English
 
Help Privacy Policy Disclaimer
  Advanced SearchBrowse

Item

ITEM ACTIONSEXPORT
 
 
DownloadE-Mail
 Local TagsRelease HistoryDetailsSummary
  GPT4MR: Exploring GPT-4 as an MR sequence and reconstruction programming assistant

Zaiss, M., Rajput, J., Dang, H., Golkov, V., Cremers, D., Knoll, F., et al. (2024). GPT4MR: Exploring GPT-4 as an MR sequence and reconstruction programming assistant. In A. Maier, T. Deserno, H. Handels, K. Maier-Hein, C. Palm, & T. Tolxdorff (Eds.), Bildverarbeitung für die Medizin 2024: Proceedings, German Conference on Medical Image Computing, Erlangen, March 10-12, 2024 (pp. 94-99). Wiesbaden, Germany: Springer Vieweg.

Item is

 

show all hide all

Basic

show hide
Item Permalink: https://hdl.handle.net/21.11116/0000-000F-1E68-C Version Permalink: https://hdl.handle.net/21.11116/0000-000F-1E6D-7
Genre: Conference Paper

Files

show Files

Locators

show
hide
Locator:
https://link.springer.com/content/pdf/10.1007/978-3-658-44037-4_28.pdf (Publisher version)
Description:
-
OA-Status:
Not specified

Creators

show
hide
 Creators:
Zaiss, M1, Author                 
Rajput , JR, Author
Dang, HN, Author
Golkov, V, Author
Cremers, D, Author
Knoll, F, Author
Maier, A, Author
Affiliations:
1Department High-Field Magnetic Resonance, Max Planck Institute for Biological Cybernetics, Max Planck Society, ou_1497796              

Content

show
hide
Free keywords: -
 Abstract: In this study, we explore the potential of generative pre-trained transformer (GPT), as a coding assistant for MRI sequence programming using the Pulseq framework. The programming of MRI sequences is traditionally a complex and time-consuming task, and the Pulseq standard has recently simplified this process. It allows researchers to define and generate complex pulse sequences used in MRI experiments. Leveraging GPT-4’s capabilities in natural language generation, we adapted it for MRI sequence programming, creating a specialized assistant named GPT4MR. Our tests involved generating various MRI sequences, revealing that GPT-4, guided by a tailored prompt, outperformed GPT-3.5, producing fewer errors and demonstrating improved reasoning. Despite limitations in handling complex sequences, GPT4MR corrected its own errors and successfully generated code with step-by-step instructions. The study showcases GPT4MR’s ability to accelerate MRI sequence development, even for novel ideas absent in its training set. While further research and improvement are needed to address complexity limitations, a well-designed prompt enhances performance. The findings propose GPT4MR as a valuable MRI sequence programming assistant, streamlining prototyping and development. The future prospect involves integrating a PyPulseq plugin into lightweight, open-source LLMs, potentially revolutionizing MRI sequence development and prototyping.

Details

show
hide
Language(s):
 Dates: 2024-02
 Publication Status: Issued
 Pages: -
 Publishing info: -
 Table of Contents: -
 Rev. Type: -
 Identifiers: DOI: 10.1007/978-3-658-44037-4_28
 Degree: -

Event

show
hide
Title: Bildverarbeitung für die Medizin 2024 (BVM 2024)
Place of Event: Erlangen, Germany
Start-/End Date: 2024-03-10 - 2024-03-12

Legal Case

show

Project information

show

Source 1

show
hide
Title: Bildverarbeitung für die Medizin 2024: Proceedings, German Conference on Medical Image Computing, Erlangen, March 10-12, 2024
Source Genre: Proceedings
 Creator(s):
Maier, A, Editor
Deserno, TM, Editor
Handels, H, Editor
Maier-Hein, K, Editor
Palm, C, Editor
Tolxdorff, T, Editor
Affiliations:
-
Publ. Info: Wiesbaden, Germany : Springer Vieweg
Pages: - Volume / Issue: - Sequence Number: - Start / End Page: 94 - 99 Identifier: ISBN: 978-3-658-44036-7
DOI: 10.1007/978-3-658-44037-4

Source 2

show
hide
Title: Informatik Aktuell
Source Genre: Series
 Creator(s):
Affiliations:
Publ. Info: -
Pages: - Volume / Issue: - Sequence Number: - Start / End Page: - Identifier: -