Skip to content

Screening of Multimodal Imaging Fusion and Segmentation Algorithm Models for Oral and Maxillofacial Tumors Based on Deep Learning

Screening of Multimodal Imaging Fusion and Segmentation Algorithm Models for Oral and Maxillofacial Tumors Based on Deep Learning

Status
Active, not recruiting
Phases
Unknown
Study type
Observational
Source
ChiCTR
Registry ID
ChiCTR2500106866
Enrollment
Unknown
Registered
2025-07-31
Start date
2025-08-01
Completion date
Unknown
Last updated
2025-08-18

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

Conditions

Oral and maxillofacial tumors

Interventions

Observation group of patients with oral and maxillofacial tumors:None

Sponsors

Peking University Hospital of Stomatology
Lead Sponsor

Eligibility

Sex/Gender
All

Inclusion criteria

Inclusion criteria: 1. The tumor is located in the deep area of the mouth (gingival region, soft palate, base of the tongue or parapharynx, etc.) or deep maxillofacial area (maxilla, ascending branch of the mandible, zygomatic bone, or skull base-subtemporal fossa, etc.), involving two or more anatomical areas; 2. Underwent CT and MRI examination before surgery, and possessed imaging data in digital imaging and communication in medicine (DICOM) format in two modalities; 3. The scanning range of imaging examination is below the upper boundary of the lower edge of the orbit and below the apex of the lower transorbital area. 4. The tumor is primary or recurrent, with or without pathological diagnosis.

Exclusion criteria

Exclusion criteria: 1. The image shows that the upper and lower teeth are not in the median position when the patient undergoes imaging examination, and the position relationship between the tumor and the surrounding anatomical structure of the patient may be offset between different images due to occlusal instability; 2. The interval between the two imaging examinations is more than 20 days, in which case the tumor may be deformed between different images due to volume growth.

Design outcomes

Primary

MeasureTime frame
Dice coefficient;MI (Mutual Information);

Countries

China

Contacts

Public ContactBinzhang Wu

Peking University School and Hospital of Stomatology

wubinzhang@bjmu.edu.cn+86 10 5329 5051

Outcome results

None listed

Source: ChiCTR (via WHO ICTRP) · Data processed: Feb 4, 2026