Deep Learning Can Help Guide Lung Cancer Treatment Decisions

A deep learning tool was able to accurately predict survival expectancy in patients with lung cancer, potentially leading to more informed care decisions by providers, according to a study published in the International Journal of Medical Informatics.

The study showed that in certain conditions, the deep learning model was more than 71 percent accurate in predicting survival expectancy of lung cancer patients, compared to other machine learning models that performed with about 61 percent accuracy.

The tool can analyze a large amount of data that describe the patients and the disease to understand how a combination of factors affect lung cancer survival periods. These factors include information like types of cancer, size of tumors, speed of tumor growth, and demographic data.

“This is a high-performance system that is highly accurate and is aimed at helping doctors make these important decisions about providing care to their patients,” said Youakim Badr, associate professor of data analytics at Penn State Great Valley. “Of course, this tool can’t be used as a substitute for a doctor in making decisions on lung cancer treatments.”

Information on a patient’s survival expectancy could help guide providers and caregivers make improved decisions on using medicines, allocating resources, and determining the intensity of care for patients.

Researchers noted that deep learning techniques are uniquely suited to address lung cancer prognosis because the technology can provide the comprehensive analysis needed in cancer research.

In deep learning, developers apply a sophisticated structure of multiple layers of artificial neurons. The learning aspect of deep learning comes from how the system learns from connections between data and labels, the team noted.

“Deep learning is a machine-learning algorithm that makes associations between the data, itself, and the labels that we use to describe the data examples," said Badr. “By making these associations, it learns from the data.”

The structure of deep learning also offers several advantages for many data science tasks, especially in cases involving large datasets.

“It improves performance tremendously. In deep learning we can go deeper, which is why they call it that. In traditional machine learning, you have a simple structure of layers of neural networks. In each layer, you have a group of cells,” said Robin G. Qiu, professor of information science and engineering and an affiliate of the Institute for Computational and Data Sciences.

“In deep learning, there are many layers of these cells that can be architected into a sophisticated structure to perform better feature transformation and extraction, which gives you the ability to further improve the accuracy of any model.”

Researchers analyzed data from the Surveillance, Epidemiology, and End Results (SEER) program. The SEER dataset is one of the biggest and most comprehensive databases on the early diagnosis information for cancer patients in the US. The program’s cancer registries cover almost 35 percent of US cancer patients.

“One of the really good things about this data is that it covers a large section of the population and it’s really diverse,” said Shreyesh Doppalapudi, a graduate-student research assistant and first author of the paper.

“Another good thing is that it covers a lot of different features, which you can use for many different purposes. This becomes very valuable, especially when using machine learning approaches.”

When the team compared several deep learning approaches to traditional machine learning models, the deep learning approaches performed much better than traditional machine learning methods.

The deep learning technique enabled researchers to find associations in the SEER dataset, which includes about 800,000 to 900,000 entries – something that would have been incredibly challenging without the assistance of AI.

“If it were only three fields I would say it would be impossible — and we had about 150 fields,” said Doppalapudi. “Understanding all of those different fields and then reading and learning from that information, would be impossible.”

Going forward, researchers will aim to improve the model and test its ability to analyze other types of cancers and medical conditions.

“The accuracy rate is good, but it’s not perfect, so part of our future work is to improve the model,” said Qiu.

Researchers also plan to connect with domain experts on specific cancers and medical conditions.

“In a lot of cases, we might not know a lot of features that should go into the model,” said Qiu. “But, by collaborating with domain experts, they could help us collect important features about patients that we might not be aware of and that would further improve the model.”