1. Artificial neural networks (ANNs), computer programs created based on the biology of functional neural networks, are ideal for analyzing complex clinical problems.
2. ANNs were better than traditional scoring indices and statistical tests at predicting 1-year survival in the setting of brain metastasis.
Evidence Rating Level: 2 (Good)
Study Rundown: Most clinical studies rely on rudimentary clinical tests such as multivariate logistic regression models to assimilate large data sets to make meaningful survival predictions. ANNs applied to these same problems of determining prognosis in the presence of a complex clinical picture outperformed typical statistical tests and prognostic indices in predicting individual outcomes. This is not surprising given that ANNs are modeled based off of biological neural networks and have their inherent capacity to learn and analyze large data sets without prior assumptions or data restrictions. Coupled with increasing amount of data collection and considerable amount of computing power currently available, the layering and continued development of ANN will present a powerful analytical tool for clinicians to utilize in the face of the most complex problems facing medicine today.
In-Depth [multi-institutional retrospective study]: The study pooled 196 patients from a multi-institutional retrospective dataset who underwent primary treatment with stereotactic radiosurgery for newly diagnosed brain metastasis. These patient data sets were then analyzed in smaller subgroups utilizing logistic regression analysis, a single ANN, and an ensemble of 5 ANNs. The Graded Prognostic Assessment (GPA) and Golden Grading Scale (GGS) were also applied to predict 1-year survival. The ensemble of 5 ANNs were significantly more effective at predicting 1-year survival than the multivariate logistic regression model (P = 0.02) and the two grading indices.
By Allen Ho
© 2013 2minutemedicine.com. All rights reserved. No works may be reproduced without written consent from 2minutemedicine.com. Disclaimer: We present factual information directly from peer reviewed medical journals. No post should be construed as medical advice and is not intended as such by the authors or by 2minutemedicine.com. PLEASE SEE A HEALTHCARE PROVIDER IN YOUR AREA IF YOU SEEK MEDICAL ADVICE OF ANY SORT. Content is produced in accordance with fair use copyrights solely and strictly for the purpose of teaching, news and criticism. No benefit, monetary or otherwise, is realized by any participants or the owner of this domain.