Clinical Prediction Models for Prognosis of Colorectal Liver Metastases: A Comprehensive Review of Regression-Based and Machine Learning Models.

Colorectal liver metastasis (CRLM) is a disease entity that warrants special attention due to its high frequency and potential curability. Identification of "high-risk" patients is increasingly popular for risk stratification and personalization of the management pathway. Traditional regression-based methods have been used to derive prediction models for these patients, and lately, focus has shifted to artificial intelligence-based models, with employment of variable supervised and unsupervised techniques. Multiple endpoints, like overall survival (OS), disease-free survival (DFS) and development or recurrence of postoperative complications have all been used as outcomes in these studies. This review provides an extensive overview of available clinical prediction models focusing on the prognosis of CRLM and highlights the different predictor types incorporated in each model. An overview of the modelling strategies and the outcomes chosen is provided. Specific patient and treatment characteristics included in the models are discussed in detail. Model development and validation methods are presented and critically appraised, and model performance is assessed within a proposed framework.

Predicting tyrosinaemia: a mathematical model of 4-hydroxyphenylpyruvate dioxygenase inhibition by nitisinone in rats.

Nitisinone or 2-(2-nitro-4-trifluoromethylbenzoyl)cyclohexane-1,3-dione is a reversible inhibitor of 4-hydroxyphenylpyruvate dioxygenase (HPPD), an enzyme important in tyrosine catabolism. Today, nitisinone is successfully used to treat Hereditary Tyrosinaemia type 1, although its original expected role was as a herbicide. In laboratory animals, treatment with nitisinone leads to the elevation of plasma tyrosine (tyrosinaemia). In rats and Beagle dogs, repeat low-dose exposure to nitisinone leads to corneal opacities whilst similar studies in the mouse and Rhesus monkey showed no comparable toxicities or other treatment related findings. The differences in toxicological sensitivities have been related to the upper limit of the concentration of tyrosine that accumulates in plasma, which is driven by the amount/activity of tyrosine aminotransferase. A physiologically based, pharmacodynamics ordinary differential equation model of HPPD inhibition to bolus exposure of nitisinone in vivo is presented. Going beyond traditional approaches, asymptotic analysis is used to separate the different timescales of events involved in HPPD inhibition and tyrosinaemia. This analysis elucidates, in terms of the model parameters, a critical inhibitor concentration (at which tyrosine concentration starts to rise) and highlights the contribution of in vitro measured parameters to events in an in vivo system. Furthermore, using parameter-fitting methods, a systematically derived reduced model is shown to fit well to rat data, making explicit how the parameters are informed by such data. This model in combination with in vitro descriptors has potential as a surrogate for animal experimentation to predict tyrosinaemia, and further development can extend its application to other related medical scenarios.

The effects of temperature, humidity and rainfall on captan decline on apple leaves and fruit in controlled environment conditions.

BACKGROUND: Captan is an important fungicide for controlling diseases in horticultural crops. Predicting its dissipation is important for estimating dietary risks and optimising pesticide application. Experiments were conducted to determine the relationship of captan loss on apple leaves with temperature, humidity and rainfall, and to investigate captan loss on fruit in dry conditions. RESULTS: There was large unit-to-unit variability in captan residues in spite of the controlled application. Temperature and humidity had negligible effects on captan loss. Captan loss is predominately due to washoff by rain, although a certain proportion of captan may bind to the plant surface tightly and hence may not be readily removed by rain. About 50% of captan can be washed off by as little as 1 mm of rain after an application, and the loss appeared not to relate to the amount of rain. Under dry conditions, daily loss of captan is estimated to be around 1% on both fruit and leaves, giving a half-life of ca 70 days. CONCLUSIONS: Captan loss on leaf and fruit surfaces is primarily due to rain washoff.

The temporal pattern of captan residues on apple leaves and fruit under field conditions in relation to weather and canopy structure.

BACKGROUND: Captan is an important fungicide for controlling diseases in horticultural crops. Understanding its dissipation is important for estimating dietary risks and optimising pesticide application. Field experiments were conducted on apple leaves and fruit to investigate (1) the temporal variability of captan residues, (2) the contribution of several factors to the variability in residues and (3) the relationship between residues and climatic conditions. RESULTS: Initial captan deposits and subsequent residues on fruit and leaves were closer to a lognormal than to a normal distribution. The unit-to-unit variation contributed most to the observed variability in the initial deposit and subsequent residues. Variability due to orchards or trees or tree-zone interactions was also frequently important, but there was no discernable trend in the effects. The variability in residues did not appear to decrease over time. Canopy structure affected greatly the initial deposition but had little direct effect on subsequent captan loss. Fruit and leaves on the outside of the tree canopy received more deposit than those on the inside, but these differences gradually decreased over time. Captan loss resulted mainly from the first rainfall after an application. CONCLUSIONS: Captan loss is mainly due to rain, and the loss is negligible under dry conditions.