INTERNATIONAL CONGRESS OF
CancerGenetics
Artificial Intelligence in oncology: Enhancing diagnosis and treatment
-
Ali Rezaei,1 Shirin Farivar,2,*
1. Department of Cell and Molecular Biology, Faculty of Life Sciences and Biotechnology, Shahid Beheshti University, Tehran, Iran
2. Department of Cell and Molecular Biology, Faculty of Life Sciences and Biotechnology, Shahid Beheshti University, Tehran, Iran
- Introduction: The world of cancer treatment is being transformed by artificial intelligence, which is improving the accuracy of cancer diagnosis and prognosis. AI algorithms, like convolutional neural networks (CNN) and support vector machines (SVM) are now being utilized in the diagnostic process. For example, Google’s LYNA (LYmph Node Assistant) and SMILY (Similar Medical Images Like Yours) are making strides in breast cancer detection. LYNA achieved a 99% success rate in identifying breast cancer on slides. Additionally, AI is playing a role in precision oncology by aiding in predicting outcomes and determining optimal treatment plans by assessing responses to treatment strategies.
- Methods: A thorough review of 10 journal articles from databases such as PubMed, Google Scholar, and Scopus was conducted to gather data on advancements in AI applications within the field of oncology focusing specifically on diagnostic methods and treatment approaches. This review of sources including primary research studies and systematic reviews offers a comprehensive analysis of how AI shapes the oncology landscape.
- Results: AI models have achieved an accuracy rate exceeding 90% in detecting glioma, breast, and prostate cancers, marking an advancement from a decade ago. Other AI applications such as BANDIT (Bayesian ANalysis to determine Drug Interaction Targets) are being used for the prediction of the target of existing molecules and drugs. Moreover, AI-enhanced imaging tools have increased detection rates in lung and prostate cancers while algorithms analyzing data are uncovering new drug targets and cancer subtypes, thus expanding the horizons of personalized medicine. Yet, there are several challenges including overfitting, black box problem, and discrepancies among facilities, especially in medical imaging.
- Conclusion: AI models have achieved an accuracy rate exceeding 90% in detecting glioma, breast, and prostate cancers, marking an advancement from a decade ago. Other AI applications such as BANDIT (Bayesian ANalysis to determine Drug Interaction Targets) are being used for the prediction of the target of existing molecules and drugs. Moreover, AI-enhanced imaging tools have increased detection rates in lung and prostate cancers while algorithms analyzing data are uncovering new drug targets and cancer subtypes, thus expanding the horizons of personalized medicine. Yet, there are several challenges including overfitting, black box problem, and discrepancies among facilities, especially in medical imaging.
- Keywords: Artificial Intelligence, Cancer Diagnosis, Precision Oncology, Deep Learning, Cancer Treatment
