A study presented at the European Respiratory Society’s International Congress has revealed that Artificial Intelligence (AI) could improve the interpretation of lung tests. This study is among the many in the recent years that have used AI to improve detection of diseases.
And, many of them are achieving near-human efficiency in detecting deadly diseases.
How are most of the lung tests conducted?
A series of tests are conducted to test the volume and speed of air during breathing. Spirometry is most commonly used lung test. A spirometer records the volume and speed at which the air is moving in and out of your lungs.
This is generally followed by a plethysmography test. This is used to measure the amount of air you can hold in your lungs. And, the last one is a diffusion test. As the name implies, the test measures how well the lungs process the air you breathe and allow the oxygen and carbon dioxide to pass in and out of your blood.
The study contends that most of the tests are conducted based on expert opinion and international guidelines.
‘Algorithm can do complex reasoning’
In this study, researchers included data from 968 people who were undergoing lung function testing for the first time. They then used machine learning techniques to understand and analyse lung functions.
Machine learning utilises algorithms that can learn from and perform predictive data analysis.
The formulated an algorithm in addition to the routine process. The algorithm factored in smoking history, body mass index, and age. Now, based on the patterns — of both the clinical and lung function data — the algorithm made suggestions for the most likely diagnosis.
Wim Janssens, the author of the study, said:
We have demonstrated that artificial intelligence can provide us with a more accurate diagnosis in this new study. The beauty of our development is that the algorithm can simulate the complex reasoning that a clinician uses to give their diagnosis, but in a more standardised and objective way so it removes any bias.
Another author Marko Topalovic said:
The benefit of this method is a more accurate and automated interpretation of pulmonary function tests, and thus better disease detection. Not only can this help non-experienced clinicians, but it also has many benefits for healthcare overall as it is time saving in achieving a final diagnosis as it could decrease the number of redundant additional tests clinicians are taking to confirm the diagnosis.