Machine learning (ML) is a set of mathematical technologies allowing computers to perform tasks without using explicit instructions relying on patterns and inference instead. These tecnologies have many potential applications in medicine but have been used also frequently for tasks they are not suited to or based on limited datasets that do not allow their proper applicability.
A group of MOLAB researchers compared the predictive performance of these methods to human knowledge-based approaches. 404 GBM patients were included (311 discovery and 93 validation). 16 morphological and 28 textural descriptors were obtained from pretreatment volumetric postcontrast T1-weighted magnetic resonance images. Different prognostic ML methods were developed. An optimized linear prognostic model (OLPM) was also built using the four significant non-correlated parameters with individual prognosis value. oLpM achieved high prognostic value (validation c-index = 0.817) and outperformed ML models based on either the same parameter set or on the full set of 44 attributes considered. Neural networks with cross-validation-optimized attribute selection achieved comparable results (validation c-index = 0.825). ML models using only the four outstanding parameters obtained better results than their counterparts based on all the attributes, which presented overfitting.
In conclusion, human-based methods studied provided the most accurate survival predictors for glioblastoma to date, due to a combination of the strength of the methodology, the quality and volume of the data used and the careful attribute selection. the ML methods studied suffered overfitting and lost prognostic value when the number of parameters was increased. This study emphasises the use of ML methods only were appropriate and with a good fundamental knowledge of their limitations.
Prognostic models based on imaging findings in glioblastoma: Human versus Machine
D. Molina-García, L. Vera, J. Pérez-Beteta, E. Arana, V. M. Pérez-García
Scientific Reports 9:5982 (2019)