Quantum Computer Spots Lung Cancer With Over 90% Accuracy in Test

Scientists used a D-Wave quantum computer to detect lung cancer in test data with more than 90% accuracy, though results are still preliminary.

Scientists have taken an interesting step toward using futuristic computers in medicine. A team led by researchers at Florida Atlantic University (FAU) tested a quantum computer — a next-generation machine that uses the strange rules of physics to solve certain problems — and found it could spot lung cancer in test data with impressive accuracy.

What they did

The researchers used a machine called D-Wave and taught it to recognize patterns using machine learning (a type of software that "learns" from examples instead of following fixed rules). Their goal was to see how well quantum software performs when given very little data to learn from.

Here are the key facts:

They started with an open dataset of 309 patient records, covering details like smoking, coughing, wheezing, fatigue, and chest pain, plus whether the person had the disease.
To balance the numbers, they added artificial (synthetic) records, bringing the total to 540 examples.
In each test, the model trained on just 108 examples and was tested on the remaining 432.

The results

The team compared three quantum learning methods. The best one used a technique called quantum annealing (a method where the computer settles into the best possible answer, a bit like a ball rolling to the lowest point).

In a simulation (a software imitation of the quantum computer), accuracy reached 94.31%.
On the real quantum hardware, accuracy dropped to 91.9%.

The researchers blamed the difference on "noise" and current limits of quantum processors. Notably, when compared to ordinary (classical) computer methods, the real quantum device actually scored the lowest.

Why it's early days

The authors stressed these are preliminary results. The dataset had no medical images, lab tests, or cancer-stage information, and the model was not tested in real hospitals.

Next, the team plans to test it on independent clinical data, expand the dataset, and combine symptom data with image analysis.

Bottom line: Quantum computers show early promise for medical diagnosis, but they aren't ready to beat today's tools just yet.

This is an AI-generated summary. Read the original article at: https://forklog.com/news/uchenye-ispytali-kvantovuyu-model-dlya-vyyavleniya-raka-legkih