EmbryoScope and AI in IVF: What They Can Refine and What They Cannot

Key Takeaways

EmbryoScope records growth inside the incubator, while AI can help the team choose an embryo for transfer. Time-lapse systems have not been shown to raise the chance of a live birth for most patients. An AI score cannot replace a chromosome test or tell you whether an embryo will implant.

Key evidence: HFEA assessment of time-lapse imaging and incubation · TILT trial: live birth with time-lapse systems (2024) · AI prediction of embryo chromosome status: review (2024)

What Is EmbryoScope?

EmbryoScope is an incubator with a built-in camera. It keeps embryos in a controlled culture chamber while taking pictures as they grow. The team can review those images without taking the culture dish out for each check.

The incubator, its image record and any AI software have different roles. Owning an EmbryoScope does not mean a clinic uses AI to choose embryos. Ask which tools the team uses and what each one adds to your care.

How Does Time-Lapse Imaging Work?

The camera takes pictures at set intervals and at different depths of focus. Software joins them into a film. This lets the team see events that a brief check under a microscope could miss.

The team can review:

• When an embryo first divides and how later divisions unfold.
• Whether the cells divide together and look similar in size.
• When the cells start to pack closely together, called compaction.
• When a fluid-filled space forms and the embryo becomes a blastocyst, a later stage of growth.

The study of these changes and their timing is called morphokinetics. The ESHRE time-lapse recommendations describe how the images can reveal unusual cell divisions.

Where Can AI Help in Embryo Selection?

AI software learns patterns from embryo images and records of their outcomes. It can use those patterns to give an embryo a score. Each model has its own scoring system; the number is not your personal chance of having a baby.

AI can help the team compare images, record features and review scores in a consistent way. That may save work, but a consistent score can still be wrong. The ASRM 2026 opinion on laboratory AI stresses the need to test tools on new data and in clinical care.

A score can be one part of the choice when two or more embryos could be transferred. It adds to the embryo grade, which describes what the embryo looks like under a microscope.

What Are the Real Advantages for Patients?

The clearest gains are in how the laboratory observes and records embryo growth. A useful lab tool does not always lead to a better treatment outcome.

• Fewer checks outside the chamber: Images are taken inside the incubator, reducing exposure to changes in temperature and gas levels during checks.
• A record to review: The team can look back at the images and explain what it saw. This does not replace the identity checks used to prevent embryo mix-ups.
• More information for comparison: If embryos look similar, their growth record may add detail to the transfer decision.

These uses are described in the ESHRE time-lapse recommendations. They should be kept separate from claims about having a baby.

What Are the Main Biological and Clinical Limitations?

1. Implantation is a multi-factorial biological process

Implantation means an embryo attaches to the lining of the womb. An image score does not assess that lining, its timing or all the other factors involved. Even the highest-scoring embryo may not implant.

2. Visual appearance does not equal chromosomal normality

An embryo can look good and still have an unusual number of chromosomes. Some AI models try to predict chromosome status from images. A prediction is different from testing cells. The Xin review found that current models cannot fully replace invasive tests for embryo chromosome status. This limit matters especially as maternal age rises and abnormal chromosome counts become more common (Franasiak et al., 2014).

3. A live-birth benefit has not been shown

Time-lapse systems have not been shown to improve the chance of a live birth for most patients. The HFEA assessment rates both manual and automated image analysis black for this outcome. That means the balance of moderate- or high-quality evidence shows no effect on the chance of having a baby.

This assessment concerns time-lapse systems as an add-on. It does not prove that every AI product has the same effect in every patient group.

4. Whether algorithms transfer between clinics

A tool that works well in one laboratory may perform differently in another. Equipment, culture methods and the patients treated can differ. The ASRM AI opinion calls for careful validation. Ask whether the tool has been checked in patients like you and in that laboratory.

Does EmbryoScope Replace Traditional Embryology?

No. The camera records growth; the team still has to interpret the findings. Embryologists check image quality, compare embryos from the same egg collection and use the treatment plan to guide the choice.

The Illingworth trial did not establish that its AI tool performed within the agreed limit of standard embryo selection. That finding does not, by itself, prove AI was worse. It also gives no basis for replacing the embryologist. The trial was funded by Vitrolife, which makes the EmbryoScope.

Does It Replace PGT-A?

No. PGT-A tests the number of chromosomes in cells taken from the embryo’s outer layer, called the trophectoderm. Images cannot supply the same test result.

PGT-A also has limits and is not a routine test for everyone. The ASRM 2024 opinion says its value as a screening test for all IVF patients has not been shown. Age, past treatment, the number of blastocysts and the possible findings all matter. Results such as mosaicism, where the sample suggests a mix of cells with different chromosome counts, need careful discussion.

Who Might Benefit Most?

The number of embryos changes what a selection tool can add. Here, a possible benefit means more information for a decision, rather than a proven rise in live births.

• Two or more embryos: The team has a choice about which to transfer first. Images or scores may help compare them.
• One embryo: There is no choice between embryos to rank. The incubator still provides a culture environment and an image record, but ranking software has little to add to that choice.
• A previous cycle with image records: The team can review unusual divisions, including reverse cleavage, when cells appear to merge again after dividing. These do not necessarily explain why a transfer failed.

