Fuzzy Logic for Knowledge-Driven and Data-Driven Modeling in Biomedical Sciences

Fuzzy logic is characterized by relevant features, such as model interpretability and the capability of dealing with uncertain or heterogeneous data, which make it particularly suitable for the definition of mathematical models in the field of biomedical sciences. We describe here two different modeling approaches, based on a knowledge-driven or data-driven strategy, that we previously developed and applied for the definition, simulation and analysis of fuzzy inference systems and fuzzy networks. To showcase their benefit in biomedical sciences, we show three applications related to: (i) the determination of minimal drug combinations to induce apoptotic death in cancer cells, (ii) the analysis of oscillatory regimes in signal transduction pathways, and (iii) the assessment of tremor severity in a neurological disorder. These models are characterized by a high level of interpretability – thanks to the use of linguistic terms, fuzzy sets, and fuzzy rules written in human-comprehensible language – making their application suitable in real-world scenarios.