Evaluation of Large Language Models in Parasternal Block Ultrasound Interpretation

October 29, 2026 updated by: Engin Ihsan Turan, Kanuni Sultan Suleyman Training and Research Hospital

Diagnostic Accuracy of Multimodal Large Language Models in Interpreting Ultrasound Images of Parasternal Intercostal Plane Blocks: A Prospective Observational Study

This prospective observational study aims to evaluate the accuracy of multimodal large language models (LLMs) in interpreting ultrasound recordings of superficial and deep parasternal intercostal plane blocks.

Ultrasound recordings obtained during routine regional anesthesia procedures will be independently evaluated by LLMs using standardized prompts. Their ability to identify anatomical structures, assess local anesthetic spread, determine block success, and recognize potential procedural risks will be compared with expert anesthesiologist assessments.

The study aims to investigate the potential role of LLMs in ultrasound-guided regional anesthesia assessment and clinical decision support.

Study Overview

Status

Status

Indicates the current recruitment status or the expanded access status.
Not yet recruiting

Conditions

Conditions

The disease, disorder, syndrome, illness, or injury that is being studied. On ClinicalTrials.gov, conditions may also include other health-related issues, such as lifespan, quality of life, and health risks.

Intervention / Treatment

Intervention / Treatment

A process or action that is the focus of a clinical study. Interventions include drugs, medical devices, procedures, vaccines, and other products that are either investigational or already available. Interventions can also include noninvasive approaches, such as education or modifying diet and exercise.

Detailed Description

Study Design and Participants This prospective observational study will evaluate the diagnostic accuracy of multimodal large language models (LLMs) in interpreting ultrasound recordings of superficial and deep parasternal intercostal plane blocks (S-PIP and D-PIP).

The study will include 32 adult patients aged 18-85 years, classified as American Society of Anesthesiologists (ASA) physical status I-III, undergoing elective surgery with an ultrasound-guided parasternal intercostal plane block.

Blocks will be performed by experienced anesthesiologists as part of routine clinical practice. Ultrasound recordings will include the local anesthetic injection and its spread within the targeted fascial plane. No additional procedures will be performed for research purposes.

Ultrasound Video Acquisition Ultrasound-guided S-PIP and D-PIP procedures will be recorded in video format, including the injection process.

Videos with inadequate visualization of relevant anatomical structures, needle placement, or local anesthetic spread will be excluded.

All recordings will be anonymized before evaluation. Each eligible recording will be assigned a unique identification code.

Large Language Model Evaluation ChatGPT (OpenAI) and Gemini (Google) will be evaluated using their latest available multimodal versions at the time of assessment. Exact model names, versions, access platforms, and evaluation dates will be documented.

The same ultrasound recording will be independently submitted to both models in MP4 format, provided that direct video analysis is supported by the selected model interfaces.

An identical standardized English-language prompt will be used for every recording and both LLMs. Each recording will be evaluated in a separate conversation to minimize contextual carryover.

Models will receive no patient-identifying information, expert interpretations, or feedback regarding the correctness of their responses.

The following eight questions will be presented in the same order:

  1. Identify the anatomical structures visible in this parasternal ultrasound recording.
  2. Based on the observed local anesthetic spread, determine whether the solution was deposited in the correct target fascial plane.
  3. Determine whether the local anesthetic spread is superficial or deep.
  4. If the observed spread is inadequate or occurs outside the target fascial plane, what corrective action would you recommend?
  5. Assess whether the needle tip is positioned within the appropriate fascial plane.
  6. Identify potential procedural risks involving adjacent structures, including the pericardium and internal mammary vessels.
  7. Which dermatomal levels are expected to be covered by this block?
  8. If the observed spread is inadequate or occurs outside the target fascial plane, would you recommend an additional injection? Models will be instructed to base their assessments exclusively on the provided ultrasound recording and to explicitly indicate when the available visual information is insufficient for a reliable conclusion.

Responses will be recorded without additional prompting or correction. Response time and other predefined performance parameters will also be documented.

