[Artificial intelligence and machine learning in oncologic imaging].

Area of Science:

• Oncology
• Medical Informatics
• Artificial Intelligence

Background:

• Machine learning (ML) is increasingly integrated into various societal sectors, notably medicine.
• The application of ML in medicine, particularly oncology, has the potential to significantly alter medical practice.
• Current research indicates ML algorithms are achieving human-level or superior performance in critical oncologic tasks.

Purpose of the Study:

• To highlight the transformative impact of machine learning in oncologic patient care.
• To discuss the capabilities of ML algorithms, such as artificial neural networks (ANNs), in medical applications.
• To project the future role of artificial intelligence (AI) in advancing oncologic diagnostics and treatment.

Main Methods:

• Review of recent publications on ML applications in oncology.
• Analysis of ML algorithm performance in tumor identification, classification, prognosis estimation, and treatment evaluation.
• Examination of the characteristics of ML algorithms, including ANNs, regarding reproducibility, speed, and cost-effectiveness.

Main Results:

• Computers, using ML, are demonstrating superior performance compared to humans in tumor identification and classification.
• ML algorithms can accurately estimate prognoses and evaluate treatment effectiveness.
• ANNs, a key ML technology, provide reproducible, fast, and inexpensive solutions for complex medical tasks.

Conclusions:

• Machine learning is poised to become an integral component of medical practice.
• AI applications offer significant advantages for oncologic diagnostics.
• The integration of ML will enhance treatment strategies and patient outcomes in oncology.