Construction of AI Model for Precision Imaging Diagnosis of Cranial Diseases

The goal of this observational study is to develop and validate a high-precision AI diagnostic model for cranial diseases by integrating clinical knowledge systems (pathophysiological classification, age stratification, and anatomical localization) to simulate radiologists' diagnostic thinking. The main question it aims to answer is: Does the AI model improve diagnostic accuracy and consistency across different hospital levels, physician qualifications, and clinical scenarios compared to traditional diagnosis? Participants' cranial MRI data (including T1, T2, FLAIR, DWI sequences) and clinical information will be collected retrospectively (2015-2025) and prospectively (2026) to train and validate the model, which will be evaluated through performance metrics (accuracy, sensitivity, specificity) and clinical efficacy assessments (doctor vs. model, with/without model assistance). This study will establish a new paradigm for clinical AI implementation, providing methodological support for precision diagnosis of neurological diseases.

Study Overview

• Status: Recruiting
• Conditions: Cranial Diseases (e.g., Brain Tumors, Stroke, Neurodegenerative Disorders)

• Study Type: Observational
• Enrollment (Estimated): 1000

Contacts and Locations

Study Locations

• China
  • Hubei
    • Wuhan, Hubei, China, 430030
      • Recruiting
      • Tongji Hospital
      • Contact: Zhu Dr. Zhu, Professor; Phone Number: 86027-83663611; Email: zhuwenzhen8612@163.com

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Child; Adult; Older Adult
• Accepts Healthy Volunteers: Yes

• Sampling Method: Probability Sample

Study Population

The study population will consist of patients diagnosed with the 48 specified cranial diseases from various hospitals, including neurology departments of tertiary hospitals, specialized clinics, and community medical centers. Participants will be selected based on clinical diagnosis and availability of complete cranial imaging data, ensuring representation of diverse disease types and severities for model training and validation.

Description

Inclusion Criteria:

Patients clinically diagnosed with one of the following 48 cranial diseases: glioma, lymphoma, meningioma, pituitary macroadenoma, ependymoma, choroid plexus papilloma, schwannoma, medulloblastoma, metastatic tumor, chordoma, craniopharyngioma, germinoma, hemangioblastoma, cholesteatoma, teratoma, viral encephalitis, brain abscess, cerebral tuberculosis infection, cryptococcal encephalitis, cerebral cysticercosis, general soft meningitis (including bacterial, fungal infections, and autoimmune meningitis), tuberculous meningitis, ischemic stroke, cerebral venous sinus thrombosis, arteriovenous malformation, cavernous hemangioma, venous developmental malformation, aneurysm, cerebral small vessel disease, epidural hemorrhage, subdural hemorrhage, intracerebral hemorrhage, cerebral contusion, subarachnoid hemorrhage, diffuse axonal injury, multiple sclerosis, hippocampal sclerosis, focal cortical dysplasia, cortical or cerebral fissure malformations, basilar invagination, Chiari malformation, Dandy-Walker malformation, Rathke's cleft cyst, pituitary hypoplasia, adrenoleukodystrophy, Alzheimer's disease, Parkinson's disease, arachnoid cyst.

Exclusion Criteria:

Patients with other severe neurological diseases not included in the 48 specified cranial diseases; Patients unable to provide complete cranial imaging data (e.g., missing images or images of poor quality); Patients with severe cognitive impairment unable to cooperate with the study (e.g., unable to understand study procedures or communicate with site personnel); Pregnant or lactating women; Patients with severe dysfunction of major organs (e.g., heart, liver, kidney) that cannot tolerate study-related examinations; Patients who refuse to sign the informed consent form.

Study Plan

How is the study designed?

Number of groups / cohorts

2

Cohorts and Interventions

Group / Cohort
Retrospective data collection from January 1, 2015 to December 31, 2025, including 70,000 subjects; including 70,000 unsupervised training cases and 2,000 supervised training cases for model development.
Prospective Cohort; Prospective data collection from March 1, 2026 to December 31, 2026, including 1,000 cases for model validation and clinical efficacy evaluation.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Diagnostic accuracy of AI model (sensitivity, specificity, AUC)	The primary outcome is the diagnostic performance of the AI model, including sensitivity, specificity, and area under the ROC curve (AUC), compared to the gold standard (e.g., histopathology or clinical diagnosis).	Within 1 month of image acquisition

Collaborators and Investigators

• Sponsor: Tongji Hospital

Study record dates

Study Major Dates

• Study Start (Actual): January 1, 2026
• Primary Completion (Estimated): December 1, 2026
• Study Completion (Estimated): December 1, 2026

Study Registration Dates

• First Submitted: February 25, 2026
• First Submitted That Met QC Criteria: February 25, 2026
• First Posted (Actual): March 3, 2026

Study Record Updates

• Last Update Posted (Actual): March 6, 2026
• Last Update Submitted That Met QC Criteria: March 5, 2026
• Last Verified: March 1, 2026

More Information

Terms related to this study

• Additional Relevant MeSH Terms: Cerebrovascular Disorders; Brain Diseases; Central Nervous System Diseases; Nervous System Diseases; Vascular Diseases; Cardiovascular Diseases; Neoplasms by Site; Neoplasms; Nervous System Neoplasms; Central Nervous System Neoplasms; Stroke; Brain Neoplasms; Neurodegenerative Diseases
• Other Study ID Numbers: TJ-IRB202601011 (Other Identifier: Tongji Hospital)

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