Over the past year, the concept of “artificial intelligence” has been juxtaposed with any industry, even some that we would not normally associate with similar concepts such as agriculture or construction. Even the medical sector and the pharmaceutical industry have not been indifferent to the latest innovations in this field.

In some ways, artificial intelligence algorithms fit perfectly with some aspects of the medical world. To work properly, AI must be trained on a large amount of relevant data (so-called machine learning). Each patient’s medical records are the perfect pool to train AI to not only be able to provide a diagnosis, but do so based on a patient’s entire medical history, returning a personalized and individual-specific report.

Although these prospects are tantalizing, there is still a long way to go before these intelligent algorithms can be fully integrated into everyday medical practice. Let’s try to see how artificial intelligence will (and is) changing the world of medicine.

AI in medical diagnosis

An algorithm before being put to work needs to be instructed with a large amount of well-structured data, i.e., previously annotated by a human being, so that the virgin algorithm is able to recognize what is submitted to it. Corresponding to this feature are the analyses performed on tumor tissues.

One of the first algorithms that can overcome the ability of physicians in the field of image classification was developed by researchers at Seoul National University Hospital and College of Medicine.

Deep Learning based Automatic Detection (DLAD) analyzes chest X-rays and helps detect abnormal cell growths, which could correspond to tumors. The artificial intelligence was able to outperform 17 of 18 physicians in a tumor tissue recognition task, paving the way for what could be a new era in disease diagnosis.

Another algorithm was developed by Google AI Healthcare and is named LYNA (Lymph Node Assistant), which identifies metastatic breast cancer tumors from lymph node biopsies, with 99% accuracy. Soon this technology may also be applied to X-rays, MRIs, and any other form of analysis that can return images of the body.

AI in the pharmaceutical industry

Artificial intelligence is increasingly being used in the discovery and design of new drugs. The main advantage is the large time savings at all stages of development, from testing to approval and distribution in the market. Time savings that translate into lower costs and higher earnings.

In addition to the actual savings, drugs developed using AI have better success rates and carry fewer health-related risks during preclinical studies.  The global market size of artificial intelligence in pharmaceuticals already amounted to $1.24 billion in 2022. In 2023 it had a compound annual growth rate of 32.8 percent, reaching a total value of $1.64 billion.

AI in personalized medicine

The great analytical power of algorithms is enabling medicine to progressively move away from “one-size-fits-all” treatments and embark on a path of evolution toward personalized medicine, which takes into account each individual’s medical profile and lifestyle in order to provide unique and specific treatments.

The use of personalized medicine based on artificial intelligence could enable:

  • Better treatment of common diseases, such as heart disease and cancer, and rare diseases, such as cystic fibrosis
  • An optimization of drug timing and dosing, based on the characteristics of each individual patient
  • Of screening patients based on their individual health profiles, instead of the current generic criteria of age and sex.

This personalized approach will lead to earlier diagnosis and more effective treatments, saving lives while making better use of resources.

AI in national health systems

Within complex systems such as the NHSs of various countries around the world, AI becomes an effective tool for addressing some of the diverse challenges that such large organizations can encounter:

  • Allocating health care staff and equipment more efficiently
  • Stimulating collaboration by forming teams of physicians, nurses, and other healthcare professionals with complementary skills
  • Ensuring better privacy for patients
  • Create of chatbots based on language models (such as ChatGPT) to enhance healthcare by providing initial support to patients who are looking for information.

In addition to the technological aspect, there are ethical, regulatory and security issues that need to be considered. The goal of AI research is not to build algorithms that can supplant physicians and healthcare professionals, but to develop systems that enable humans to take their capabilities to the next level.