Related Experiment Video
Updated: Aug 9, 2026
Vessel-Sparing Microsurgical Longitudinal Intussusception Vasoepididymostomy to Treat Epididymal Obstructive Azoospermia
Published on: May 27, 2022
The use of Data Mining in the categorization of patients with Azoospermia
Themistoklis Mikos1, Nikolaos Maglaveras, Konstantinos Pantazis
1Unit of Reproductive Endocrinology, 1st Department of Obstetrics & Gynaecology, Aristotle University of Thessaloniki, Greece.
Objective:
Data Mining is a relatively new field of Medical Informatics. The aim of this study was to compare Data Mining diagnosis with clinical diagnosis by applying a Data Miner (DM) to a clinical dataset of infertile men with azoospermia.
Design:
One hundred and forty-seven azoospermic men were clinically classified into four groups: a) obstructive azoospermia (n=63), b) non-obstructive azoospermia (n=71), c) hypergonadotropic hypogonadism (n=2), and d) hypogonadotropic hypogonadism (n=11). The DM (IBM's DB2/Intelligent Miner for Data 6.1) was asked to reproduce a four-cluster model.
Results:
DM formed four groups of patients: a) eugonadal men with normal testicular volume and normal FSH levels (n=86), b) eugonadal men with significantly reduced testicular volume (median 6.5 cm3) and very high FSH levels (n=29), c) eugonadal men with moderately reduced testicular volume (median 14.5 cm3) and raised FSH levels (n=20), and d) hypogonadal men (n=12). Overall DM concordance rate in hypogonadal men was 92%, in obstructive azoospermia 73%, and in non-obstructive azoospermia 69%.
Conclusions:
Data Mining produces clinically meaningful results but different from those of the clinical diagnosis. It is possible that the use of large sets of structured and formalised data and continuous evaluation of DM results will generate a useful methodology for the Clinician.