Application of machine learning tools for evaluating the impact of premenopausal hysterectomy on serum anti-mullerian hormone levels

Boddupally, Sreenu ; SV, Kameswari ; Shaik Mohammad, Naushad ; Vijay Kumar, Kutala

Abstract

Women who have had premenopausal total hysterectomy at a young age could probably experience partial or total loss of ovarian function. The purpose of this retrospective cross-sectional study is to investigate the ovarian function in women underwent hysterectomy at an early age. A total of 1165 subjects comprised of 685 hysterectomised women and 480 age matched controls were enrolled in the study. We found that there is a steady decline in serum Anti Mullerian Hormone (AMH) levels, a marker of ovarian function after every five years post - hysterectomy in early age groups (20-30 yr and 31-40 yr) followed by loss of ovarian function in the age group of 40-50 yr. The application of multiple linear regression and machine learning tools has revealed that AMH is positively correlated with LH and estradiol and negatively correlated with age, FSH, years since hysterectomy and vitamin D. Serum AMH level of <0.08 ng/ml is associated with the increased of FSH, decreased LH and estradiol. The decreased ovarian function is associated with lower calcium levels, which are likely to influence the bone health. In conclusion, by utilizing multiple linear regression and machine learning tools, we found that serum FSH is the most important in predicting the AMH-mediated ovarian function.

Keyword(s)

Anti-Mullerian Hormone (AMH); Hysterectomy; Pre-menopause; Ovarian function; Oxidative stress

Full Text: PDF (downloaded 91 times)

Refbacks

  • There are currently no refbacks.
This abstract viewed 110 times