Prediction of pain intensity with uterine morphological features and brain microstructural and functional properties in women with primary dysmenorrhea

Primary dysmenorrhea (PDM), defined as painful menstrual cramps of uterine origin, could cause brain structural and functional changes after long-term menstrual pain. Here, we aimed to investigate the predictive value of uterine morphological features and microstructural/functional properties of the...

Full description

Saved in:
Bibliographic Details
Published in:Brain imaging and behavior Vol. 15; no. 3; pp. 1580 - 1588
Main Authors: Quan, Shilan, Yang, Jing, Dun, Wanghuan, Wang, Ke, Liu, Hongjuan, Liu, Jixin
Format: Journal Article
Language:English
Published: New York Springer US 01.06.2021
Springer Nature B.V
Subjects:
Biomedical and Life Sciences
Biomedicine
Brain
Cramps
Diffusion
Feature extraction
Iterative methods
Magnetic resonance imaging
Menstruation
Morphology
Myometrium
Neural networks
Neuropsychology
Neuroradiology
Neurosciences
Optimization
Original Research
Pain
Performance prediction
Permutations
Predictions
Psychiatry
Regression analysis
Regression models
Structure-function relationships
Substantia alba
Substantia grisea
Texture
Thalamus
Uterus
Primary dysmenorrhea
Uterine morphology
Left thalamus
Prediction
Machine learning
ISSN:1931-7557, 1931-7565, 1931-7565
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Primary dysmenorrhea (PDM), defined as painful menstrual cramps of uterine origin, could cause brain structural and functional changes after long-term menstrual pain. Here, we aimed to investigate the predictive value of uterine morphological features and microstructural/functional properties of the brain extracted from periovulatory phases for the intensity of menstrual pain as rated by women with PDM during their subsequent menstrual period. Forty-five women with PDM were recruited and classified into the high and mild pain intensity groups. Pelvic MRI was employed to extract the uterine texture features. White matter diffusion properties, grey matter and functional connectivity features were extracted as brain features. Multivariate logistic regression models with iteration optimization were built for classifying different pain intensity groups. Texture features from myometrium and uterine junction zone had outstanding prediction performance with an area under the receiver operating characteristic (AUC) of 0.96 ( P  < 0.05, permutation test), and diffusion properties along the thalamic fiber bundles were the most discriminative features with AUC of 0.95. Applying features from uterus and brain together, we could gain better prediction performance. Our results indicated that accumulated differences in menstrual pain were associated not only with uterine structure but also diffusion properties of thalamic-related fiber tracts, suggesting that treatment options of PDM patients may be expanded from only being able to manage pain in the uterus focusing on the functional/structural modifications of the pain processing system.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
ISSN:1931-7557
1931-7565
1931-7565
DOI:10.1007/s11682-020-00356-w