Automatic segmentation of vestibular schwannomas using machine learning (ML) and correlation to its clinical symptoms (hearing loss, vertigo and tinnitus)
- Conditions
- D33.3Cranial nerves
- Registration Number
- DRKS00034593
- Lead Sponsor
- niversitätsklinikum Freiburg, Klinik für Hals-, Nasen- und Ohrenheilkunde
- Brief Summary
Not available
- Detailed Description
Not available
Recruitment & Eligibility
- Status
- Complete
- Sex
- All
- Target Recruitment
- 216
Inclusion Criteria
MR image in T1 present
vestibular schwannoma has been diagnosed
Exclusion Criteria
postoperative patients
patients with neurofibomatosis type II
no sufficient information about the patient available
Study & Design
- Study Type
- observational
- Study Design
- Not specified
- Primary Outcome Measures
Name Time Method Correlation of symptoms (occurrence and intensity) with machine learning segmentated volume of vestibular schwannoma.<br>Is a larger volume of an acoustic neuroma associated with a higher frequency or greater intensity of symptoms such as hearing loss, vertigo and tinnitus?<br>Is a specific growth pattern (maximum length*width*height as well as intra- or extrameatal fraction) associated with increased symptom severity?
- Secondary Outcome Measures
Name Time Method Which frequency is most affected by hearing loss by the growth of an acoustic neuroma? <br>How well does automated segmentation by a neural network correlate with or overlap manual segmentation?<br>Can automated segmentation of vestibular schwannoma give us information about the likelihood of occurence and intensity of symptoms?<br>How ist growth of acoustic neuroma ifluenced by stereotaktic radiotherapie?<br>Do Gender, Age or side of the AN korrelate with size and growth rate of AN?