Lunit AI Solution Confirms More Than Double Increase in Lung Nodule Detection Rate

Lunit's AI chest X-ray image analysis solution 'Lunit INSIGHT CXR'
[Photo by Lunit]

[Asia Economy Reporter Chunhee Lee] Medical artificial intelligence (AI) company Lunit announced on the 14th that a study published in the international journal Radiology showed that using AI image analysis solutions increases the detection rate of lung nodules during screening. Radiology is an SCI-level international journal published by the Radiological Society of North America (RSNA) and is the most prestigious international journal in the field of radiology, with a global Impact Factor of 29.146.

This study was conducted on 10,476 patients who underwent chest X-rays at the Health Promotion Center of Seoul National University Hospital from June 2020 to December 2021. The first group (AI group) used the chest X-ray image analysis solution 'Lunit Insight CXR' during the reading process, while the second group (non-AI group) did not use AI. The detection rates of actionable nodules suspected of malignancy were compared and analyzed between the two groups.

As a result, the lung nodule detection rate in the AI group was 0.59% (31 out of 5,238), which was more than twice significantly higher than the 0.25% (13 out of 5,238) detection rate in the non-AI group. Notably, among these, 8 patients in the AI group were ultimately diagnosed with lung cancer, whereas none were detected in the non-AI group.

Analysis of lung nodule sensitivity on chest X-rays through three-dimensional chest computed tomography (CT) scans showed that the AI group had a sensitivity of 56.4%, higher than the 23.2% in the non-AI group. On the other hand, the false referral rate was 45.9% in the AI group and 56.0% in the non-AI group, showing no statistically significant difference.

Professor Jinmo Koo of the Department of Radiology at Seoul National University Hospital, who led the study, said, "The study was conducted under conditions identical to the clinical environment of a university hospital to ensure the reliability of the results," adding, "The results strongly confirm that AI can provide practical assistance in reading chest X-rays and will especially contribute to the early detection of chest diseases and lung cancer."

Seobeom Seok, CEO of Lunit, also stated, "This is the first study to prove the efficacy of AI through a large-scale prospective, randomized controlled clinical trial (RCT) related to chest X-ray reading in actual medical settings," and added, "We expect that AI solutions will establish themselves as the standard of care for chest X-ray imaging in large-scale screening environments and contribute to the spread and settlement of AI in the field of radiology."

The study results were presented at the Asia-Oceania Congress of Radiology (AOCR 2023) held in Bangkok, Thailand, from the 9th to the 12th, and will also be presented at the Society of Thoracic Radiology (STR 2023) in Charleston, South Carolina, USA, from the 5th to the 8th of next month.

This content was produced with the assistance of AI translation services.