Expert Assessment and Blinding Two experienced anesthesiology professors will participate in the expert assessment process. The first expert (Expert A) will independently review each ultrasound recording and answer the same eight standardized questions. These assessments will serve as the expert reference. LLM-generated responses and Expert A's responses will subsequently be anonymized and randomly coded to conceal their sources.

The second expert (Expert B), blinded to response origin, will independently evaluate the anonymized responses regarding their accuracy, clinical appropriateness, and agreement with the expert reference.

Responses will be assessed using predefined scoring criteria, including content similarity and an overall quality score ranging from 0 to 10.

Outcome Assessment The primary objective is to compare the diagnostic accuracy of ChatGPT and Gemini in identifying anatomical structures and determining whether the local anesthetic solution was correctly deposited within the intended fascial plane during parasternal intercostal plane blocks.

Secondary assessments will include classification of superficial versus deep local anesthetic spread, needle tip position, procedural risk identification, appropriateness of clinical recommendations, predicted dermatomal coverage, and agreement with expert assessments. Model response time, response quality, and reported confidence will also be evaluated.

Data Protection and Clinical Safety All recordings will be de-identified before submission to the LLM platforms. Model-generated interpretations will be used solely for research purposes and will not influence clinical decision-making or patient management.

The study will be conducted in accordance with the approved research protocol and applicable ethical and data protection requirements.

Study Type

Study Type

Describes the nature of a clinical study. Study types include interventional studies (also called clinical trials), observational studies (including patient registries), and expanded access.
Observational

Enrollment (Estimated)

Enrollment

The number of participants in a clinical study. The "anticipated" enrollment is the target number of participants that the researchers need for the study.
32

Contacts and Locations

This section provides the contact details for those conducting the study, and information on where this study is being conducted.

Study Contact

Study Contact

The name and contact information for the person who can answer enrollment questions for a clinical study. Each location where the study is being conducted may also have a specific contact, who may be better able to answer those questions.

Study Locations

  • Turkey (Türkiye)
      • Istanbul, Turkey (Türkiye), 34303
        • Health Science University İstanbul Kanuni Sultan Süleyman Education and Training Hospital

Participation Criteria

Researchers look for people who fit a certain description, called eligibility criteria. Some examples of these criteria are a person's general health condition or prior treatments.

Eligibility Criteria

Eligibility Criteria

The key requirements that people who want to participate in a clinical study must meet or the characteristics they must have. Eligibility criteria consist of both inclusion criteria (which are required for a person to participate in the study) and exclusion criteria (which prevent a person from participating). Types of eligibility criteria include whether a study accepts healthy volunteers, has age or age group requirements, or is limited by sex.

Ages Eligible for Study

18 years and older (Adult, Older Adult)

Accepts Healthy Volunteers

No

Genders Eligible for Study

All

Sampling Method

Probability Sample

Study Population

32

Description

Inclusion Criteria:

  • Patients aged 18-85 years.
  • American Society of Anesthesiologists (ASA) physical status I-III.
  • Undergoing elective surgery with an ultrasound-guided superficial or deep parasternal intercostal plane block.
  • Complete ultrasound video recording of local anesthetic injection.
  • Written informed consent for study participation.

Exclusion Criteria:

  • Technical failure during block performance.
  • Inadequate ultrasound recording quality preventing clear visualization of the needle tip or local anesthetic spread.
  • Missing demographic data.
  • Withdrawal of informed consent at any stage of the study.

Study Plan

This section provides details of the study plan, including how the study is designed and what the study is measuring.

How is the study designed?

Design Details

  • Observational Models: COHORT
  • Time Perspectives: PROSPECTIVE

Cohorts and Interventions

Group / Cohort

Group / Cohort

A group or subgroup of participants in an observational study that is assessed for biomedical or health outcomes.
Intervention / Treatment

Intervention / Treatment

A process or action that is the focus of a clinical study. Interventions include drugs, medical devices, procedures, vaccines, and other products that are either investigational or already available. Interventions can also include noninvasive approaches, such as education or modifying diet and exercise.
Patients Undergoing Parasternal Intercostal Plane Blocks

