.png){kind=logo}
How Artificial Intelligence in Oncology Is Evolving Diagnosis, Treatment, and Drug Development to Improve the Lives of Cancer Patients
Artificial intelligence in oncology is transforming the landscape of cancer diagnosis and treatment, offering new and powerful tools to improve patient outcomes. From early detection to personalized treatment, AI provides solutions that are accelerating progress in the fight against cancer.
Artificial intelligence in oncology has enabled advancements in cancer diagnosis. According to shared articles, the FDA has approved more than 521 medical devices using AI and machine learning, primarily focused on radiology. These devices have been particularly effective in diagnosing breast, lung, and prostate cancers, representing 31%, 8.5%, and 8.5%, respectively, of the cases where AI has the most impact.
Artificial Intelligence Tool Developed to Predict Risk of Lung Cancer
Sybil is an artificial intelligence tool developed by researchers from MIT and the Mass General Cancer Center that uses low-dose CT scans (LDCT) to predict the risk of developing lung cancer up to six years before the first symptoms appear. This algorithm was validated by analyzing over 6,000 scans from the National Lung Screening Trial (NLST), 8,821 patients from Massachusetts General Hospital (MGH), and 12,280 from Chang Gung Memorial Hospital in Taiwan, achieving up to 94% accuracy in predicting lung cancer in some cases. Sybil does not require clinical data or radiologist annotations, allowing its real-time implementation in radiology stations to support decision-making. While the results are promising, researchers note that further prospective studies are needed to validate its effectiveness in more diverse populations.
Artificial intelligence in oncology transforms the present while also promising great advances for the future in areas such as drug development, virtual assistants for patients, and the optimization of healthcare processes.
AlphaFold, a technology developed by Google DeepMind, has revolutionized the field of structural biology by accurately predicting the 3D structure of 350,000 proteins based solely on their genetic sequence. This breakthrough has enabled researchers to better understand how proteins fold, which is essential for drug design as folding affects how they interact. Thanks to this technology, potential therapeutic targets can be identified more quickly and accurately, significantly reducing experimentation costs in the lab.
Additionally, AI tools like those developed by iNetMed are transforming preclinical drug discovery by creating computational models that simulate how a disease evolves based on changes in gene expression. These models allow predictions of how a drug might interact with a tumor, improving precision in the preclinical phase and increasing the chances of success in clinical trials.
Virtual assistants are also beginning to play a key role in cancer management. For example, Penn Medicine developed a chatbot called Penny that guides patients through their complex treatment regimens, helping them reduce errors and improve treatment adherence by 70%.
Available at: https://www.healthcareitnews.com/news/penn-medicine-uses-ai-chatbot-penny-improve-cancer-care
In a study conducted by Massive Bio and presented at the American Society of Clinical Oncology (ASCO), the viability of the DLVTB technology, which consists of a "board" of models that process and structure information from medical texts, predict evidence-based treatment protocols, and generate a final report using deep learning, called the Deep Learning Virtual Tumor Board (DLVTB), was demonstrated. This technology was used to improve the treatment of patients with advanced colorectal adenocarcinoma, identifying the most appropriate treatment for each patient. 63% of patients became eligible for at least one clinical trial, representing a significant improvement over the national average of 3% eligibility for such studies.
The impact of artificial intelligence in oncology is expected to continue growing. While challenges such as data bias, especially in Black populations, still exist, AI has the potential to transform cancer care globally.
Conclusion:
Artificial intelligence in oncology is helping to detect cancers at earlier stages, improve diagnostic accuracy, and personalize treatments for each patient. As AI continues to evolve, its role in the fight against cancer will become increasingly crucial, benefiting millions of people worldwide.
Sources:
Desai, A., Loaiza-Bonilla, A., Culcuoglu, C., Johnston, K., & Kurnaz, S. (2021). Outcomes and applicability of a deep learning virtual tumor board (DLVTB) in community-dwelling colorectal cancer (CRC) patients. Journal of Clinical Oncology, 39(15_suppl), e18635. https://doi.org/10.1200/JCO.2021.39.15_suppl.e18635
Kekatos, M. (2023). How artificial intelligence is being used to detect, treat cancer -- and the potential risks for patients. ABC News. https://abcnews.go.com/Health/artificial-intelligence-detect-treat-cancer-potential-risks-patients/story?id=101611751
Sava, J. (2023). Current applications of artificial intelligence in oncology. Targeted Oncology. https://www.targetedonc.com/view/current-applications-of-artificial-intelligence-in-oncology
Luchini, C., Pea, A., & Scarpa, A. (2022). Artificial intelligence in oncology: current applications and future perspectives. British Journal of Cancer, 126(1), 4-9. https://doi.org/10.1038/s41416-021-01633-1
Vachon, C. M., Scott, C. G., Norman, A. D., Khanani, S. A., Jensen, M. R., Hruska, C. B., Brandt, K. R., Winham, S. J., & Kerlikowske, K. (2023). Impact of artificial intelligence system and volumetric density on risk prediction of interval, screen-detected, and advanced breast cancer. Journal of Clinical Oncology, 22(11), 1153. https://doi.org/10.1200/JCO.22.01153
Johns Hopkins Kimmel Comprehensive Cancer Center. (2022). DeepTCR program for predicting immunotherapy responses. Johns Hopkins Medicine. https://www.hopkinsmedicine.org/kimmel_cancer_center/research/clinical_trials/
MIT and Mass General Cancer Center. (2023). Sybil AI tool for lung cancer detection. Massachusetts General Hospital. https://www.massgeneral.org/cancer-center
Penn Medicine. (2023). Penny: The virtual assistant for cancer patients. Penn Medicine. https://www.pennmedicine.org/
Marquedant, K. (2023). Artificial intelligence tool developed to predict risk of lung cancer. Massachusetts General Hospital. https://www.massgeneral.org/news/press-release/artificial-intelligence-tool-developed-to-predict-risk-lung-cancer
