Comparative analysis of Artificial Intelligence applications in oncology: Assessing progress in lung cancer and brain cancer diagnosis and treatment
Comparative analysis of Artificial Intelligence applications in oncology: Assessing progress in lung cancer and brain cancer diagnosis and treatment
Paniz Sanjari,1,*
1. Iran university of medical science
Introduction: Cancer refers to a group of diseases characterized by abnormal cell growth, where early detection is essential for the treatment and survival of patients. Given the increasing use of artificial intelligence (AI) in clinical medicine, the present review study aims to explore the applications of artificial intelligence in cancer diagnosis and treatment, comparing advancements in lung cancer and malignant brain tumors.
Methods: The current review study followed the PRISMA protocol for data collection and search, using MESH terms including “artificial intelligence,” “lung cancer,” “oncology,” “diagnosis,” and “brain tumor.” The search was conducted in English databases (PubMed, Scopus, Google Scholar) for studies published between 2018 and 2024. Inclusion criteria consisted of systematic reviews, meta-analyses, cohort studies, case-control studies, and randomized controlled trials (RCTs). Exclusion criteria included abstracts without full-text articles and studies outside the specified timeframe. A total of 51 articles were retrieved, of which 19 met the inclusion criteria after applying the exclusion criteria.
Results: Among the articles, 36% focused on the general applications of artificial intelligence in neuro-oncology, 31% on lung cancer patients, and the remaining 31% on brain tumor patients. The review indicates that artificial intelligence has shown high potential in pathological assessments, initial screening, prognostic assessment, surgery, and immunotherapy for lung cancer. Additionally, artificial intelligence algorithms have demonstrated success in brain tumor segmentation, diagnosis, differentiation, grading, treatment response, and clinical outcome predictions for brain tumors. It has been found that limitations still exist in the use of artificial intelligence in both areas but the results for lung cancer diagnosis appear to be more reliable.
Conclusion: Although artificial intelligence algorithms have advanced in cancer diagnosis and treatment in recent years, their progress seems to vary across different cancer types. Moreover, significant limitations in their clinical application continue to be reported.