NHS AI Blood Test Could Spare Thousands of Women from Invasive Womb Cancer Checks

Introduction

The National Health Service (NHS) in England is preparing to roll out an artificial intelligence-powered blood test designed to help doctors assess women referred for suspected womb cancer, potentially reducing the number of invasive diagnostic procedures patients have to undergo. The test, developed by the Leeds-based company PinPoint Data Science, uses machine learning to analyze blood markers and generate a cancer risk score, giving clinicians another tool to decide who needs urgent investigation and who can be safely monitored.

This development marks another step in the NHS’s growing use of AI across cancer diagnostics, following similar deployments in lung cancer screening, infection risk detection, and app-based patient triage. For the tens of thousands of women referred each year over concerns about possible womb cancer, this AI blood test could mean fewer uncomfortable procedures, faster answers, and more efficient use of NHS resources.

The Scale of the Problem

Every year, around 90,000 postmenopausal women in England are referred by their GPs for further investigation after experiencing postmenopausal bleeding, a key warning sign of womb (endometrial) cancer. Of these referrals, roughly 10,000 women are ultimately diagnosed with womb cancer, and sadly, about 2,700 women die from the disease annually.

These numbers highlight a persistent challenge in cancer diagnostics: while it’s essential to investigate every case of concerning symptoms, the vast majority of women referred through these urgent pathways do not actually have cancer. This means thousands of women each year go through invasive, uncomfortable, and sometimes distressing procedures that ultimately rule out disease rather than confirm it. Streamlining this process without compromising patient safety has long been a goal for NHS clinicians and researchers, and this is precisely where the new AI blood test aims to make a difference.

How the PinPoint AI Blood Test Works

The PinPoint test uses machine learning algorithms to interpret patterns across approximately 30 different blood markers. Rather than looking at a single biomarker in isolation, the AI model considers the combined signal across all these markers to produce a risk classification: low, elevated, or high risk of cancer.

This risk score is designed to slot into existing NHS cancer referral pathways, giving GPs and specialists additional information to help decide the next steps for a patient. According to PinPoint, the test costs around £30 to administer, a relatively modest cost when weighed against the expense, discomfort, and anxiety associated with more invasive procedures.

Importantly, PinPoint positions this as a multi-cancer test rather than one narrowly focused on gynaecological cancers alone. The company reports that its technology has already been applied across several other cancer pathways, including lung cancer, upper gastrointestinal cancer, head and neck cancer, and lower gastrointestinal cancer. This broader application suggests the underlying AI model may have long-term potential well beyond womb cancer detection, potentially supporting a more integrated, blood-based approach to early cancer triage across multiple disease areas.

Trial Results: A Closer Look at the Data

Before being introduced into frontline NHS use, the PinPoint test was evaluated in a large-scale trial involving 16,481 patients who had been referred through urgent suspected cancer pathways across the Yorkshire region. This included women who had been referred specifically due to symptoms raising concern about womb or other gynaecological cancers.

The trial results are striking. Roughly one in ten women referred for heavy or postmenopausal bleeding were ultimately found to have cancer. Against this backdrop, the PinPoint test demonstrated strong diagnostic performance:

• The test correctly identified 99.1% of actual cancer cases as either elevated or high risk, meaning very few cancers were missed.
• For women placed in the lowest-risk category, the test achieved a negative predictive value of 99.8%, meaning that when the test indicated a very low risk, it was correct in ruling out cancer nearly all of the time.

These figures suggest that the test could reliably identify the women who are highly unlikely to have cancer, allowing them to potentially skip more invasive procedures altogether, while still ensuring that women at genuine risk are flagged for prompt further investigation.

Where the Test Is Being Introduced

Based on these results, two major NHS trusts are moving forward with implementation. Mid Yorkshire NHS Teaching Trust plans to use the AI blood test across six different types of gynaecological and upper gastrointestinal cancers, while Leeds Teaching Hospitals NHS Trust intends to apply the test specifically within its gynaecological cancer pathway.

This phased, region-specific rollout is a common approach for new NHS diagnostic technologies, allowing trusts to build real-world experience and further evidence before wider adoption is considered across other parts of the country.

