Percorrer por autor "Rossi, Camila Maria"
A mostrar 1 - 1 de 1
Resultados por página
Opções de ordenação
- Evaluation of fungal growth dynamics in food remains under indoor conditions for forensic temporal estimationPublication . Rossi, Camila Maria; Barroso, Maria Helena; Mascarenhas, Paulo; Silva, Zoé daEstimating the post-mortem interval (PMI) remains a central challenge in forensic science, especially in closed environments, where classical methods—such as the assessment of cadaveric signs and entomology—are limited by the absence of insects and the relative stability of the microclimate. In this context, the study of fungal growth emerges as a promising alternative, given that fungi follow predictable temporal patterns during the decomposition of food remains. This study investigates the use of common food remains (ripe bananas and prepared soups) as model substrates to identify patterns of microbial colonisation that are potentially useful in estimating PMI. The effects of temperature, humidity, and initial substrate condition were evaluated using observations in a domestic environment and temperature-controlled tests on some bananas—complemented by morphological and molecular laboratory analyses. Distinct colonisation profiles were identified: Cladosporium cladosporioides, Meyerozyma caribbica and Penicillium citrinum predominated in soups, while Fusarium verticillioides dominated in bananas, after an initial stage of colonisation by yeasts, followed by the growth of filamentous fungi. The relationship between growth area and time was modelled using machine learning techniques (ridge, elastic net polynomials, gradient boosting and random forest), which revealed good accuracy for short intervals (≈24–72 h), especially in species with stable growth, such as P. citrinum, and lower performance in more variable cases, such as F. verticillioides. The results show that temperature and humidity are determinants of the rate of colonisation; temperature-controlled trials on some bananas reduced uncertainty and increased the accuracy of predictive models. Despite limitations, the data suggest that food waste can function as complementary “microbial clocks” in closed environments, contributing to the reconstruction of the chronology of events and reinforcing the potential of forensic microbiology in estimating PMI.