Artificial intelligence (AI) is already leaving its mark on medicine, but there is a bottleneck that no one can avoid: access to clinical data in quantity and quality, without putting patient privacy at risk. At that point—right where the rope between innovation and confidentiality is tightened—synthetic data begins to make a real difference.
Synthetic data are sets of information generated by algorithms capable of replicating the statistical properties of real clinical data, but without exposing personal identities or histories. They are not “made-up” bases: they are new data that reproduce plausible clinical patterns. As Diego Pereyra, Healthcare Global Director at Softtek, explains: “This technology opens a new horizon for health research: it accelerates clinical trials, the development of new drugs and allows hospitals to build reliable data networks, with representative samples on a regional or even global scale.”
This combination allows something that was almost impossible before: simulating complex medical scenarios, training diagnostic support algorithms or validating digital therapies without resorting to real clinical bases. It also makes it possible to work with rare diseases or very rare combinations, where a lack of volume usually slows down research.
A boost for Argentina and the region. In Latin America, where the fragmentation of health systems and the shortage of robust clinical repositories remain challenges, synthetic data becomes an accelerator. In Argentina, in particular, technology began to gain ground both in academic settings and in public initiatives linked to healthcare digitalization.
During the last year, meetings such as the Datathon 2025 of the Ministry of Health They put the topic on the table: interoperability, secure research and new ways to train algorithms without compromising privacy. Local universities and hospitals also experimented with rare disease simulations and diagnostic model validation.
“There are companies specialized in the generation of synthetic data that allow privacy to be preserved without sacrificing the quality or clinical value of the information. These solutions comply with the most demanding regulatory frameworks, such as GDPR in Europe (Europrivacy) and HIPAA internationally,” highlights Pereyra, adding, “In our country, synthetic data can democratize medical innovation, allowing hospitals and startups to research and develop AI solutions without depending on international repositories or violating the patient confidentiality.
bconcrete benefits for hospitals and patients
- More robust clinical models: They generate more balanced bases, reduce biases and improve diagnostic precision.
- Secure collaboration: Hospitals, universities and companies can exchange knowledge without sharing sensitive data.
- Democratized access: In regions with scarce or fragmented repositories, they function as a technological “equalizer.”
- Frictionless innovation: They allow algorithms to be validated, accelerate solutions and improve care without putting privacy at risk.
Synthetic data do not replace real clinical information, but rather complement it. “They are a bridge that accelerates research and makes medical innovation safer, especially in countries where access to large volumes of data is limited,” says Pereyra.
Challenges and next steps. Although the potential is enormous, specialists emphasize that generative models must be carefully evaluated. The key will be to ensure that the synthetic data maintains clinical value and does not introduce distortions. Transparency, medical validation and clear standards will be essential to build trust between professionals, institutions and patients.
“Synthetic data enables a new business unit, based on the responsible monetization of reliable data, always under secure collaboration schemes, promoting better informed decisions, more agile research and sustainable innovation in the health system,” concludes the expert.
* Diego Pereyra is a professional specialist at Softek.
by Diego Pereyra
