Lung cancer is very common form of cancer which is highly diagnosed and death rate is high. The role of technology can be of great help in such a dangerous disease. With the continuous development of science and technology, artificial intelligence (AI) can help early diagnosis of lung cancer.
According to a group of doctors in Boston – AI model can predict a risk of death from lung cancer on easily obtainable x-rays; which could help in planning of treatment and personalized prevention.
A Convolutional Neural Network (CNN) model developed via researchers group led by Dr. Jacob Weiss of Massachusetts General Hospital to estimate the risk of lung cancer death using a chest x-ray as the sole input. In a test on three datasets containing more than 15,000 individuals, the model performed well.
In study, published on Nature Communications, group of researchers wrote “Our findings motivate the use of deep learning to identify individuals at high risk of lung disease mortality from easily obtainable and low-cost chest radiograph images. These findings may allow for improved risk assessment of those who would benefit most from personalized prevention and treatment strategies.”
Researchers also explained that chest x-ray is a common early test for lung cancer, but it is still difficult for humans to detect the actionable early stage of the disease by X-ray reading. AI can play a important role in the risk and extent of lung disease beyond established methods.
The AI CNN model developed by group named CXR-Lung-Risk to identify person on x-ray who may be at high risk for lung cancer death.
In this study researchers include 147,497 x-rays of 40,643 persons and tested the model in three independent cohorts comprising 15,976 individuals between 35 and 90 years of age enrolled in previous trials. In this lymphadenopathy, lung nodules, pleura and lung fibrosis, cardiac abnormalities, chronic obstructive pulmonary disease (COPD)/emphysema, atelectasis, opacities, and bone/chest wall lesions x-rays are included. Also, the data included the dates of actual observed deaths of patients.
According to researchers the probability of risk is usually expressed in a percentage, which can be difficult to understand. So, they created a method for model from the output risk from a single X-ray expressed in years.
The group concluded with “CXR-Lung-Risk could be used with minimal disruption of current clinical workflows and automatically analyze the latest radiograph of a patient at high speed and low additional cost37. As such, CXR-Lung-Risk could serve as an early warning system to triage patients into existing screening and chronic pulmonary disease pathways, and to both provide more accurate risk assessments for those programs and increase adherence to guidelines-based therapies.”
The information shared in this blog is for educational purposes only and is certainly not a substitute for medical advice. Please consult your healthcare practitioner for any medical needs.