AI Tools Support Lung Cancer Diagnosis

AI Tools Support Lung Cancer Diagnosis

Lung cancer is the leading cause of cancer death in both men and women. The American Cancer Society estimates 2022 lung cancer cases and deaths in the United States will be:

  • Approximately 236,740 new cases of lung cancer
    • 117,910 in men and 118,830 in women
  • About 130,180 deaths from lung cancer
    • 68,820 in men and 61,360 in women

Survival in patients with lung cancer varies depending on the stage of cancer when it is diagnosed; the earlier the diagnosis, the better the survival rate.

November is lung cancer awareness month, and this week’s Domain Knowledge reviews AI tools used in the detection of lung cancer.

Case Study: AI Enhanced Peer Review Finds Undocumented Lung Nodules

A large United States based health system implemented an AI platform to deploy a natural language processing algorithm. The process augmented the work of the radiologists, ran a second review in the background on thousands of CT scans, and had a limited impact on the radiology workflow.

After review, specific scans were flagged if a lung nodule was detected but not mentioned in the radiology report. The flagged scans were further evaluated through the health system’s quality review program to determine if the flagged nodules were actual misses.

The health system found the use of AI to identify undocumented lung nodules supports early lung cancer detection.

Case Study: Detection of Multiple Lung Nodules on Chest X-ray Supports Diagnosis

Detecting lung nodules on a chest X-ray image is an important first step in assessing a patient. These images can help the physician identify the size, shape, and location of lung nodules, as well as other important characteristics. From there, the physician may recommend additional tests to rule out cancer or to determine other underlying issues.

In this case study, AI not only detected a nodule area but also highlighted more than 1 nodule area in both lungs.

Case Study: Using AI to Distinguish Lung Nodule from Other Easily Confused Lesions

Confusion in distinguishing between similar-looking findings on x-ray images is a problem that can be encountered by physicians (e.g., sclerotic rib lesions can mimic pulmonary nodules). The key is to look at prior radiographs. If the nodule remains projected over exactly the same rib site despite projectional differences, the physician can be confident that the lesions are within the rib rather than the lung.

AI can distinguish lung nodules from other easily confused lesions, which helps reduce false positives and reduce imaging scans needed for normal diagnosis purposes. 

That’s a wrap for this week’s review of news and happenings in the healthcare AI space. In closing, I’ll leave you with an invitation to learn more about the benefits of using an AI Hub to manage multiple applications across your clinical needs and offer you a personal demo of Ferrum’s platform and growing AI catalog.

If you have AI tips, suggestions, or resources you’d like to share, leave us a note below, and please feel free to suggest topics you would like to see covered in future posts.

Picture of Kathleen Poulos

Kathleen Poulos

Kathleen is a registered nurse with a digital marketing background, a love for using technology to solve healthcare challenges and a passion for improving patient care.

Contact Us

CASE STUDY

ARA Health Specialists

Use the button below to download your free case study and learn how our approach to validation has improved the number of clinically significant findings in AI software.

CASE STUDY

Sutter Health

Use the button below to download your free case study and learn how our approach to validation has improved the number of clinically significant findings in AI software.