectly performing the arithmetic for the computation. We hope our study highlights the quantitative knowledge and reasoning gaps in LLMs within medical settings, encouraging future improvements of LLMs for various clinical calculation tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2406_12036
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle MedCalc-Bench: Evaluating Large Language Models for Medical Calculations
Khandekar, Nikhil
Jin, Qiao
Xiong, Guangzhi
Dunn, Soren
Applebaum, Serina S
Anwar, Zain
Sarfo-Gyamfi, Maame
Safranek, Conrad W
Anwar, Abid A
Zhang, Andrew
Gilson, Aidan
Singer, Maxwell B
Dave, Amisha
Taylor, Andrew
Zhang, Aidong
Chen, Qingyu
Lu, Zhiyong
Computation and Language
Artificial Intelligence
As opposed to evaluating computation and logic-based reasoning, current benchmarks for evaluating large language models (LLMs) in medicine are primarily focused on question-answering involving domain knowledge and descriptive reasoning. While such qualitative capabilities are vital to medical diagnosis, in real-world scenarios, doctors frequently use clinical calculators that follow quantitative equations and rule-based reasoning paradigms for evidence-based decision support. To this end, we propose MedCalc-Bench, a first-of-its-kind dataset focused on evaluating the medical calculation capability of LLMs. MedCalc-Bench contains an evaluation set of over 1000 manually reviewed instances from 55 different medical calculation tasks. Each instance in MedCalc-Bench consists of a patient note, a question requesting to compute a specific medical value, a ground truth answer, and a step-by-step explanation showing how the answer is obtained. While our evaluation results show the potential of LLMs in this area, none of them are effective enough for clinical settings. Common issues include extracting the incorrect entities, not using the correct equation or rules for a calculation task, or incorrectly performing the arithmetic for the computation. We hope our study highlights the quantitative knowledge and reasoning gaps in LLMs within medical settings, encouraging future improvements of LLMs for various clinical calculation tasks.
title MedCalc-Bench: Evaluating Large Language Models for Medical Calculations
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2406.12036