public/uploads/demo/banner-tai-nguyen-1.jpg

Blogs

Welcome to our blog section on Artificial Intelligence (AI)! Here, we will explore in-depth one of the fastest and most exciting technological fields of the modern era.

Current and Future Perspectives: Artificial Intelligence in Oncology

09/01/2024 admin

Current and Future Perspectives: Artificial Intelligence in Oncology

AI in Oncology: Transforming Cancer Care with Advanced Technologies

From Concept to Transformative Reality

The integration of artificial intelligence (AI) in oncology has progressed from a concept to a tangible reality, offering transformative possibilities for cancer care.

Ongoing studies reveal expanding applications in cancer diagnosis, treatment, and research. Dr. Tufia C. Haddad of Mayo Clinic highlights AI's prevalence in cancer diagnostics, particularly in enhancing accuracy through machine learning and deep learning models. Numerous AI technologies, FDA-approved for oncology applications, are notably concentrated in radiology, with a focus on breast cancer detection.

Beyond radiology, AI demonstrates effectiveness in colorectal cancer screening, pathology slide analysis, and drug discovery for oncology. Dr. Laurie Margolies notes AI's crucial role in breast radiology, while Dr. Haddad emphasizes its impact on radiation treatment planning and remote patient assessment for supportive care.

Applications Under Investigation

AI research in oncology reveals diverse promising applications. Dr. Margolies anticipates AI breakthroughs in breast imaging, potentially surpassing family history in identifying breast cancer risks. Dr. Haddad explores AI's potential in identifying high-risk pancreatic cancer patients for early detection through abdominal imaging and health records. AI's transformative potential extends to noninvasive "virtual biopsy" through MRI radiomics for cancer diagnosis based on genetic mutations like IDH1 or BRAF.

Dr. Elemento highlights large language models' (LLMs) exciting potential in oncology, offering accurate responses to medical questions and personalized treatment recommendations through analyzing medical data. Recent studies show ChatGPT's 88% accuracy in responding to breast cancer screening questions, but limitations exist, indicating the need for refinement.

Natural language processing models are under investigation for predicting cancer patient survival outcomes. Nunez et al demonstrated the models' ability to predict survival based solely on initial oncology consultation data, suggesting a potential advancement in personalized patient care.

Optimizing AI for Clinical Use

Optimizing AI usage in clinical oncology requires addressing multiple limitations and issues, as emphasized by experts. Dr. Haddad stresses the ethical imperative of developing AI models with diverse datasets for accurate representation in patient cohorts. Dr. Elemento identifies data-sharing reluctance as a barrier, suggesting normalized incentives or alternative training algorithms for local model training.

Despite AI's progress in medicine, existing models need assessment in clinical practice. Dr. Haddad underscores the impact of true AI in oncology when actively used in clinical care, providing value to clinicians, enhancing patient outcomes, and improving health systems. Dr. Elemento highlights implementation challenges, urging solutions for integrating AI models into electronic health records, user-friendly interfaces, and high training costs to avoid limiting competition and access.

Establishing guidelines for AI safety, transparent dataset characteristics, and patient confidentiality protection is crucial, according to Dr. Haddad. She anticipates a transformative moment in oncology with the emergence of generative AI and large language models, expecting an unprecedented impact on cancer research and care delivery.

Disclosures: Dr. Elemento is a co-founder of OneThree Bio, a company utilizing AI for drug discovery in oncology. Dr. Margolies is on the medical advisory board for Screenpoint Medical. Dr. Haddad has no relevant disclosures to report.

Introducing DrAid™ Oncology D&T

VinBrain introduces revolutionary healthcare innovations, DrAid™ MRI Rectal Cancer D&T and DrAid™ CT Liver Cancer D&T, powered by AI. These advancements utilize advanced machine-learning technology to assist radiologists in automatically identifying abnormal lesions in MRI and CT scans, aiding in the early diagnosis of rectal and liver cancer. Beyond early detection, DrAid™ MRI Rectal Cancer D&T and DrAid™ CT Liver Cancer D&T play a crucial role in creating personalized treatment plans for oncologists and radiologists.

Explore the future of cancer diagnostics with these cutting-edge technologies and stay at the forefront of healthcare advancements!

---

Source: 

Rodriguez, Tori. "AI in Oncology: Current and Future Applications." CancerTherapy Advisor, 10 Nov. 2023, www.cancertherapyadvisor.com/home/cancer-topics/general-oncology/ai-in-oncology/. Accessed 10 Jan. 2024.