Artificial intelligence improves diagnosis of lung disease: study

Posted: 19 September 2018 | | No comments yet

Diagnosing lung disease using AI was found to be accurate in twice as many cases as the diagnosis of pulmonologists…

Artificial intelligence can significantly improve the diagnosis of lung disease, suggests a new study.

Artificial intelligence (AI) can improve the diagnosis of lung disease by helping doctors interpret respiratory symptoms more accurately, according to new research.

An AI computer algorithm using high quality data proved more consistent and accurate in interpreting respiratory test results and suggesting diagnoses than lung specialists, revealed recent research presented at the European Respiratory Society International Congress in Paris, France.

“Pulmonary function tests provide an extensive series of numerical outputs and their patterns can be hard for the human eye to perceive and recognise; however, it is easy for computers to manage large quantities of data like these and so we thought AI could be useful for pulmonologists,” said Dr Marko Topalovic, a postdoctoral researcher at the Laboratory for Respiratory Diseases, Catholic University of Leuven, Belgium.

The study included 120 pulmonologists from 16 hospitals and researchers used historical data from 1,430 patients from 33 Belgian hospitals. The data were assessed by the pulmonologists and interpretations were measured against gold standard guidelines from the European Respiratory Society and the American Thoracic Society. The expert panel considered patients’ medical histories, results of all Pulmonary function tests and any additional tests, before agreeing on the correct interpretation and diagnosis for each patient.

‘More accurate in twice as many cases’

“We found that diagnosis by AI was more accurate in twice as many cases as diagnosis by pulmonologists,” said Dr Topalovic. “These results show how AI can serve as a second opinion for pulmonologists when they are assessing and diagnosing their patients.”

The use of good quality data for training the AI algorithm was of key importance, he added. An expert panel examined all the results from the pulmonary function tests, and the other tests plus medical information. They used these to reach agreement on final diagnoses before developing an algorithm to train the AI. Next, this was validated by incorporating it into real clinical practice at the University Hospital Leuven. The key challenge was to ensure the algorithm recognised patterns of up to nine different diseases.

AI-based software

Dr Topalovic said: “The interpretation of pulmonary function tests and the diagnosis of respiratory disease by pulmonologists is not an easy task. It takes more information and further tests to reach a satisfactory level of accuracy. On the other hand, the AI-based software has superior performance and therefore can provide a powerful decision support tool to improve current clinical practice. Feedback from doctors is very positive, particularly as it helps them to identify difficult patterns of rare diseases.”

Two Belgian hospitals are already using the AI-based software to improve interpretations and diagnoses. Dr Topalovic said the technology can empower doctors to make their interpretations and diagnoses easier, faster and better. But AI will not replace doctors because they are able to see a broader perspective than that presented by pulmonary function tests alone. “However, it is evident that AI will augment our abilities to accomplish more and decrease chances for errors and redundant work,” he added. “The AI-based software has superior performance and therefore may provide a powerful decision support tool to improve current clinical practice.”

The next step was said to be to persuade more hospitals to use the technology and investigate transferring the AI technology to primary care, where the data would be captured by general practitioners (GPs) to help them make correct diagnoses and referrals.