Diagnosis
Doctors report first pregnancy using AI system to detect sperm in men previously considered infertile
A couple who had been trying to conceive for 19 years have become the first to achieve pregnancy using an artificial intelligence system designed to detect sperm in men with azoospermia – a condition where no sperm are visible in semen.
The pregnancy was announced in March 2025 by doctors at Columbia University Fertility Center.
It followed the development of the STAR (Sperm Track and Recovery) system, led by Dr Zev Williams, director of the centre.
Male factors are thought to account for around 40 per cent of infertility cases in the US, with azoospermia responsible for roughly 10 per cent of male infertility.
Until recently, there was little doctors could do beyond recommending donor sperm.
Williams explained that semen samples from men with azoospermia can appear normal, but microscopic analysis shows no visible sperm among other cell debris.
As sperm are the smallest cells in the human body, even highly trained technicians often fail to identify them.
The Columbia team spent five years developing STAR, which combines an AI algorithm trained to detect sperm with a fluidic chip that channels the semen through microscopic tubules.
When the AI detects sperm, that portion of the sample is diverted into a separate channel for collection.
The recovered sperm can then be frozen or used for fertilisation.
Inspired by methods used by astrophysicists to find stars and planets, the STAR system is trained to detect what Williams calls “really, really, really rare sperm.”
He said: “If you can look into a sky that’s filled with billions of stars and try to find a new one, or the birth of a new star, then maybe we can use that same approach to look through billions of cells and try to find that one specific one we are looking for.”
He likened the process to finding “a needle hidden within a thousand haystacks,” and noted that STAR is able to do this within a couple of hours and gently enough to preserve sperm for use in IVF.
What makes STAR distinct, he added, is that it not only detects the presence of sperm, but also isolates them automatically—a step that sets it apart from many other AI diagnostic tools.
The system can scan around eight million images in an hour.
Williams recalled the moment he became convinced of the system’s potential: before discarding semen samples deemed sperm-free by embryologists, they were run through STAR.
In one case, after two days of unsuccessful manual searching, STAR found 44 sperm in one hour.
Rosie, 38, who asked to use a pseudonym to protect her identity, and her husband became the first couple to achieve pregnancy using STAR.
They had spent nearly two decades trying to conceive and had undergone 15 unsuccessful IVF cycles. Their Orthodox Jewish faith, Rosie said, kept them hopeful throughout.
Before using STAR, they had explored multiple options to address her husband’s azoospermia, including surgery and bringing in a specialist from abroad to manually search sperm samples.
They also looked into chemical extraction methods, which posed risks to sperm quality.
Rosie said: “There really was nothing else out there.
“Especially because I am running quite a few years ahead of where we should be [for fertility]. I’m not that old, but in fertility years—egg-wise—I was reaching my end.”
They learned about Williams’ programme through a community group and quickly familiarised themselves with the technology.
She said: “We knew exactly what it was, and knew what they were trying to do.
“If they could get sperm in a more natural way without chemicals and hopefully chose the good ones—if the programme was able to do that, we knew we had a better chance.”
The IVF cycle using STAR did not require any extra testing or procedures and followed the same steps as previous attempts.
Rosie said: “We were keeping our hopes to a minimum after so many disappointments.
“We came in, did what we had to do for the cycle, knowing there was probably a very small chance of anything happening. Why should this be any different from every other time?”
Williams explained that in conventional IVF, sperm typically outnumber eggs by a large margin, but in azoospermia cases, the reverse is true.
To maximise the chances of success, his team used STAR to collect several sperm samples in advance, which were frozen.
On the day of egg retrieval, they processed a fresh semen sample through STAR and used any recovered sperm to fertilise the eggs.
The frozen samples were kept in reserve in case no viable sperm were found in the fresh sample.
Within two hours, Rosie was told her eggs had fertilised successfully.
She said: “After the transfer, it took me two days to believe I was actually pregnant.”
Now four months into her pregnancy, Rosie is receiving standard obstetric care, and doctors say everything is progressing normally.
She said: “I still wake up in the morning and can’t believe if this is true or not.
“And I still don’t believe [I’m pregnant] until I see the scans.”
Williams said that azoospermia is just one fertility challenge where AI could be transformative.
Williams said: “There are things going on that we are blind to right now. But with the introduction of AI, we are being shown what those things are.
“The dream is to develop technologies so that those who are told ‘you have no chance of being able to have a child’ can now go on to have healthy children.”
AI may help speed breast cancer diagnosis for high-risk women after abnormal mammograms, a study suggests.
Women with abnormal mammograms often wait weeks to learn whether they have breast cancer.