The Current Diagnostic Pathway and Its Drawbacks

To understand why this AI test matters, it helps to look at what women currently go through when referred for suspected womb cancer. Under the standard pathway, women are typically given a pelvic examination that includes a transvaginal ultrasound scan. This procedure involves inserting an ultrasound probe into the vagina to measure the thickness of the womb lining. While clinically valuable, many women find this scan uncomfortable, embarrassing, or even painful.

If the ultrasound results still raise concern, patients may then be referred for additional, more invasive procedures, including a biopsy and a hysteroscopy, a direct examination of the inside of the womb using a small camera. Each additional step adds time, cost, physical discomfort, and emotional stress for patients who, in the majority of cases, will ultimately be found not to have cancer.

PinPoint’s stated goal is to intervene earlier in this pathway, using the blood test to identify women at very low risk before they are subjected to the transvaginal ultrasound scan and subsequent invasive checks. According to the company, this approach could spare around one in five referred women from needing a transvaginal ultrasound at all, which would equate to roughly 18,000 women across England each year avoiding an uncomfortable procedure they never actually needed.

What Clinicians Are Saying

Several senior clinicians involved in cancer care have spoken about the potential benefits of the new test. Professor Sean Duffy, chief medical officer at PinPoint Data Science and a former NHS England national clinical director for cancer, has emphasized that the primary value of the test lies in its ability to confidently rule out cancer in women at very low risk, freeing up clinical capacity for those who need it most.

Dr. Jacinta Walsh, a GP based in Normanton, West Yorkshire, has pointed out that under the current system, some patients can require up to six separate GP visits before cancer is definitively ruled out. She suggested that a reliable blood test could significantly shorten this process, reducing the burden on both patients and primary care services.

Meanwhile, Tracy Jackson, a consultant gynaecologist and cancer unit lead at Leeds Teaching Hospitals NHS Trust, noted that the majority of women who go through the current referral pathway do not have cancer, even though the investigations involved can be uncomfortable and distressing. She sees the AI blood test as a valuable triage tool, allowing low-risk patients to be reassured and managed within primary care, while higher-risk patients can be fast-tracked for the specialist investigations they need.

Part of a Broader Wave of NHS AI Adoption

The PinPoint blood test is not an isolated initiative. It fits into a broader pattern of the NHS integrating artificial intelligence tools across multiple areas of patient care and diagnostics.

At Kent and Canterbury Hospital, East Kent Hospitals University NHS Foundation Trust has deployed an AI system called MEMORI, which analyzes routine patient data, including blood tests, blood pressure readings, temperature, other clinical observations, medications, and demographic information, to assess a patient’s risk of infection.

Separately, NHS England has introduced an AI-powered triage tool within the NHS App. This tool is projected to reach more than 200,000 patients within its first year of operation, with plans to make it available to all NHS App users by April 2028.

In the field of lung cancer detection, the UK government has committed £20 million toward expanding AI-powered chest X-ray analysis tools to every NHS trust in England by 2029. These tools are already operational in roughly half of NHS trusts and have reportedly supported the assessment of more than four million patients being investigated for possible lung cancer.

Together, these initiatives paint a picture of an NHS system increasingly willing to integrate AI-driven decision support into everyday clinical workflows, from initial patient triage through to diagnostic interpretation.

Caution and the Need for Further Evidence

Despite the promising trial data, experts are urging measured optimism rather than unqualified enthusiasm. Cancer Research UK has described the PinPoint test as a promising development but stressed that more research is needed to fully understand its real-world benefits for both patients and the wider NHS system.

Samantha Harrison, a spokesperson for the charity, has highlighted that while early detection saves lives, current diagnostic pathways are not always fast enough to catch cancer at the earliest, most treatable stages. She noted that a blood test capable of ruling out endometrial cancer for certain women, without requiring further invasive investigation, could represent a meaningful improvement in how quickly and comfortably patients receive accurate answers.

Further evaluation will be needed to understand exactly how the test performs at scale across more diverse patient populations, how it affects long-term patient outcomes, how it influences referral decision-making across GP practices, and what its broader impact will be on NHS diagnostic capacity and waiting times.

