Author: Matthew Reiss, MSE, PhD, Manager, Precision Medicine & Navigation, GO2 for Lung Cancer

You’ve likely started to see artificial intelligence (AI) being used more and more in your everyday life. While the benefits and challenges of using AI in fields such as healthcare must be carefully considered, we asked one of the most popular generative AI platforms, ChatGPT, what it thinks of AI’s future role in lung cancer care—plus we’ve added some thoughts of our own. Some highlights of the conversation are below.

Author’s note: Throughout this article, AI-generated responses are shown in purple text, while the author’s thoughts are shown in black text.

We asked ChatGPT: What are some of the major ways AI can revolutionize lung cancer care?

Chat GPT: AI has the potential to revolutionize the landscape of lung cancer research, diagnosis, and treatment in numerous ways. Here are some key areas where AI could play a transformative role in the future.

Early detection

Chat GPT: AI algorithms can analyze medical imaging data, such as chest X-rays and CT scans, with high accuracy. By identifying subtle patterns indicative of lung cancer at an early stage, AI can contribute to early detection, significantly improving patient outcomes.

GO2: Early detection is one of the ways AI has the most potential to impact lung cancer care. Computers are great at finding patterns hidden in data. Machine learning is a subfield of AI. An example of machine learning is when a programmer trains a computer to take information from an image, such as an X-ray or CT scan, and accurately predict whether a person is at an increased risk of developing lung cancer. A team of researchers from MIT are working on this exact problem and their program, Sybil, has exciting early results.

This technology could be taken a step further in a holistic approach to healthcare. It could be used to analyze changes in behavior, voice patterns, or social media activity to evaluate distress or changes in mental well-being and connect individuals with supportive resources earlier so fewer people are left behind. Using AI could allow those at an increased risk of lung cancer access to comprehensive screening and resources earlier, leading to better early-stage diagnoses and better long-term outcomes.

Precision medicine

Chat GPT: AI can analyze vast amounts of genomic and molecular data to identify specific genetic mutations and biomarkers associated with lung cancer. This information can be used to tailor personalized treatment plans, ensuring more effective and targeted therapies for individual patients.

GO2: Another thing that AI excels at over humans is processing large, complex datasets to predict outcomes. Understanding an individual’s history and molecular profile already helps us create personalized treatment plans for individuals. This is exactly what our LungMATCH program aims to do, but AI can help us look deeper into the data.

When expert navigators or healthcare professionals make treatment decisions, they have an inherent bias due to their own experiences and expertise. Additionally, even the smartest individuals can only consider so much information at once. These limitations do not exist for AI; the right AI tool can integrate data from patient records, molecular profile testing, real-time health data, and clinical trial data simultaneously and more consistently and without bias to predict the best personalized outcome for an individual. Given the right information, an AI tool could predict everything from a treatment’s effectiveness to its side effects and when to expect disease progression. AI also has enormous potential to help identify new, targetable biomarkers for lung cancer, which can lead to the development of more effective therapies.

Drug discovery

Chat GPT: AI algorithms can accelerate drug discovery by analyzing large datasets to identify potential drug candidates and predict their effectiveness. This can significantly reduce the time and resources required for developing new and more targeted treatments for lung cancer. 

GO2: Research shows that more than of new drugs that make it to clinical trials fail before reaching the market. Toss in an average development time of 10-15 years and $1-2 billion for each new drug approved, and you can understand why AI algorithms are already helping companies determine which potential drug candidates will deliver the most targeted and effective treatments. AI has enormous potential to help cut down on development time and costs, which means newer and more effective treatments can get people diagnosed with lung cancer sooner and potentially save more lives.

Community building and support

Chat GPT: AI-driven platforms can facilitate the creation of online communities for lung cancer patients, survivors, and caregivers. These platforms can provide a supportive space for sharing experiences, insights, and advice, fostering a sense of community, and reducing the isolation often associated with a cancer diagnosis. AI can also assist in taking complex genetic information and presenting it in a more understandable format. This can empower patients to make informed decisions and better comprehend their risk factors.

GO2: I completely agree with this response. AI can provide a new and faster way to connect people impacted by lung cancer to one another for support and community. The ability to share personal experiences, advice, and insights is part of the human experience — AI can make connecting like-minded people and sharing information easier.

AI can also act as a real-time, informed moderator, helping make understanding complex information easier by presenting that information in a practical, more approachable way. An example of this would be AI’s ability to rapidly share information in an individual’s primary language of choice, helping to overcome language barriers that currently prevent accessing support resources in underserved communities.

Many AI tools also allow you to ask clarifying or related questions based on previously generated responses, allowing individuals to get many of the answers they seek from the comfort of their own homes. It’s important to note that this is still an active area of development for AI, with experts working hard to ensure that the AI-generated information is both empathetic and accurate.

The future of AI and lung cancer

Chat GPT: As AI technologies continue to evolve, their integration into lung cancer care holds tremendous promise. The integration of these technologies can create a more comprehensive and patient-centric approach to lung cancer management by improving outcomes, enhancing efficiency, and ultimately transforming how we approach and manage this challenging disease.

In simple terms, AI is like a smart helper for doctors and scientists, making sure they do their best to keep you healthy and happy!

GO2: I wholeheartedly agree. These highlights just scratch the surface of what we’re already working on with AI for lung cancer care. There are far more ways AI is currently being integrated into our daily lives and ways it can be used in the future to answer questions we haven’t even thought of yet.

Additionally, there are caveats that we must give careful thought and consideration to as we continue to integrate AI more and more into our daily lives. Perhaps the most challenging consideration is that AI models require large, diverse datasets for training to ensure accurate outcomes and avoid bias. Data sets such as these can be difficult to collect in lung cancer research and algorithms that perform well in controlled research environments will require validation of their reliability with real-world data. This also presents important ethical and legal considerations to balance the need for access to diverse patient datasets and the protection of patient privacy.

While from my current standpoint, I don’t believe AI will fully replace the role that humans play in lung cancer care, I do believe it offers an incredibly powerful toolset we can make use of to greatly improve patient-centric care for individuals living with lung cancer.