Enhancing Medical Researchers' Self-learning With an Intelligent Language Model

A Superiority Randomized Controlled Trial of the Effect of a Novel Intelligent Language Model on the Self-learning Ability of Medical Researchers

Solving medical scientific problems is a crucial driving force behind the advancement of medical disciplines. As the complexity of scientific questions increases, an increasing number of problems require interdisciplinary collaboration to be resolved. However, most medical researchers lack interdisciplinary background knowledge and require substantial time to systematically learn relevant knowledge and skills. Furthermore, the continuous emergence of new knowledge and skills emphasizes the importance of researchers' ability for autonomous learning in the medical field. Therefore, to promote the development of medical disciplines, there is an urgent need for an effective method to enhance researchers' self-directed learning abilities for conducting interdisciplinary research.

The next-generation artificial intelligence language models, exemplified by ChatGPT, hold great potential in assisting researchers to access knowledge and information from various domains. Whether researchers can leverage such AI tools to enhance their self-directed learning abilities for conducting interdisciplinary research remains to be further explored. Additionally, concerns have been raised regarding the potential degradation of cognitive abilities through their use, although valid evidence is currently lacking.

To investigate whether AI tools, represented by ChatGPT, can effectively assist medical researchers in conducting interdisciplinary research and whether their usage may negatively impact researchers' cognitive abilities, a randomized controlled trial is warranted. This trial aims to ascertain the potential benefits and risks associated with utilizing AI tools in the medical research domain.

Study Overview

• Status: Not yet recruiting
• Conditions: Medical Artificial Intelligence; Interdisciplinary Research; Self-Directed Learning
• Intervention / Treatment: Other: Intelligent Language Model; Other: control

• Study Type: Interventional
• Enrollment (Estimated): 60
• Phase: Not Applicable

Contacts and Locations

• Study Contact: Name: wenben chen, Doctor; Phone Number: +8618819472798; Email: weberchan@foxmail.com
• Study Contact Backup: Name: yuanjun shang, Doctor; Phone Number: +8613003970091; Email: shangyj@mail2.sysu.edu.cn

Study Locations

• China
  • Guangdong
    • Guangzhou, Guangdong, China, 510060
      • Zhongshan Ophthalmic Center, Sun Yat-sen University

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Adult
• Accepts Healthy Volunteers: Yes

Description

Inclusion Criteria:

• Junior ophthalmologist with 1-3 years of clinical experience
• 20-28 years old, regardless of gender
• No prior experience in interdisciplinary research involving digital medicine
• Self-reported a minimum of 20 hours of participation in this study during the trial period
• Agree to participate in this study and sign informed consent

Exclusion Criteria:

• Individuals with reading difficulties or reading disabilities
• Reluctance to participate in this study

Study Plan

How is the study designed?

Design Details

• Primary Purpose: Other
• Allocation: Randomized
• Interventional Model: Parallel Assignment
• Masking: Single

Number of Arms

2

Arms and Interventions

Participant Group / Arm	Intervention / Treatment
Experimental: Intelligent Language Model Group; Subjects must use the intelligent language model to complete the retrieval and protocol design execution of an interdisciplinary task, in addition to Google search, literature search and book query.	Other: Intelligent Language Model; Subjects must use the intelligent language model to complete the retrieval and protocol design execution of an interdisciplinary task, in addition to Google search, literature search and book query.
Placebo Comparator: Control Group; Subjects can only use Google search, literature retrieval and book query, and cannot use any AI-driven conversational natural language processing tools to complete the retrieval and protocol design execution of an interdisciplinary task.	Other: control; Subjects can only use Google search, literature retrieval and book query, and cannot use any AI-driven conversational natural language processing tools to complete the retrieval and protocol design execution of an interdisciplinary task.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
completion rate	The number of people who completed the task within the given time / the total number of people in the group	through study completion, an average of 9 months

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Feasibility of the research program	The feasibility of the scheme is scored by a scoring group composed of experts. The feasibility is divided into 1-5 points according to the correctness and integrity of the key steps and details of the test. The higher the score, the higher the feasibility. The 1 point represents more than half of the key steps are missing or wrong, and the 5 point represents all the key steps and the details are appropriate.	through study completion, an average of 9 months

Collaborators and Investigators

• Sponsor: Zhongshan Ophthalmic Center, Sun Yat-sen University

Study record dates

Study Major Dates

• Study Start (Estimated): August 20, 2023
• Primary Completion (Estimated): October 31, 2023
• Study Completion (Estimated): April 30, 2024

Study Registration Dates

• First Submitted: August 21, 2023
• First Submitted That Met QC Criteria: August 28, 2023
• First Posted (Actual): August 29, 2023

Study Record Updates

• Last Update Posted (Actual): August 29, 2023
• Last Update Submitted That Met QC Criteria: August 28, 2023
• Last Verified: July 1, 2023

More Information

Terms related to this study

• Keywords: Self-Directed Learning; Large Language Model; medical artificial intelligence; superiority randomized controlled trial
• Other Study ID Numbers: 2023KYPJ222

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No