AI and Cancer Treatment

Introduction

60% of Americans would feel uncomfortable if their health care provider relied on artificial intelligence (AI) for their medical care; only 38% believe that AI used for disease diagnosis and treatment recommendation improves health. A recent survey done by Pew Research Center reveals how the public views the use of AI in their own health care. While the majority expected difficulties in the patient-provider relationships, 40% believe AI can reduce the number of mistakes by healthcare providers.

How can AI support cancer treatment?

1. AI in cancer imaging

AI models detect, characterize, and monitor tumors through cancer imaging. During detection, the tool highlights suspicious regions in images and gives high false-positive rates and overdiagnosis. Then, it provides basic information on intra-tumor heterogeneity and variability in characterization. Lastly, the AI model can monitor tumor changes and detect many discriminative traits in the images over time that go beyond those measured by human readers. In cancer imaging, computer-assisted detection systems will reduce the mistakes of humans and lead to better outcomes for patients.

2. AI in killing cancer cells

What if a predictive model could advise researchers on the specific commands to incorporate into cells for desired functions? A collaboration between UC San Francisco and IBM Research employed combinatorial signaling motif libraries and machine learning to decipher the phenotype of CAR-T cells. By leveraging their understanding of algorithm data, machine learning determines CAR costimulatory signaling rules, facilitating the design of enhanced non-natural costimulatory domains. This results in improved in vitro and in vivo phenotypes, utilizing signaling motif libraries. This groundbreaking technology anticipates the most suitable natural or synthesized elements to introduce into cells, enabling precise responses to complex diseases.

3. Application of ChatGPT

ChatGPT (Chat Generative Pre-Trained Transformer) is an AI chatbot trained to follow instructions in a prompt and provide a detailed response as if you are having a human-like conversation. This technology has grown dramatically and is currently influencing various industries, including healthcare. The chatbot can be used for clinical decision support, medical recordkeeping, interpreting medical literature, and disease surveillance. But, it is very difficult to trust the given response by ChatGPT 100% due to the blending of inaccuracies and insights. To improve this limitation, OpenAI, the company that developed chatGPT, is planning to launch a medical version.

Conclusion

Moreover, as AI technology progresses, it opens promising avenues for cancer treatment and healthcare progress. However, developing stringent safeguards and protocols to safeguard patient privacy and data integrity is of utmost importance. Addressing ethical concerns, including accountability and bias mitigation in algorithms, should remain a top priority during integration. By harmonizing innovation with responsible deployment, AI can revolutionize healthcare, ensuring personalized and efficient cancer treatments for all patients and ultimately contributing to improving communities' well-being.

Sources

• https://www.pewresearch.org/science/2023/02/22/60-of-americans-would-be-uncomfortable-with-provider-relying-on-ai-in-their-own-health-care/

• https://acsjournals.onlinelibrary.wiley.com/doi/abs/10.3322/caac.21552

• https://www.ucsf.edu/news/2022/12/424406/how-ai-found-words-kill-cancer-cells

• https://www.science.org/doi/10.1126/science.abq0225

• https://www.forbes.com/sites/michaelmillenson/2023/03/19/in-cancer-patient-empowering-ai-begins-to-change-care-relationships

• https://www.forbes.com/sites/bernardmarr/2023/05/19/a-short-history-of-chatgpt-how-we-got-to-where-we-are-today