AI & ChatGPT - what lies ahead in medicine?
What role does artificial intelligence play in the patient's journey in modern medicine? Artificial intelligence applied in medical practice can be an effective assistant to doctors. Radiodiagnostics is one of the most fertile fields for the application of AI in oncology. AI can count tiny details and deviations and raise suspicions where the human eye is not capable. Today, oncology is moving towards tumor diagnosis at the gene level, and AI can be a key tool in improving diagnostics. Can AI already evaluate genomic analysis and identify genetic mutations of the disease?
Artificial intelligence is advancing at a dizzying pace, but its safe use in healthcare has clear boundaries. Health data are enormous, inconsistent, and physicians’ decision-making is sensitive to the quality of information. The lecture showed where AI is already genuinely helping, and why general-purpose chatbots are not yet ready for clinical practice. Healthcare data are vast and complex: there are approximately 70,000 ICD diagnostic codes, and oncology patients often have multiple diagnoses at once. Over a lifetime, a patient generates roughly a million gigabytes of data—from MRI images and laboratory biomarkers to data from smartwatches. That is a volume comparable to hundreds of millions of books, which a person cannot “read” even with the help of a team of experts. When making decisions, physicians typically need 5 to 10 key data points, often more in oncology. In practice they encounter “regulars” (standard cases), “repeaters” (challenging, frequently recurring cases), and “strangers” (unusual, hard-to-read findings). If there is too much data, or it is incomplete or distorted, the risk of errors increases, which can be associated with worse treatment outcomes. It therefore makes sense to look for ways to manage large volumes of data and use them meaningfully.Why healthcare needs AI
