I hold a degree and PhD in physics and I am co-founder of PerMediQ GmbH. The mission of our company is to develop innovative and transparent machine learning and mathematical modeling solutions to improve the quality of disease management. Together with our partners and our team, we develop and implement recommendation systems for medicine based on ethical AI, which lead to greater transparency, robustness and reduced use of data volumes, while maintaining decentralized data structures to avoid data silos and maintain the data sovereignty of citizens.
In addition to innovation and project management for my company, I work about ethical artificial intelligence (AI), semantics and machine learning, fundamental aspects of complex systems, systems biology and evolution, as well as fundamentals of statistical physics and philosophical foundations of physics and biology. In addition to this scientific background, I have experience in startups, innovation and product development.
Inspired by my experience modeling complex systems in biology, I have observed that they differ from purely mechanistic systems. To assess the degree of observability of a system, I helped to define a novel complexity measure and researched the relationship between computer science and biological processes in evolution.
Also, motivated by the goal to avoid the suffering of animals in in-vivo experiments in different industries, like cosmetics and pharmaceutical industries, I collaborated with different colleagues across the European union in the COSMOS project in the development of Insilico organisms and whole body models for insilico tests of substances with dermal and oral absorption.
Other topics in my research career have been the development of novel uses of game theory for the modelling of soft matter, the exploration and quantification of envy between the networking of agents in the wake of social networks and links between soft matter and astrophysics (see selected publications below).
I’m also exploring the possibility that biology cannot simply be reduced as an emerging physical phenomenon, in part due to the persistent incompleteness (ambiguity) that has arisen in the various interconnected scales, and that “meaning” is essential in living systems, which is a non-physical element that can only be understood with other approaches such as the combination of biology and informatics, which limits the objective observability of biological systems and the ability to describe them with mathematical models.
I have also explored the future of the pharmacology/toxicology considering novel technologies like machine learning and semantic internet, and wrote some essays exploring from a philosophical perspective the interlinking between biology and physics.
Besides several honorary activities, including my activity as referee in several journals, I presented introductory lectures about general relativity for children (in the framework of the Hector Akademie, Stuttgart.
My education spans two continents: I wrote my PhD work at Kurt Binder’s group at the Johannes Gutenberg Universität in Mainz and wrote my degree (recognized as Diplom Physiker in Germany) at the Universidad Nacional de Colombia in Bogotá. Other stations were institute of theoretical physics at the Bremen University, the Max Planck Institute in Magdeburg, and the Astronomical Observatory in Bogotá, Colombia.
Below I list the chapters of the relevant topics of my work with links to some relevant publications:
Medical recommender systems that emulate human rational processes.
How to help physicians in their daily work using ethical AI? In this work, we report the development of a novel recommendation system of therapies in medicine based on the combination of continuous logic and deep learning algorithms (Logical Neuronal Networks, or LONNs). The advantage of this method is that, first, it is transparent, since the model results emulate human logical decision-making using logical combinations of input parameters, and second, it is safer against attacks than conventional deep learning methods, since it drastically reduces the number of trainable parameters in the model. This development is a contribution for future ethical AI applications.
Biology is much more than mechanisms, and informatics can be fundamental to better understand what is life.
Explanations based on low-level interacting elements are valuable and powerful since they contribute to identify the key mechanisms of biological functions. However, many dynamic systems based on low-level interacting elements with unambiguous, finite, and complete information of initial states generate future states that cannot be predicted, implying an increase of complexity and open-ended evolution. Such systems are like Turing machines, that overlap with dynamical systems that cannot halt. We argue that organisms find halting conditions by distorting these mechanisms, creating conditions for a constant creativity that drives evolution. We introduce a modulus of elasticity to measure the changes in these mechanisms in response to changes in the computed environment.
In-silico individuals for pharmacology and toxicology
In this work we opened de possibility to use in-silico models to represent toxic effects of substances (acetaminophen) only using mathematical models, as a framework for efficiently integrating inter-individual variability data into models, paving the way for personalized or stratified predictions of drug toxicity and efficacy.
Insilico Models to assess Toxicology.