Evolution of Artificial Intelligence in Oncology
Artificial intelligence (AI) has revolutionized many areas of medicine, and oncology is not an exception. From the beginning, AI has transformed the diagnosis, treatment, and management of cancer, providing more precise and personalized tools to combat this disease.
The use of AI in oncology began to gain ground in the early 21st century, when advances in machine learning and data processing enabled the development of algorithms capable of analyzing large volumes of medical data. Initially, these efforts focused on the diagnosis and early detection of cancer through the analysis of medical images. Let’s highlight some good examples:
In 2012, IBM Watson for Oncology pioneered the use of AI for cancer diagnosis, using machine learning to analyze clinical data and provide personalized treatment recommendations based on medical literature and previous cases.
In 2016, an emerging company called PathAI began developing AI algorithms to enhance the accuracy of pathological diagnosis, analyzing histological images to detect anomalies that could indicate cancer.
As technology advanced, AI started playing a more integral role in oncology, covering not only diagnosis but also treatment planning and administration.
In AI-guided radiotherapy: In 2014, the MRIdian® MR Linac system was launched. This system combines magnetic resonance imaging (MRI) with a linear accelerator (Linac) to offer adaptive real-time radiotherapy. It uses AI to monitor tumor movement and adjust treatment, improving precision and reducing side effects.
Varian’s Ethos™ therapy, an AI-driven adaptive system, received FDA approval on February 11, 2020. Developed by Varian, this system uses AI to adapt radiotherapy plans to daily changes in patient anatomy, optimizing radiation delivery.
In personalized chemotherapy: Tempus, founded in 2015 by Eric Lefkofsky, focuses on precision medicine, using AI to analyze genomic and clinical data. Tempus helps oncologists select the most effective treatments for each patient based on their genetic profile.
In robotic-assisted oncologic surgery: CyberKnife®, developed by Accuray Incorporated, uses stereotactic radiotherapy for precise tumor treatment. This robotic radiosurgery system uses AI to deliver highly precise radiotherapy treatments, especially for hard-to-reach or moving tumors during breathing.
When focusing on the economic impact of AI in oncology, it is significant in terms of cost savings and improved health outcomes.
In resource optimization: AI can help reduce costs associated with unnecessary tests and procedures through more precise and personalized diagnostics.
Automated systems enable faster and more accurate treatment planning and administration, reducing waiting times and operational costs, thereby directly impacting efficiency improvement.
AI allows treatments to be tailored to each patient’s specific needs, personalizing treatment, improving outcomes, and reducing the incidence of severe side effects and hospital readmissions.
However, despite numerous benefits, the use of AI in oncology also presents challenges and potential disadvantages. Some relate to difficulties in access, as not all patients have access to advanced technologies, exacerbating inequalities in cancer treatment. Additionally, implementing AI systems can be costly, limiting adoption in resource-constrained environments.
Although technological advances are here to stay, there are concerns about dependency on technology. Some argue that doctors may become too reliant on AI systems, potentially reducing their clinical skills and ability to make decisions without technological assistance. Moreover, AI algorithms are not infallible and can make mistakes, especially if trained on low-quality or biased data.
As with almost all technology and health-related discussions, privacy concerns arise when using large volumes of medical data, raising issues about patient information privacy and security. Ethical questions also arise regarding AI-driven automated decision-making in cancer treatment, concerning patient responsibility and consent.
In April 2023, I wrote about the use of machine learning in health economics, and the use of artificial intelligence would only useful to square the circle in health systems launched in the last century. Private health care knows it and tries to use that space.
EDUCATIONAL PILLS:
International Master in Medical Affairs. Program part of the educational offerings of the BioTech and Pharma School. In Memoriam César Nombela. Certified by the Catholic University of Ávila. Information
Advanced Management Program. IE Business School. Executive Development Programs. Information
READING PILLS:
Respected Cost Containment and Data Management Vendor Inks Deal with Promising New TPA. Health Cost IQ.
It’s not price-only theory. Economic Forces by Brian Albrecht.
Inflammation: its impact on aging and disease risk, and how to identify, prevent, and reduce it. Petter Attia.
Critical Thinking in Academic Research — Second Edition. Robing Ewing and Cindy Gruwell. Download for free.
STARTUP PILLS:
Varda Hopes New Research Draws More Drugmakers to Space Factories. Wall Street Journal.
Investment Opportunities. Capital Cell
PODCAST PILLS:
Launch of Tome Biosciences, AI scientist, mRNA vaccines. Touching Base.
EVENT PILLS:
ISPOR. Digital Real-World Evidence Generation Approaches in Rare Diseases and Oncology. June, 5–6 Register.
Skills to Thrive in the AI World: Critical thinking and emotional intelligence. June, 7 13h (BST) Life Science Access Academy.
The Summer 2024 HESG conference registration is now open. The conference will be held on the central campus at the Scarman Conference Centre, University of Warwick, Coventry, CV4 7SH from the 3rd to 5th July 2024. More details are available on the Warwick HESG welcome page.
HESG Summer Conference: 3rd to 5th July 2024. More info & registration.
