Artificial Intelligence (AI) has been in existence since 1956, and in the 1970s, it shared its first experience in the healthcare sector with the so-called Mycin, an expert system aimed at detecting infectious blood diseases. Mycin reasoned, communicated in natural language with the user, and prescribed personalized medications to each patient.
In medicine, the use of machine learning models to search medical data and discover insights that help improve health outcomes and patient experiences is known as AI in modern medicine. Thanks to recent advancements in computer science and informatics, AI is rapidly becoming an integral part of current healthcare. AI-driven algorithms and applications are used to assist medical professionals in clinical settings and ongoing research.
Based on current advancements, AI holds the potential to be useful in the development of antibodies and drugs that aid in enhancing the immune system response. It also collaborates in clinical studies, molecule design, human genome study, medical robotics, prosthetic development, simulation, training, and augmented reality.
Applications of AI in Modern Medicine
Some notable examples of AI application in the field of healthcare are as follows:
- Google Brain, Google’s AI research team, has developed a system that detects protein crystallization with a reliability rate of 94%. This technology enables pharmacists to understand how the application of a specific molecule benefits the treatment of a particular disease through more precise mechanisms and in less time. Work is being done on developing neural networks that can work with protein language models, similar to the language models used by applications like ChatGPT, but instead of texts and information, protein sequences are input. These advancements can expedite the processing of models that allow for generating proteins with a precise structure to fulfill their function and even generate new proteins adapted to disease mutations, which could aid in combating mutations such as COVID-19 and improve vaccine efficacy.
- In a study conducted in Los Angeles and published in the scientific journal Nature, the capacity of AI to analyze echocardiograms was observed. Cardiologists made fewer corrections to the analyses performed by AI (16.8% of the total) compared to those analyzed by ultrasound technicians (27.2% corrected). Additionally, cardiologists could not distinguish which analyses were conducted by AI and which were the product of humans. Another study published in The Lancet Digital Health, conducted in Germany, found better breast cancer diagnosis when doctors and AI work together.
- Recently, scientists from the University of Sydney (Australia) and Boston University (USA) have designed a tool that utilizes AI principles with the aim of diagnosing Parkinson’s disease before the first symptoms appear in patients. Through a neural network system based on machine learning, biomarkers present in the blood that could have gone unnoticed by traditional techniques were explored. Although this development was validated in less than forty individuals and requires further refinement, it represents a promising first step.
Specialized Education in AI in the Field of Health
Currently, there are specialized programs in this field, such as those offered at the renowned Favaloro University in Argentina, which cover concepts such as machine learning, biostatistics, data science, neural networks, among others.
While AI used in a favorable manner can be highly beneficial for the healthcare field, it is important to exercise caution. The AI revolution is advancing rapidly, seemingly without anyone’s permission. It is crucial to maintain an ethical approach and carefully consider the potential impacts and challenges that may arise when implementing these technologies in the healthcare domain.