In a new study, scientists from Caltech and the Washington University School of Medicine in St. Louis have taken a big step towards solving a long-standing challenge in cancer treatment.
They’ve used artificial intelligence (AI) to predict which lung cancer patients are more likely to see their cancer spread, particularly to the brain. This is a significant advancement, especially for those diagnosed with non-small cell lung cancer (NSCLC), the most common type of lung cancer.
Typically, doctors face a tough choice after lung surgery in early-stage NSCLC patients: whether to recommend further aggressive treatments like chemotherapy or radiation.
These treatments are potent and can have severe side effects, so they’re only worth considering if there’s a high risk of the cancer spreading.
Historically, predicting this risk has been tricky, leading to situations where patients undergo these harsh treatments unnecessarily.
The study, recently published in the Journal of Pathology, introduces AI as a promising solution to this dilemma. The researchers fed the AI program with biopsy images from 118 NSCLC patients and taught it to discern patterns that might predict the likelihood of the cancer spreading to the brain.
The AI’s performance was impressive, making accurate predictions 87% of the time. This is a significant improvement over the 57% accuracy rate of expert pathologists who participated in the study.
Changhuei Yang, a professor at Caltech and one of the study’s lead researchers, emphasized the potential of AI to reduce over-treatment in cancer patients. He highlighted that this study is just the beginning and that more extensive research is needed to confirm these promising results.
The AI’s ability to outperform human experts is particularly notable in the early-stage NSCLC patients, suggesting that AI could play a critical role in guiding treatment decisions.
This technology doesn’t just offer a way to make better decisions about post-surgery treatments; it also opens up possibilities for new therapeutic approaches.
The researchers aim to understand precisely what the AI sees in the biopsy images that humans don’t, which could lead to the development of targeted treatments for indicators of metastasis.
The AI’s workings remain somewhat of a mystery, as it doesn’t explain the reasoning behind its predictions. This black box aspect of AI technology is a double-edged sword.
While it can identify patterns and make predictions beyond human capabilities, understanding the “why” behind these predictions is crucial for translating AI insights into practical medical interventions.
Looking ahead, the team is also focusing on improving the quality and uniformity of biopsy images used in training AI systems.
They believe that by designing imaging instruments optimized for AI rather than human use, they can further enhance the accuracy and reliability of AI predictions in cancer treatment.
This study not only showcases the potential of AI in improving cancer treatment decisions but also underscores the need for interdisciplinary collaboration in medical research.
As scientists uncover more about how AI can be used in diagnosing and treating diseases, the hope is that patients will benefit from more personalized and accurate medical care, reducing the burden of unnecessary treatments and focusing on those most likely to improve outcomes.
If you care about cancer, please read studies that low-carb diet could increase overall cancer risk, and new way to increase the longevity of cancer survivors.
For more information about cancer, please see recent studies about how to fight cancer with these anti-cancer superfoods, and results showing daily vitamin D3 supplementation may reduce cancer death risk.
The research findings can be found in The Journal of Pathology.
Copyright © 2024 Knowridge Science Report. All rights reserved.