Adult patients undergoing ultrasound-guided superficial or deep parasternal intercostal plane blocks as part of routine clinical care. Anonymized ultrasound recordings of local anesthetic injection will be independently evaluated by ChatGPT and Gemini using identical standardized prompts. Model assessments will be compared with expert anesthesiologist evaluations.
Diagnostic test: Multimodal Large Language Model-Based Ultrasound Assessment

Anonymized ultrasound recordings of parasternal intercostal plane blocks will be independently evaluated using the latest available multimodal versions of ChatGPT and Gemini. Both models will receive the same MP4 recordings and identical standardized prompts to assess anatomical structures, correct deposition of local anesthetic solution within the target fascial plane, injection spread, and potential procedural risks. Responses will be compared with expert anesthesiologist assessments. No AI-generated recommendations will influence patient care.

What is the study measuring?

Primary Outcome Measures

Primary Outcome Measures

In a clinical study's protocol, the planned outcome measure that is the most important for evaluating the effect of an intervention/treatment. Most clinical studies have one primary outcome measure, but some have more than one.
Outcome Measure
Measure Description
Time Frame
Correct Local Anesthetic Deposition
Time Frame: During the ultrasound-guided block procedure

The accuracy of ChatGPT and Gemini in determining whether the local anesthetic solution was deposited within the intended fascial plane will be evaluated using ultrasound recordings. Model assessments will be compared with the expert reference assessment. Accuracy will be calculated as the percentage of correctly classified recordings.
During the ultrasound-guided block procedure

Secondary Outcome Measures

Secondary Outcome Measures

In a clinical study's protocol, a planned outcome measure that is not as important as the primary outcome measure for evaluating the effect of an intervention but is still of interest. Most clinical studies have more than one secondary outcome measure.
Outcome Measure
Measure Description
Time Frame
Anatomical Identification Accuracy
Time Frame: During the ultrasound-guided block procedure

The ability of ChatGPT and Gemini to correctly identify anatomical structures visible in parasternal intercostal plane block ultrasound recordings will be evaluated against expert anesthesiologist assessments. Anatomical identification performance will be scored using predefined criteria.
During the ultrasound-guided block procedure
Needle Tip Position
Time Frame: During the ultrasound-guided block procedure

The accuracy of ChatGPT and Gemini in determining whether the needle tip is located within the intended fascial plane will be compared with the expert reference assessment. Results will be reported as the percentage of correctly classified recordings.
During the ultrasound-guided block procedure
Local Anesthetic Spread
Time Frame: During the ultrasound-guided block procedure

The ability of each model to classify local anesthetic spread as superficial or deep and identify the observed spread pattern will be compared with expert assessments. Classification accuracy will be reported as a percentage.
During the ultrasound-guided block procedure
Procedural Risk Identification
Time Frame: During the ultrasound-guided block procedure

The ability of each model to identify potential procedural risks involving adjacent anatomical structures, including the internal mammary vessels and pericardium, will be evaluated against expert assessments. Accuracy will be reported as the percentage of correctly classified recordings.
During the ultrasound-guided block procedure
Clinical Recommendation Quality
Time Frame: During the ultrasound-guided block procedure

The appropriateness of recommendations regarding inadequate local anesthetic spread, corrective actions, and additional injection will be assessed by an independent expert anesthesiologist using a predefined scoring system ranging from 0 to 10, with higher scores indicating greater clinical appropriateness.
During the ultrasound-guided block procedure
Agreement With Expert Assessment
Time Frame: During the ultrasound-guided block procedure

Agreement between ChatGPT, Gemini, and expert anesthesiologist assessments will be evaluated for predefined ultrasound interpretation parameters. Agreement will be quantified using Cohen's kappa for applicable categorical assessments.
During the ultrasound-guided block procedure

Collaborators and Investigators

This is where you will find people and organizations involved with this study.

Sponsor

Sponsor

The organization or person who initiates the study and who has authority and control over the study.
Kanuni Sultan Suleyman Training and Research Hospital

Study record dates

These dates track the progress of study record and summary results submissions to ClinicalTrials.gov. Study records and reported results are reviewed by the National Library of Medicine (NLM) to make sure they meet specific quality control standards before being posted on the public website.