Researchers at UC San Francisco and UC Berkeley said an AI-guided workflow could help reduce that wait by quickly identifying those most likely to have the disease. Some women could move from imaging to evaluation, and sometimes biopsy, in a single day.
Dr Maggie Chung, first author of the study, said: “This is a really an exciting time.
“This moves us closer to personalised care, where we can tailor a plan so that each patient gets the right intervention at the right time.”
The study used an open-source AI model called Mirai.
The model was trained on hundreds of thousands of mammograms linked to patients’ cancer outcomes.
A mammogram is an X-ray scan of the breast used to look for signs of cancer. A biopsy involves taking a small tissue sample to test for disease.
The AI tool is designed to detect subtle patterns in screening mammograms and predict a woman’s cancer risk.
Researchers at UC San Francisco and UC Berkeley applied the model to more than 4,100 screening mammograms at Zuckerberg San Francisco General Hospital and Trauma Center.
Mirai identified 525 women, about 12.7 per cent of screened patients, as high risk.
Those patients could receive an interpretation of their mammograms immediately after the scan and have additional diagnostic imaging for suspicious areas on the same day.
Some women who needed biopsies were also able to have them on the same day.
The researchers said Mirai reduced the wait time for diagnostic evaluation from several weeks to about an hour.
For women who were ultimately diagnosed with breast cancer, it reduced the average wait for biopsy from more than two months to fewer than 10 days.
The researchers stressed that Mirai does not replace radiologists or make diagnoses on its own.
Instead, it acts as a triage tool to help physicians identify the patients who can benefit most from accelerated care.
The team analysed more than 114,000 archival mammograms before launching the programme, to ensure the model would capture enough high-risk patients without overloading the clinic with too many expedited evaluations.
The researchers said they hope AI will support a more personalised approach to breast cancer screening tailored to each patient’s breast cancer risk.
Chung said: “Right now, many women follow the same screening schedule but their individual risk can be very different.
“AI risk assessment gives us the chance to identify the women most likely to benefit from expedited care and get them what they need.”
Adam Yala, senior author of the study and a data scientist at UC Berkeley, said: “This is a powerful example of how AI can be a collaborative partner for physicians.
“It shows how we can improve care when we bring clinicians and data scientists together to design these systems.”
Type 2 diabetes diagnoses are rising twice as fast in women under 40 as in women over 40, new data shows.
Type 2 diabetes is a serious condition and can lead to complications such as heart attacks and strokes. When it develops in younger people, it can be more aggressive and have more severe and acute effects.
Diagnoses in women under 40 rose by 47 per cent between 2017/18 and 2023/24. By comparison, diagnoses rose by 22 per cent in women aged 40 to 79.
During the same period, type 2 diabetes diagnoses in men under 40 increased by 34 per cent.
Diabetes UK said it is concerned about the follow-up care offered to women who have had gestational diabetes, also known as GDM, which increases the risk of developing type 2 diabetes after pregnancy.
Gestational diabetes is high blood sugar that develops during pregnancy and usually goes away after birth, but it raises the risk of type 2 diabetes later.
Colette Marshall, chief executive at Diabetes UK, said: “These figures should be a wake-up call. Type 2 diabetes is rising twice as fast in younger women compared to older women, and a crucial opportunity for prevention is being missed. Every diagnosis is life-changing, but when it develops in younger people, type 2 diabetes is even more aggressive.
“Pregnancy shouldn’t be a pathway to ill health. Yet despite facing a much higher risk of type 2 diabetes, too many women with GDM receive little or no follow-up care after pregnancy.
“As the Government turns its Strategy into action, support for women who have had gestational diabetes must not be overlooked.”
Last year, the NHS published the first national GDM audit for England in 2024/25, which revealed inconsistencies in follow-up care.
Only 57 per cent of women with GDM received an annual HbA1c test, which should be offered to every woman with GDM.
An HbA1c test measures average blood sugar levels over the previous two to three months.
Only 4.5 per cent of women had received support through the NHS Diabetes Prevention Programme.
The report also found that 11 per cent of women developed prediabetes within five years of having GDM, while 15 per cent developed type 2 diabetes within 10 years.
Prediabetes means blood sugar levels are higher than normal and a person has a higher risk of developing type 2 diabetes.
A recent survey funded by Diabetes UK also found that more than a third of women with GDM felt abandoned by healthcare services after giving birth.
If you live in England and have had gestational diabetes, you can self-refer to the NHS Diabetes Prevention Programme, which supports people at risk of developing type 2 diabetes. If you live in Northern Ireland, Scotland or Wales, you can speak to your GP about support.
