Experimental exposure to gasohol impairs sperm quality with recognition of the classification pattern of exposure groups by machine learning algorithms

Abstract

Contamination caused by leakage at gas stations leads to possible exposure of the general population when in contact with contaminated water and soil. The present study aimed to evaluate the reproductive effects of exposure of adult male rats to gasohol and evaluate the performance of machine learning (ML) algorithms for pattern recognition and classification of the exposure groups. Rats were orally exposed to 0 (control), 16 (EA), 160 (EB), or 800 mg kg⁻¹ bw day⁻¹ of gasohol (EC), for 30 consecutive days. Sperm quality of the groups exposed to two higher doses was reduced in comparison to the control group. The sperm parameters decreased were: daily sperm production, sperm number in the caput/corpus epididymis, progressive motility, mitochondrial activity, and acrosomal membrane integrity. Sperm transit time in the epididymis cauda and sperm isolated head were increased in EB and EC. Sertoli cells number was decreased in these groups, but their support capacity was maintained. ML methods were used to identify patterns between samples of control and exposure groups. The results obtained by ML methods were very promising, obtaining about 90% of accuracy. It was concluded that the exposure of rats to different doses of gasohol impair spermatogenesis and sperm quality, with a recognizable classification pattern of exposure groups at ML.

https://ift.tt/2Pxge1q