ChatGPT in the Diagnosis and Management of Complex Polyneuropathies: Comparative Analysis With Neurologists Using Real-World Cases

NCT07493681 | Completed

• 1 other identifier: 485/24
• Study Type: observational
• Target: 100
• Locations: 1 country
• Sites: 1

Brief Summary

BACKGROUND AND PURPOSE Polyneuropathies are diseases affecting the peripheral nerves that occur in approximately 1% of the general population, rising to up to 13% among older adults. Despite their prevalence, accurate diagnosis is often challenging and requires specialist expertise that is not uniformly available. Patients evaluated in primary care or non-specialist settings frequently experience diagnostic delays or misdiagnoses, highlighting the need for innovative tools to support clinicians at critical points in the diagnostic process. Artificial intelligence (AI) large language models (LLMs), such as ChatGPT, are increasingly being explored as potential aids in clinical diagnosis. These tools can process complex clinical information and generate diagnostic suggestions at low cost and with broad accessibility. However, their performance in specialised neurological conditions, particularly complex polyneuropathies, has not yet been rigorously evaluated in real-world settings. STUDY OBJECTIVES This study aims to evaluate the diagnostic performance of ChatGPT-4o on real-world polyneuropathy cases and to compare it with that of peripheral nerve disease specialists and non-specialist neurologists. A secondary objective is to assess whether exposure to ChatGPT-4o outputs influences and potentially improves neurologist diagnostic accuracy. STUDY DESIGN This will be a comparative diagnostic accuracy study conducted at two tertiary referral centres for peripheral neuropathies in Milan, Italy. One hundred patients with confirmed polyneuropathy diagnoses will be randomly selected from consecutive outpatients. Each case will be summarised in a standardised format including demographics, symptom history, neurological examination findings, nerve conduction study results, and screening laboratory data. Only cases with a diagnosis confirmed after at least 12 months of clinical follow-up will be included. ChatGPT-4o will be presented with each case using a structured prompt, and will be asked to provide: (1) a leading diagnosis, (2) two alternative differential diagnoses, and (3) a single recommended confirmatory diagnostic test. The model will be run in two independent trials to assess response consistency. The same 100 cases will also be reviewed by neurologists from multiple international centres. Participants will be classified as either peripheral nerve disease specialists, neurologists routinely practising in tertiary polyneuropathy centres, or non-specialists, including general neurologists or those sub-specialised in other fields. Neurologists will first provide their own diagnostic assessments independently, and will subsequently be shown ChatGPT-4o's output with the option to revise their responses. EXPECTED SIGNIFICANCE This study will provide evidence on whether AI-based LLMs can serve as reliable diagnostic aids in complex polyneuropathy cases.

Conditions

Polyneuropathies

Keywords

Polyneuropathies; Artificial Intelligence; Large Language Models; ChatGPT

Outcome Measures

Primary Outcomes (1)

• Diagnostic Accuracy of Leading Diagnosis

  Proportion of cases in which the leading diagnosis proposed by ChatGPT-4o, specialist neurologists, or non-specialist neurologists matches the patient's final confirmed etiological diagnosis.

  through study completion, an average of 6 months

Secondary Outcomes (3)

• Accuracy of Differential Diagnoses

  through study completion, an average of 6 months

• Change in Neurologist Diagnostic Accuracy After AI Review

  through study completion, an average of 6 months

• Consistency of ChatGPT-4o Responses Across Independent Trials

  through study completion, an average of 6 months

Study Arms (1)

Polyneuropathy patients

Patients with a confirmed etiological diagnosis of polyneuropathy evaluated at tertiary neuromuscular referral centers. Clinical data from these patients will be used to generate standardized anonymized case summaries that will be evaluated by neurologists and by an artificial intelligence-based large language model (ChatGPT-4o) for diagnostic accuracy.

