Evaluation of Cavernous Sinus Invasion by Pituitary Adenoma Using Deep Learning Based Denoising MR

Asan Medical Center

Status: Completed

Conditions: Cavernous Sinus Invasion by Pituitary Adenoma

Treatments: Diagnostic Test: MRI with deep learning based denoising

Study type: Interventional

Funder types: Other

Identifiers

NCT04268251

AsanMCHSKim_06

Details and patient eligibility

About

Preoperative evaluation of cavernous sinus invasion by pituitary adenoma is critical for performing safe operation and deciding on surgical extent as well as for treatment success. Because of the small size of the pituitary gland and sellar fossa, determining the exact relationship between the pituitary adenoma and cavernous sinus can be challenging. Performing thin slice thickness MRI may be beneficial but is inevitably associated with increased noise level. By applying deep learning based denoising algorithm, diagnosis of cavernous sinus invasion by pituitary adenoma may be improved.

Enrollment: 67 patients

Sex: All

Ages: 18+ years old

Volunteers: No Healthy Volunteers

Inclusion criteria

• Patients undergoing preoperative brain MR for pituitary adenoma

Exclusion criteria

• Patients who have any type of bioimplant activated by mechanical, electronic, or magnetic means (e.g., cochlear implants, pacemakers, neurostimulators, biostimulates, electronic infusion pumps, etc), because such devices may be displaced or malfunction
• Patients who are pregnant or breast feeding; urine pregnancy test will be performed on women of child bearing potential
• Poor MRI image quality due to artifacts

Trial design

Primary purpose: Diagnostic

Allocation: N/A

Interventional model: Single Group Assignment

Masking: None (Open label)

---

67 participants in 1 patient group

Deep learning based denoising MR

Experimental group

Description:

1 mm slice thickness coronal contrast-enhanced T1 weighted imaging with deep learning based denoising vs. 3 mm slice thickness coronal contrast-enhanced T1 weighted imaging

Treatment:

Diagnostic Test: MRI with deep learning based denoising