Study Major Dates

Study Start (Estimated)

Study Start

The actual date on which the first participant was enrolled in a clinical study. The "anticipated" study start date is the date that the researchers think will be the study start date.
November 2, 2026

Primary Completion (Estimated)

Primary Completion

The date on which the last participant in a clinical study was examined or received an intervention to collect final data for the primary outcome measure. Whether the clinical study ended according to the protocol or was terminated does not affect this date. For clinical studies with more than one primary outcome measure with different completion dates, this term refers to the date on which data collection is completed for all the primary outcome measures. The "anticipated" primary completion date is the date that the researchers think will be the primary completion date for the study.
April 2, 2027

Study Completion (Estimated)

Study Completion

The date on which the last participant in a clinical study was examined or received an intervention/treatment to collect final data for the primary outcome measures, secondary outcome measures, and adverse events (that is, the last participant's last visit). The "anticipated" study completion date is the date that the researchers think will be the study completion date.
April 3, 2027

Study Registration Dates

First Submitted

First Submitted

The date on which the study sponsor or investigator first submitted a study record to ClinicalTrials.gov. There is typically a delay of a few days between the first submitted date and the record's availability on ClinicalTrials.gov (the first posted date).
October 29, 2026

First Submitted That Met QC Criteria

First Submitted That Met QC Criteria

The date on which the study sponsor or investigator first submits a study record that is consistent with National Library of Medicine (NLM) quality control (QC) review criteria. The sponsor or investigator may need to revise and submit a study record one or more times before NLM's QC review criteria are met. It is the responsibility of the sponsor or investigator to ensure that the study record is consistent with the NLM QC review criteria.
October 29, 2026

First Posted (Actual)

First Posted

The date on which the study record was first available on ClinicalTrials.gov after National Library of Medicine (NLM) quality control (QC) review has concluded. There is typically a delay of a few days between the date the study sponsor or investigator submitted the study record and the first posted date.
November 4, 2026

Study Record Updates

Last Update Posted (Actual)

Last Update Posted

The most recent date on which changes to a study record were made available on ClinicalTrials.gov. There may be a delay between when the changes were submitted to ClinicalTrials.gov by the study's sponsor or investigator (the last update submitted date) and the last update posted date.
November 4, 2026

Last Update Submitted That Met QC Criteria

Last Update Submitted That Met QC Criteria

The most recent date on which the study sponsor or investigator submitted changes to a study record that are consistent with National Library of Medicine (NLM) quality control (QC) review criteria. It is the responsibility of the sponsor or investigator to ensure that the study record is consistent with the NLM QC review criteria.
October 29, 2026

Last Verified

Last Verified

The most recent date on which the study sponsor or investigator confirmed the information about a clinical study on ClinicalTrials.gov as accurate and current. If a study with a recruitment status of recruiting; not yet recruiting; or active, not recruiting has not been confirmed within the past 2 years, the study's recruitment status is shown as unknown.
October 1, 2026

More Information

Terms related to this study

Other Study ID Numbers

Other Study ID Numbers

Identifiers or ID numbers other than the NCT number that are assigned to a clinical study by the study's sponsor, funders, or others. These numbers may include unique identifiers from other trial registries and National Institutes of Health grant numbers.
  • LLMs success-Parasternal Block

Drug and device information, study documents

Studies a U.S. FDA-regulated drug product

No

Studies a U.S. FDA-regulated device product

No

This information was retrieved directly from the website clinicaltrials.gov without any changes. If you have any requests to change, remove or update your study details, please contact register@clinicaltrials.gov. As soon as a change is implemented on clinicaltrials.gov, this will be updated automatically on our website as well.

Cookie Settings

We use cookies to ensure that we give you the best experience on our website. For more details carefully read our Privacy and Cookies Policy. ...>>>You can change your preferences or withdraw your consent at any time by deleting the cookies from your website or computer as described in the policy. Please accept it and choose your preferences by ticking the corresponding boxes in the "Manage Settings" section below. Please carefully read our Terms of Service before you proceed with using any part of this website.
«Manage Settings