Complexity in Tumour Growth Patterns
Gabriel Landini · Birkhäuser Basel eBooks · 1998
This paper investigates the use of two types of local dimension analysis for objective characterisation of the complexity of epithelial profiles of normal, premalignant and malignant tumour tissues. The results indicate that tumour profile geometry can be characterised quantitatively using fractal concepts and suggest that local fractal dimension analysis provides sufficient information for profile classification purposes. Comparison of the data with an equivalent task carried out by human observers indicates that the characterisation of tumour profiles is of practical use only if unbiased, quantitative methods are applied. These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.