Diabetes UK has written to women’s health minister Baroness Merron calling for urgent improvements to postnatal support for those diagnosed with GDM during pregnancy.
GDM affects between 10 and 20 per cent of pregnant women, but Diabetes UK said cases have long been underreported and UK-wide data on the condition has not been readily available.
The charity said poor follow-up care for women who have had GDM may be contributing to rising rates of type 2 diabetes in younger women.
It is calling for consistent postnatal follow-ups for women after GDM, more referrals to the NHS Diabetes Prevention Programme, greater accountability for improvements in postnatal care, and action on inequalities affecting women from deprived and minority ethnic communities.
A genomic test may help some women with breast cancer avoid chemotherapy, with near-identical outcomes in an international trial.
The findings suggest patients with a low test score could be treated with hormone therapy alone without increasing the risk of their cancer returning.
Researchers said the results could support more personalised treatment decisions and spare some women the side-effects of chemotherapy.
Prof Rob Stein, the trial’s chief investigator and a professor of breast oncology at UCL, said: “Optima addresses a longstanding challenge in breast cancer care: identifying who truly benefits from chemotherapy and who does not.
“Our findings show that many patients can safely avoid chemotherapy without compromising their outcomes.
“These results mark an important and significant step toward more personalised treatment.
“The trial has successfully used tumour biology to guide decisions rather than relying solely on traditional clinical features.”
Breast cancer treatment usually involves surgery to remove tumours. Chemotherapy is then often recommended if doctors believe there is a risk the disease will return.
Chemotherapy can cause side-effects including hair loss, rashes, nausea, insomnia and fatigue. Some women may also face longer-term consequences such as infertility, cognitive impairment or early menopause.
The Optima trial followed more than 4,000 patients with newly diagnosed breast cancer in the UK, Norway, Sweden, Australia, New Zealand and Thailand.
The trial was led by University College London.
One woman who took part in the trial told the Guardian that being able to skip chemotherapy felt “like Christmas”. Nine years after being diagnosed, taking the test and skipping chemotherapy, she is healthy and enjoying a full and active life.
The trial tested whether a genomic test could identify which patients need chemotherapy and which could safely avoid it.
The Prosigna test, made by diagnostics company Veracyte, analyses the activity of 50 genes in tumour tissue. It identifies the molecular subtype of the cancer and gives a score estimating the risk of breast cancer returning in the next 10 years.
The randomised trial involved 4,429 patients aged 40 or over with hormone-positive breast cancer. Hormone-positive breast cancer grows in response to hormones such as oestrogen or progesterone. It is the most common form of breast cancer, accounting for up to 80 per cent of cases globally.
Participants were assigned to one of two groups. In the standard treatment group, patients received chemotherapy followed by hormone therapy.
In the second group, patients had their tumours analysed using the genomic test. Those with a high score received chemotherapy and hormone therapy. Those with a low score received hormone therapy alone.
Radiotherapy and other treatments were given as usual in both groups.
In the second group, outcomes were very similar whether chemotherapy was given or not. Five years after treatment, 95 per cent of patients who had chemotherapy and hormone therapy were alive and free from breast cancer recurrence, while 94 per cent of those who skipped chemotherapy were also alive and recurrence-free.
The findings suggest chemotherapy offered little or no additional benefit for patients with low test scores.
Some men also took part in the study, but researchers said there were too few to draw firm conclusions for this group.
The trial received funding from the National Institute for Health and Care Research, Veracyte and cancer charities.
Prof Iain MacPherson, a co-chief investigator and professor of breast oncology at the University of Glasgow, said: “Optima provides robust, practice-changing evidence that we can safely reduce the use of chemotherapy for many patients with hormone-sensitive breast cancer.
“These findings represent a major step forward in delivering more personalised, precise care, ensuring that treatment decisions are driven by what will genuinely improve outcomes for patients, while avoiding unnecessary toxicity.
“The potential impact for both patients and health services is substantial.”
Menopause1 week agoPerimenopause misinformation ‘putting women at risk’
Insight4 weeks agoNIH Grant terminations disproportionately impact minority scientists, research finds
Adolescent health4 weeks agoWUKA brings Period-Positive Pool Party to London Aquatics Centre to keep girls swimming through puberty
Insight3 weeks agoPCOS renamed after decade-long campaign to end ‘cyst’ misconception
Events4 weeks agoWHIS 2026 unveils agenda and first speakers for the leading women’s health summit
Menopause4 weeks agoCBT shows promise for menopause insomnia and hot flashes
Hormonal health2 weeks agoNHS urged to update website following renaming of PCOS
News6 days agoThree menopause innovators shortlisted for Femtech World Award