Artificial Intelligence in Modern Fertility Technologies

In vitro fertilisation (IVF) represented a major technological breakthrough when it first became available more than forty years ago. However, despite incremental improvements, overall IVF success rates have plateaued in recent years.

Now fertility clinics and technology developers are using AI to improve clinical outcomes once more. AI is being applied across multiple stages of the reproductive journey, including ovulation tracking, fertility diagnostics, fertilisation, and embryo selection. In each case, the underlying purpose of the AI is similar: to identify and analyse subtle patterns in large and complex biological datasets, which may not be readily identifiable through conventional analytical techniques.

This blog highlights some AI-enabled fertility technologies currently implemented by UK-based organisations.

Ovulation Tracking

Ovul is an ovulation tracking system that combines a reusable saliva‑testing device with an AI‑powered mobile application. Unlike conventional ovulation prediction methods based on calendar estimates, basal body temperature measurements, or urine-based hormone tests, Ovul monitors oestrogen-related changes reflected in saliva.

As oestrogen levels rise in the days leading up to ovulation, saliva exhibits a well-documented phenomenon known as ferning, in which in which crystallisation patterns form as the saliva dries. Ovul’s AI system has been trained on more than 10,000 images to predict ovulation by analysing these crystallisation patterns. In use, a small saliva sample is applied to the device. An internal optical sensor captures an image of the saliva crystallisation pattern, which is then analysed by the AI algorithm to estimate proximity to ovulation. The analysis results are delivered directly to the user via the Ovul application.

Ovul’s approach demonstrates the ability of AI to enable continuous and real-time ovulation tracking.

Fertility Diagnostics

AI also plays an increasingly important role in fertility diagnostics.

ExSeed is an at‑home male fertility testing system designed to provide clinical‑grade sperm analysis using consumer hardware. The system effectively converts a smartphone into a digital microscopy platform. A semen sample is placed on a slide, inserted into the ExSeed device, and imaged using the phone’s camera at approximately 200× magnification while a short video is recorded. This video data is analysed using ExSeed’s proprietary AI-driven sperm analysis algorithm. From the captured video data, the AI algorithm quantifies key parameters including sperm motility, concentration, and volume, to an accuracy of up to 97% compared to standard laboratory testing.

ExSeed illustrates how AI-based analysis can migrate traditionally laboratory-bound diagnostic procedures into a home setting without sacrificing analytical rigour.

Fertilisation

AI technologies are also being deployed by fertility clinics to support the selection of sperm and oocytes for fertilisation.

Avenues, a London-based fertility clinic, applies AI-driven analysis throughout the reproductive process. In the context of fertilisation, AI is used to analyse and rank individual spermatozoa within a sperm sample. By identifying subtle patterns in microscopy images, the AI system is able to detect indicators of sperm health that may not be readily apparent to human observers. Avenues similarly applies AI in the assessment of egg viability. This model analyses features associated with oocyte quality and has been reported to outperform manual assessment methods. By applying AI to both sperm and egg selection, clinicians are better equipped to identify combinations most likely to result in successful fertilisation.

These applications highlight the role of AI as a decision-support tool, augmenting rather than replacing clinical expertise.

Embryo evaluation and selection

Another use of AI by fertility clinics lies in embryo evaluation and selection.

Care Fertility uses artificial intelligence to support embryo evaluation and selection through their in‑house “Caremaps-AI” system. Caremaps-AI is designed to assist embryologists in identifying the embryos most likely to result in a live birth.

The system operates on time-lapse image data acquired during embryo development. Embryos are cultured in incubators equipped with integrated cameras that automatically capture images at regular intervals from fertilisation through to the blastocyst stage. AI analyses the full developmental trajectory captured in these time-lapse images to identify temporal and morphological patterns associated with embryo viability. Each embryo is then assigned a ranking score representing its likelihood of resulting in a live birth.

The Caremaps-AI system demonstrates the capability of AI in optimising embryo evaluation and selection in a crucially non-invasive manner.

Concluding Remarks

Across ovulation tracking, diagnostics, fertilisation and embryo selection, AI technologies are enabling new forms of assessment in fertility treatment. By uncovering patterns within complex physiological and developmental data, these systems support more informed clinical decision-making while reducing reliance on subjective, lab-based, or invasive techniques. As datasets continue to expand and models mature, AI is likely to play an increasingly central role in modern fertility care.