Swedish Scientists Are Developing A Cancer-Sniffing Robot With 97% Accuracy For Ovarian Cancer

Ovarian cancer, rather unfairly, is a master of disguise. Its early symptoms: bloating, abdominal discomfort, vague fatigue are so indistinct that they could just as easily be blamed on a heavy lunch or a stressful week. As a result, it is often discovered late, when treatment becomes more complicated and survival rates decline. In 2022 alone, roughly 325,000 new cases were diagnosed worldwide, and more than 200,000 people died from the disease. The numbers, according to global projections, are expected to rise sharply by 2050.

At Linköping University in Sweden, Associate Professor Donatella Puglisi and her colleagues have been working on a device that mimics the mammalian sense of smell. The aim is admirably straightforward: teach a machine to detect what the human body whispers long before it begins to shout. The project received funding from Swedish research bodies including Vinnova, Formas, the Swedish Energy Agency, and the Swedish Research Council, with computational support from the National Academic Infrastructure for Supercomputing in Sweden.

Ovarian cancer (Getty Images)

The Swedish team’s approach is disarmingly clever. Instead of hunting for a single elusive biomarker, they decided to examine the overall “smell profile” of blood plasma. Cancers release volatile organic compounds. Different cancers release different blends. In a sense, each has its own aromatic signature, though not one you would bottle and sell.

“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” says Donatella Puglisi.

The device itself contains 32 sensors, each responding to various volatile substances emitted from the sample. Electronic nose technology has, in fact, been around for about 60 years. Previously, it has been used in everything from food quality control to environmental monitoring. What has changed dramatically is the rise of machine learning.

The sensors gather a dense cloud of chemical information. Then an algorithm (trained on known samples from a biobank) analyzes patterns within that cloud. Instead of looking for one specific molecule, it looks at the orchestra. According to the researchers, the system can distinguish ovarian cancer not only from healthy samples but also from endometrial cancer. And it does so with a reported accuracy of 97%!

“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,” says Jens Eriksson, associate professor at LiU and CTO at VOC Diagnostics AB, the company developing the electronic nose.

The electronic nose doesn’t need to know what it is searching for in advance. It surveys the entire chemical landscape and lets the algorithm decide whether the pattern resembles known cases of disease. The test reportedly takes about 10 minutes and is relatively inexpensive.

This is still a pilot study, published in Advanced Intelligent Systems, and much work remains before such technology becomes a routine part of screening programmes, from clinical validation, regulatory approval to large-scale trials. But the researchers are hopeful that within three years, the method could become part of formal cancer screening.

Source:

https://advanced.onlinelibrary.wiley.com/doi/10.1002/aisy.202500838

Read more:

1. Are You At The Risk Of Ovarian Cancer? Know The Primary Factors
2. Everyday Things In Your House That Are Putting You At Risk Of Cancer
3. Centre To Roll Out HPV Vaccination Across The Country To Prevent Cervical Cancer
4. 3 Things Men Should Never Do Because It Increases Their Partner's Risk Of Cervical Cancer | 2026 Cancer Prevention Action Week