Skip to content

AI-assisted diagnosis of spinal lesions using multimodal multisequence MRI

AI-assisted diagnosis of spinal lesions using multimodal multisequence MRI

Status
Active, not recruiting
Phases
Unknown
Study type
Observational
Source
ChiCTR
Registry ID
ChiCTR2600123943
Enrollment
Unknown
Registered
2026-05-02
Start date
2026-05-02
Completion date
Unknown
Last updated
2026-05-04

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

vertebral lesions

Interventions

Vertebral lesion observation group:None

Sponsors

Qilu Hospital of Shandong University
Lead Sponsor

Eligibility

Sex/Gender
All

Inclusion criteria

Inclusion criteria: 1. Retrospective component (1) Patients who visited Qilu Hospital of Shandong University between January 2012 and December 2023; (2) Patients with vertebral tumors confirmed by surgical biopsy/resection pathology, or vertebral infections confirmed by histopathology or bacterial culture; (3) Complete preoperative MRI data, including at least T1-weighted imaging (T1WI), T2-weighted imaging (T2WI), and contrast-enhanced T1WI, with acceptable image quality; (4) Availability of the basic clinical information required for model operation, including complete medical records and basic laboratory test results. 2. Prospective component (1) Newly presenting patients with suspected vertebral lesions consecutively enrolled at Qilu Hospital of Shandong University from April 2026 to July 2028; (2) Patients undergoing MRI examination for clinical purposes, with complete MRI data including at least T1WI, T2WI, and contrast-enhanced T1WI, and with acceptable image quality; (3) Patients for whom subsequent diagnostic confirmation can be obtained and who can complete the follow-up assessment required by the study protocol; (4) Availability of the basic clinical information required for model operation, including complete medical records and basic laboratory test results.

Exclusion criteria

Exclusion criteria: 1. Retrospective component (1) Cases with insufficient diagnostic evidence, for which a definite diagnosis of vertebral tumor or vertebral infection could not be established; (2) Missing key preoperative MRI sequences, including T1-weighted imaging (T1WI), T2-weighted imaging (T2WI), or contrast-enhanced T1WI; (3) MRI images of inadequate quality that could not be used for lesion detection, segmentation, or classification; (4) Missing key clinical data or basic laboratory test results required for model operation. 2. Prospective component (1) Failure to complete the required MRI examination, or incomplete key MRI sequences; (2) MRI images of inadequate quality that could not be used for model analysis; (3) Inability to obtain subsequent diagnostic confirmation or to complete the required follow-up assessment; (4) Missing key clinical data or basic laboratory test results required for model operation; (5) Refusal to sign the informed consent form or withdrawal from the study during the study period.

Design outcomes

Primary

MeasureTime frame
Area under the receiver operating characteristic curve (AUC);Accuracy;Sensitivity;Specificity;F1-score;

Countries

China

Contacts

Public ContactHaipeng Si

Qilu Hospital of Shandong University

sihaipeng1978@email.sdu.edu.cn+86 185 6008 2571

Outcome results

None listed

Source: ChiCTR (via WHO ICTRP) · Data processed: May 7, 2026