Lung cancer screening with low-dose computed tomography (LDCT) has been shown to significantly reduce mortality in high-risk patients.1 However, LDCT is not without its drawbacks: exposure to radiation, overdiagnosis, and false positives that lead patients to undergo invasive procedures such as biopsies and surgeries.2

Deep learning, a subset of artificial intelligence (AI), may help mitigate these risks, according to a study in Nature.3

In deep learning systems, human-made algorithms are responsible for sorting through and learning from data. Large quantities of data are fed into an artificial neural network – a set of algorithms modeled after the human brain – and then interpreted. Programmers don’t tell the machine what to do with the data; rather, the machine analyzes the data and makes decisions based on what it has learned. In most cases, the more data it receives, the more accurate its decisions will be.4

Continue Reading

Researchers at Google developed an artificial neural network with several layers of processing. They fed the network with CT scans from patients: some who had lung cancer, some who did not, and some who had nodules that later developed into malignancies.3

The technology showed a remarkable ability to detect lung cancer. When pitted against 6 radiologists, the system outperformed all 6 with an 11% reduction in false positives and a 5% reduction in false negatives. The algorithm performed on par with the radiologists when they were allowed to view prior scans. Overall, the model was 94.4% accurate.

Related Articles

Although Google’s deep learning system is not ready for widespread use, the results offer a glimpse into AI’s potential role in medicine. “This creates an opportunity to optimize the screening process via computer assistance and automation,” the authors wrote. “While the vast majority of patients remain unscreened, we show the potential for deep learning models to increase the accuracy, consistency, and adoption of lung cancer screening worldwide.”

References

  1. Ostrowski M, Marjanski T, Rzyman W. Low-dose computed tomography screening reduces lung cancer mortality. Adv Med Sci. 2018;63(2):230-236.
  2. Pinsky PF. Assessing the benefits and harms of low-dose computed tomography screening for lung cancer. Lung Cancer Manag. 2014;3(6):491-498.
  3. Ardila D, Kiraly AP, Bharadwaj S, et al. End-to-end lung cancer screening with three-dimensional deep learning on low-dose chest computed tomography (published online May 20, 2019]. Nat Med. 2019. doi:10.1038/s41591-019-0447-x
  4. Marr B. What is deep learning AI? A simple guide with 8 practical examples. Forbes. October 1, 2018. Accessed June 5, 2019.

Topics:

Lung Cancer