Predicting DNA fragmentation: A non-destructive analogue to chemical assays using machine learning

TL;DR

This study develops a non-destructive framework to predict sperm DNA fragmentation from unstained images by fusing hand-crafted morphometrics with a vision transformer backbone trained through transfer learning. Using data from 65 men across multiple chemical assays (AB, TB, AO, CMA3, TUNEL, SCD), the approach builds per-assay predictive models and assesses performance on brightfield and phase-contrast images, with TUNEL showing the strongest signal (AUC up to ~0.82). The results reveal assay- and modality-specific differences, with phase-contrast imagery improving several predictions while others favor brightfield, and demonstrate the feasibility of guiding sperm selection for ART without destroying samples. The work highlights the potential for clinical impact in ART by enabling rapid, non-destructive DNA integrity assessment and lays groundwork for integration into real-time decision-support tools to improve IVF/ICSI outcomes.

Abstract

Globally, infertility rates are increasing, with 2.5\% of all births being assisted by in vitro fertilisation (IVF) in 2022. Male infertility is the cause for approximately half of these cases. The quality of sperm DNA has substantial impact on the success of IVF. The assessment of sperm DNA is traditionally done through chemical assays which render sperm cells ineligible for IVF. Many compounding factors lead to the population crisis, with fertility rates dropping globally in recent history. As such assisted reproductive technologies (ART) have been the focus of recent research efforts. Simultaneously, artificial intelligence has grown ubiquitous and is permeating more aspects of modern life. With the advent of state-of-the-art machine learning and its exceptional performance in many sectors, this work builds on these successes and proposes a novel framework for the prediction of sperm cell DNA fragmentation from images of unstained sperm. Rendering a predictive model which preserves sperm integrity and allows for optimal selection of sperm for IVF.

Figures (11)

• Figure 1: Schematic of ensemble model architecture for binary classification problem.
• Figure 2: Aniline Blue ROC Curves
• Figure 3: Aniline Blue Example Images
• Figure 4: Acridine Orange ROC Curves
• Figure 5: Acridine Orange Example Images