What This Means for Patients

For the many thousands of women referred each year for suspected womb cancer, this AI blood test offers the prospect of a less invasive, faster, and more reassuring diagnostic journey. Rather than immediately undergoing an uncomfortable transvaginal ultrasound scan, eligible women could first receive a simple blood test that helps determine whether further investigation is truly necessary.

For women at high risk, this system is designed to expedite their pathway to specialist care rather than delay it, ensuring that those with concerning symptoms are prioritized for the detailed investigations they need. For low-risk women, it may mean avoiding unnecessary discomfort, anxiety, and repeated hospital visits altogether.

The introduction of PinPoint’s AI-powered blood test into NHS cancer pathways reflects a broader shift toward smarter, data-driven triage in modern healthcare. By harnessing machine learning to interpret dozens of blood markers simultaneously, the test offers a low-cost, minimally invasive way to help distinguish between women who urgently need further investigation and those who can be safely reassured.

While the early trial results, showing strong sensitivity for detecting cancer and high accuracy in ruling it out for low-risk patients, are encouraging, healthcare experts rightly caution that continued research and real-world monitoring will be essential. As NHS trusts in Yorkshire begin using the test in gynaecological and related cancer pathways, the coming months and years will reveal how significant an impact this technology can have on reducing unnecessary invasive procedures, easing pressure on NHS diagnostic services, and ultimately improving the experience and outcomes for the tens of thousands of women referred for suspected womb cancer each year.

As AI continues to make inroads into NHS diagnostics, from blood-based cancer risk scoring to AI-assisted chest X-ray analysis and app-based patient triage, tools like the PinPoint test may represent an early glimpse of a future where cancer detection is faster, less invasive, and more precisely targeted to those who need it most.

Frequently Asked Questions

1. What is the NHS AI blood test for womb cancer?

It’s a machine learning-based blood test developed by PinPoint Data Science that analyzes around 30 blood markers to assess a patient’s risk of womb (endometrial) cancer, classifying them as low, elevated, or high risk.

2. How does the PinPoint AI test work?

The test looks at patterns across roughly 30 blood markers using a machine learning model, generating a risk score that clinicians can use alongside existing NHS cancer referral pathways to help decide next steps.

3. How accurate is the PinPoint blood test?

In a trial of 16,481 patients across Yorkshire, the test correctly flagged 99.1% of actual cancer cases as elevated or high risk, and achieved a 99.8% negative predictive value for patients placed in the lowest-risk group.

4. Which NHS trusts are using this AI blood test?

Mid Yorkshire NHS Teaching Trust plans to use it across six gynaecological and upper gastrointestinal cancer types, while Leeds Teaching Hospitals NHS Trust plans to apply it within its gynaecological cancer pathway.

5. Will this test replace transvaginal ultrasound scans?

Not entirely. It’s designed to be used before an ultrasound, helping identify very low-risk women who may be able to skip the scan. PinPoint estimates this could spare around 18,000 women in England annually.

6. How much does the test cost?

PinPoint has stated the test costs around £30 per patient, a relatively low cost compared to the resources required for invasive follow-up procedures.

7. Who is eligible for this test?

It’s aimed at women referred by GPs for suspected womb cancer, typically postmenopausal women experiencing symptoms like abnormal or heavy bleeding. Around 90,000 women a year in England fall into this referral group.

8. Is this the only AI tool the NHS is using for cancer diagnosis?

No. The NHS is also using AI for lung cancer detection via chest X-ray analysis, infection risk assessment through a system called MEMORI, and an AI triage tool built into the NHS App.

9. What do experts say about the test’s reliability?

Clinicians involved in the trial, including gynaecologists and GPs, describe it as a promising triage tool, though Cancer Research UK notes that more research is needed to confirm its real-world impact on patient outcomes and NHS capacity.

10. When will this test be available more widely across the NHS?

Currently, it’s being introduced at NHS trusts in Yorkshire following successful trial results. Broader rollout across England would depend on further evidence gathered from these initial deployments.