Pompa Marcello, Panunzi Simona, Borri Alessandro, D'Orsi Laura, De Gaetano Andrea
Institute of Systems Analysis and Informatics "A. Ruberti" (IASI), National Research Council of Italy, Rome, Italy.
Institute for Biomedical Research and Innovation (IRIB), National Research Council of Italy, Palermo, Italy.
PLoS One. 2025 Jan 2;20(1):e0316401. doi: 10.1371/journal.pone.0316401. eCollection 2025.
The growing availability of patient data from several clinical settings, fueled by advanced analysis systems and new diagnostics, presents a unique opportunity. These data can be used to understand disease progression and predict future outcomes. However, analysing this vast amount of data requires collaboration between physicians and experts from diverse fields like mathematics and engineering.
Mathematical models play a crucial role in interpreting patient data and enable in-silico simulations for diagnosis and treatment. To facilitate the creation and sharing of such models, the CNR-IASI BioMatLab group developed the "Gemini" (MoSpec/Autocoder) system, a framework allowing researchers with basic mathematical knowledge to quickly and correctly translate biological problems into Ordinary Differential Equations models. The system facilitates the development and computation of mathematical models for the interpretation of medical and biological phenomena, also using data from the clinical setting or laboratory experiments for parameter estimation.
Gemini automatically generates code in multiple languages (C++, Matlab, R, and Julia) and automatically creates documentation, including code, figures, and visualizations.
This user-friendly approach promotes model sharing and collaboration among researchers, besides vastly increasing group productivity.
在先进分析系统和新型诊断技术的推动下，来自多个临床环境的患者数据日益丰富，这带来了独特的机遇。这些数据可用于了解疾病进展并预测未来结果。然而，分析如此大量的数据需要医生与数学和工程等不同领域的专家合作。
数学模型在解释患者数据方面发挥着关键作用，并能够进行计算机模拟以辅助诊断和治疗。为了促进此类模型的创建和共享，意大利国家研究委员会智能系统与信息技术研究所（CNR-IASI）的生物数学实验室小组开发了“双子座”（MoSpec/自动编码器）系统，这是一个框架，使具有基本数学知识的研究人员能够快速、正确地将生物学问题转化为常微分方程模型。该系统有助于开发和计算用于解释医学和生物学现象的数学模型，还可利用临床环境或实验室实验的数据进行参数估计。
双子座能自动生成多种语言（C++、Matlab、R和Julia）的代码，并自动创建文档，包括代码、图表和可视化内容。
这种用户友好的方法除了极大提高团队生产力外，还促进了研究人员之间的模型共享与合作。