Your wider outlook still depends on age and other clinical factors. Our guide to IVF success rates explains why a clinic-wide rate is not your personal chance.

Practical Questions to Ask Your Fertility Clinic

Ask what the technology would change in your care:

1. Is time-lapse culture standard here, or does it cost extra? Ask what the fee covers.
2. Which software do you use? Check whether the team uses the film, an AI score or both, and who makes the final choice.
3. What has the tool been tested for? A score that predicts embryo appearance is not proof of more live births. Ask about checks in that laboratory.
4. What would change with one embryo, or with two? One leaves no ranking choice; two allow comparison.
5. How do you resolve a score that conflicts with other findings? Ask how the team weighs it alongside the embryo grade, any PGT-A result and the plan for preparing the womb lining.

Clinical Note

In daily practice, EmbryoScope and AI are most useful when they make a laboratory decision clearer and more consistent. When several blastocysts look similar under the microscope, reviewing their development over time can add useful information when deciding which embryo to transfer first.

However, I always remind patients that implantation depends on more than an embryo’s image or score. Technology can refine a laboratory decision, but it cannot promise implantation or a live birth.

— Dr. Senai Aksoy

Frequently Asked Questions for Dr. Aksoy

Is EmbryoScope worth paying extra for?

EmbryoScope is a useful laboratory tool. It lets us follow embryo growth without taking embryos out of the incubator. But large, reliable studies have not shown a rise in live-birth rates compared with standard embryo culture.

I therefore start by asking how many embryos the patient has, what the clinic’s standard laboratory conditions are, how much the extra fee is and what this tool would change in their case. Particularly if the patient has only one or very few embryos, the expected extra help in choosing between embryos is limited.

I am not against using EmbryoScope. But I do not see a high extra fee as necessary when it is paid in the expectation of a better chance of pregnancy.

— Dr. Senai Aksoy

Sources

• Apter S, Ebner T, Freour T, Guns Y, Kovacic B, Le Clef N, et al. Good practice recommendations for the use of time-lapse technology. Human Reproduction Open. 2020;2020(2):hoaa008. PubMed · DOI
• Armstrong S, Bhide P, Jordan V, Pacey A, Marjoribanks J, Farquhar C. Time-lapse systems for embryo incubation and assessment in assisted reproduction. Cochrane Database of Systematic Reviews. 2019;(5):CD011320. PubMed · DOI
• Bhide P, Chan DYL, Lanz D, Alqawasmeh O, Barry E, Baxter D, et al. Clinical effectiveness and safety of time-lapse imaging systems for embryo incubation and selection in in-vitro fertilisation treatment (TILT): a randomised controlled trial. The Lancet. 2024;404(10449):256-265. PubMed · DOI
• Franasiak JM, Forman EJ, Hong KH, Werner MD, Upham KM, Treff NR, et al. The nature of aneuploidy with increasing age of the female partner: a review of 15,169 consecutive trophectoderm biopsies evaluated with comprehensive chromosomal screening. Fertility and Sterility. 2014;101(3):656-663.e1. PubMed · DOI
• Illingworth PJ, Venetis C, Gardner DK, Nelson SM, Berntsen J, Larman MG, et al. Deep learning versus manual morphology-based embryo selection in IVF: a randomized, double-blind noninferiority trial. Nature Medicine. 2024;30(11):3114-3120. PubMed · DOI
• Xin X, Wu S, Xu H, Ma Y, Bao N, Gao M, et al. Non-invasive prediction of human embryonic ploidy using artificial intelligence: a systematic review and meta-analysis. EClinicalMedicine. 2024;77:102897. PubMed · DOI
• Practice Committee and Genetic Counseling Professional Group of the American Society for Reproductive Medicine. The use of preimplantation genetic testing for aneuploidy: a committee opinion. Fertility and Sterility. 2024;122(3):421-434. ASRM PGT-A opinion (2024) · DOI
• Practice Committee of the American Society for Reproductive Medicine. Artificial intelligence in the in vitro fertilization laboratory: a committee opinion. Fertility and Sterility. 2026;126:494-498. ASRM laboratory AI opinion (2026)
• Human Fertilisation and Embryology Authority. Time-lapse imaging and incubation. Treatment add-on evidence assessment; accessed 31 October 2026. HFEA

Dr. Senai Aksoy

Dr. Senai Aksoy studied and trained in France before returning to Turkey, where he was a member of the ICSI team at Sevgi Hospital, Ankara — the country's first ICSI centre (1994-95) — and a co-author on the first Turkish ICSI publications produced in collaboration with the Brussels Van Steirteghem group (Human Reproduction, 1996; PMID 8671323). He helped build the IVF programme at the American Hospital Istanbul and has been running his own fertility practice since 1998.

Verified profiles: PubMed · ORCID · LinkedIn

Read full profile → Instagram YouTube

The content has been created by Dr. Senai Aksoy and medically approved.