- Researchers are reporting that an artificial intelligence (AI) program can detect lung nodules earlier.
- In rare cases, the nodules can be a sign of lung cancer.
- Experts say AI is increasingly being used in the medical profession.
Artificial Intelligence, also known as AI, has found its way into many sectors of our lives.
AI, which essentially teaches machines to learn so they can make decisions that mimic human intelligence, is ubiquitous, even if you can’t always see it.
There are AI applications being used now in education, commerce, manufacturing, financial services, telecommunications, energy, aviation, and drones.
But perhaps the most significant and potentially game-changing uses for AI are in healthcare, specifically cancer.
AI is being used to assist with clinical decisions for cancer diagnosis and screening, processing medical data, and early cancer detection with deep-learning strategies.
A study published today in the journal Radiology looked at the effect AI-based software had in a real-world oncological clinical practice.
In it, researchers reported that AI “significantly” improved the detection of lung nodules on chest X-rays.
Lung nodules are abnormal growths that form on the lungs. They’re common and typically form from previous lung infections.
But in rare instances, they can be a sign of lung cancer.
One of the common screening methods used for identifying lung nodules is chest X-rays.
Dr. Jin Mo Goo, a study co-author and a professor at the Department of Radiology at Seoul National University Hospital in Korea, told Healthline that AI can be a powerful tool to help identify lung nodules, especially when radiologists are experiencing a high volume of cases.
“The detection of cancer at an earlier stage is a crucial issue. As many solid tumors could be identified in imaging studies, the detection of potential early cancers such as lung nodules in lung cancer, more effectively is the first step in improving the outcomes of cancer patients,” said Goo.
In the study, researchers included 10,476 people with an average age of 59 who had undergone chest X-rays at a health screening center between June 2020 and December 2021.
Participants completed a self-reported health questionnaire to identify baseline characteristics such as age, sex, smoking status, and past history of lung cancer.
About 11% of the participants were current or former smokers.
The study participants were randomly divided evenly into two groups — AI or non-AI.
The first group’s X-rays were analyzed by radiologists aided by AI while the second group’s X-rays were interpreted without the AI results.
Solid nodules with diameters either larger than 8 millimeters or subsolid nodules with a solid portion larger than 6 millimeters were identified as actionable, meaning that the nodule required follow-up under lung cancer screening criteria.
Lung nodules were identified in 2% of the participants. Analysis showed that the detection rate for actionable lung nodules on chest X-rays was higher when aided by AI (0.59%) than without AI assistance (0.25%).
While older age and a history of lung cancer or tuberculosis were associated with positive reports, these and other health characteristics did not have an impact on the efficacy of the AI system, the researchers reported.
This suggests that AI may work consistently across different populations, even for those with diseased or postoperative lungs, the scientists concluded.
Goo said the study provided strong evidence that AI can help in interpreting chest radiography and that this will contribute to identifying chest diseases, especially lung cancer, more effectively and at an earlier stage.
Imaging-based cancer screening is tedious work because the prevalence of cancer is typically low in the screening population.
“The value of computer-aided detection and diagnosis has been investigated to reduce missed cancers for decades,” Goo said.
“The recent introduction of deep-learning technology enhanced the performance of traditional machine-learning techniques not only in identifying lesions but also in measuring and characterizing lesions,” he added.
Thomas Swalla, the chief executive officer and board director at Dotmatics, a global company of 850 scientists and employees worldwide who are focused on deploying AI in multiple platforms, was not involved in this study.
He told Healthline the findings are just another example of what AI can do in the cancer sector.
“The impact of AI on cancer is twofold. By using AI, it costs less money to find new therapies and allows greater access to care,” said Swalla.
“The use of AI will drive down the cost of healthcare to consumers,” he added. “And that will lead to more discovery in cancer as well as rare diseases, which historically simply did not have a successful business model.”
Dr. Steven Quay is the founder of Atossa Therapeutics, a biopharmaceutical company developing novel therapeutics and delivery methods for breast cancer and other breast conditions.
Quay said a new era of AI in oncology is just beginning.
“We are seeing a point now where AI can partner with research to develop new cancer therapies as well as treatment modalities,” he told Healthline. “I think that soon AI will be leading the charge in cancer care.”
Quay said the new way to use AI in cancer is to let it play against itself.
“You supply the data set and let it work and it learns from that,” he said. “The process then goes beyond the human knowledge that it is creating. It works in ways humans do not predict and that is creativity. AI also has a major role now in cancer research at a baseline level.”