Other: ChatGPT-4o; Other: chatGPT

Interventions

ChatGPT-4o OTHER

ChatGPT-4o Enterprise (OpenAI) is evaluated as an artificial intelligence-based large language model for clinical diagnostic reasoning. Standardized anonymized clinical case summaries will be presented to the model, which will generate a leading diagnosis, two alternative differential diagnoses, and one recommended confirmatory diagnostic test for each case.

Polyneuropathy patients

chatGPT OTHER

Case summaries will

Polyneuropathy patients

Eligibility Criteria

• Sex: all
• Healthy Volunteers: No
• Age Groups: Child (0-17), Adult (18-64), Older Adult (65+)
• Sampling Method: Non-Probability Sample

Study Population

Cases will be identified from patients with a confirmed diagnosis of polyneuropathy consecutively attending the polyneuropathy outpatient clinics of the two participating institutions. Demographic, clinical, and diagnostic data will be extracted from medical records at two time points: the initial neurological consultation and the first follow-up visit after completion of standard screening investigations, including laboratory tests and nerve conduction studies.

You may qualify if:

• Confirmed diagnosis of polyneuropathy of any etiology
• Final etiological diagnosis confirmed after at least 12 months of clinical follow-up
• Availability of sufficient clinical, electrophysiological, and laboratory information in the medical record to allow preparation of a standardized case summary

You may not qualify if:

• \- Absence of a confirmed etiological diagnosis (idiopathic neuropathy)

Sponsors & Collaborators

• Istituto Clinico Humanitas: lead
• Fondazione I.R.C.C.S. Istituto Neurologico Carlo Besta: collaborator

Study Sites (1)

Humanitas Research Hospital

Milan, MI, 20089, Italy

Location

Related Publications (4)

• Fonseca A, Ferreira A, Ribeiro L, Moreira S, Duque C. Embracing the future-is artificial intelligence already better? A comparative study of artificial intelligence performance in diagnostic accuracy and decision-making. Eur J Neurol. 2024 Apr;31(4):e16195. doi: 10.1111/ene.16195. Epub 2024 Jan 18.

  PMID: 38235841 BACKGROUND

• Kanjee Z, Crowe B, Rodman A. Accuracy of a Generative Artificial Intelligence Model in a Complex Diagnostic Challenge. JAMA. 2023 Jul 3;330(1):78-80. doi: 10.1001/jama.2023.8288.

  PMID: 37318797 BACKGROUND

• Uwishema O, Boon P. Bridging the Gaps: Addressing Inequities in Neurological Care for Underserved Populations. Eur J Neurol. 2025 Feb;32(2):e70073. doi: 10.1111/ene.70073. No abstract available.

  PMID: 39912252 BACKGROUND

• Elafros MA, Kvalsund MP, Callaghan BC. The Global Burden of Polyneuropathy-In Need of an Accurate Assessment. JAMA Neurol. 2022 Jun 1;79(6):537-538. doi: 10.1001/jamaneurol.2022.0565. No abstract available.

  PMID: 35404377 BACKGROUND

MeSH Terms

Conditions

Polyneuropathies

Condition Hierarchy (Ancestors)

Peripheral Nervous System Diseases; Neuromuscular Diseases; Nervous System Diseases

Study Officials

• Pietro Emiliano Doneddu, MD, Neurologist

  Humanitas Research Hospital IRCCS, Rozzano-Milan

  PRINCIPAL INVESTIGATOR

Study Design

• Study Type: observational
• Observational Model: OTHER
• Time Perspective: OTHER
• Target Duration: 12 Months
• Sponsor Type: OTHER
• Responsible Party: SPONSOR

Study Record Dates

• First Submitted: March 15, 2026
• First Posted: March 25, 2026
• Study Start: June 12, 2023
• Primary Completion: January 15, 2025
• Study Completion: March 1, 2025
• Last Updated: March 25, 2026

Record last verified: 2026-03

Data Sharing

• IPD Sharing: Will share
• Shared Documents: STUDY PROTOCOL, SAP, ICF, CSR, ANALYTIC CODE
• Time Frame: From March 2026 onwards
• Access Criteria: Reasonable academic request

Locations

IT (1)