Researchers in Sweden have trained an AI-guided electronic nose to detect ovarian cancer from a simple blood sample. The fast, low-cost test could one day help catch deadly cancers much earlier.
An artificial “nose” guided by artificial intelligence can detect early signs of ovarian cancer from a simple blood sample, offering a potential new way to screen for one of the deadliest gynecologic cancers.
Researchers at Linköping University in Sweden report that their machine learning–driven device can distinguish ovarian cancer from both endometrial cancer and healthy controls by analyzing the mix of volatile substances rising from blood plasma. In tests on stored samples, the tool reached 97% accuracy.
The work, published in the journal Advanced Intelligent Systems, builds on decades-old “electronic nose” technology but combines it with modern AI to tackle a major gap in women’s health: the lack of a reliable, routine screening test for ovarian cancer.
Ovarian cancer is often called a silent killer because its early symptoms — bloating, abdominal discomfort, changes in appetite — are vague and easily mistaken for more common conditions. As a result, many patients are diagnosed only after the disease has spread, when treatment is harder and survival rates are much lower.
Globally, hundreds of thousands of people are diagnosed with ovarian cancer each year, and more than 200,000 die from the disease. International cancer organizations expect those numbers to rise sharply by 2050 as populations grow and age.
The team set out to recreate one of biology’s most powerful sensing systems, according to co-corresponding author Donatella Puglisi, an associate professor at Linköping University.
“We’re trying to mimic the mammalian sense of smell artificially. We’ve now developed an algorithm that can distinguish ovarian cancer from endometrial cancer and healthy control groups, using data from an electronic nose,” Puglisi said in a news release.
Electronic noses have been around for roughly 60 years. The prototype used in this study has 32 commercially available sensors that respond to different volatile compounds — tiny molecules that evaporate from a liquid or solid and can be detected as odors.
Each type of cancer releases its own pattern of volatile substances, essentially giving it a distinct chemical “scent.” Instead of hunting for a single biomarker, such as a specific protein in the blood, the LiU device reads the overall pattern from all 32 sensors at once.
Those raw sensor readings are then fed into advanced machine learning models. By training the models on known samples from a biobank — some from people with ovarian cancer, some with endometrial cancer, and some from healthy individuals — the system learns to recognize the complex patterns that signal disease.
The result is a test that does not depend on identifying one particular molecule and that can handle the messy reality of human biology, where many factors can influence any single biomarker.
Today’s blood-based cancer screening tests typically look for one or a few biomarkers tied to a specific cancer. Those tests can be slow to analyze and often lack the sensitivity and precision needed to reliably catch ovarian cancer early.
Co-corresponding author Jens Eriksson, an associate professor at Linköping University and the chief technology officer at VOC Diagnostics AB, the company developing the electronic nose, emphasizes that is a major limitation of current care.
“Unlike in breast cancer, there is currently no reliable ovarian cancer screening method. These tests are often based on a single biomarker and lack the precision required to detect the disease at an early stage. Our method is therefore far ahead not only in terms of accuracy but also in the ability to identify early disease,” Eriksson said in the news release.
The new approach is designed to be fast and practical. According to the team, the electronic nose test takes about 10 minutes from sample to result.
“It’s a simple test that takes 10 minutes and gives a clear result. Our method can test many people at a low cost and is much more accurate than what’s on the market today. This study is a pilot, but we hope it will be used as part of cancer screening within three years. Right now, we’ve focused on detecting cancer, but the applications are endless,” Eriksson added.
If that vision is realized, the technology could make screening more accessible in both high- and low-resource settings. The sensors themselves are relatively simple and already available on the market, which could help keep costs down compared with more complex imaging or lab-based tests.
Puglisi emphasized the broader public health stakes as cancer diagnoses rise worldwide.
“More and more people are being diagnosed with cancer, especially young adults, and this is alarming. If screening were more accessible, both in terms of cost and location, it would be possible to improve early diagnosis. Our approach could facilitate the adoption of new screening protocols and the development of new diagnostic methods, improving survival rates, quality of life, and overall clinical outcomes,” she said.
While this study is still a pilot and larger clinical trials will be needed to confirm the results in real-world settings, the concept points to a future where AI-enhanced chemical sensing could help detect many different cancers from a simple blood draw.
Beyond ovarian cancer, the same kind of electronic nose could potentially be trained to recognize the volatile signatures of other diseases, from additional cancer types to metabolic or infectious conditions. Because the system is “biomarker-agnostic,” researchers can focus on teaching it to recognize patterns rather than first identifying every molecule involved.
For now, the LiU team is working to refine the technology, validate it in bigger and more diverse patient groups, and navigate the regulatory steps required before any screening test can be used in clinics.
If successful, an AI-powered nose that can “smell” cancer from blood could become a powerful new tool in early detection — and a reminder that sometimes, the oldest senses can inspire the newest science.
Source: Linköping University