PREDICTION OF POSTERIOR-FOSSA TUMOR TYPE IN CHILDREN BY MEANS OF MAGNETIC-RESONANCE IMAGE PROPERTIES, SPECTROSCOPY, AND NEURAL NETWORKS

Recent studies have explored characteristics of brain tumors by means of magnetic resonance spectroscopy (MRS) to increase diagnostic accura cy and improve understanding of tumor biology. In this study, a comput er-based neural network was developed to combine MRS data (ratios of N -acetyl-aspartate, choline, and creatine) with 10 characteristics of t umor tissue obtained from magnetic resonance (MR) studies, as well as tumor size and the patient's age and sex, in hopes of further improvin g diagnostic accuracy. Data were obtained in 33 children presenting wi th posterior fossa tumors. The cases were analyzed by a neuroradiologi st, who then predicted the tumor type from among three categories (pri mitive neuroectodermal tumor, astrocytoma, or ependymoma/other) based only on the data obtained via MR imaging. These predictions were compa red with those made by neural networks that had analyzed different com binations of the data. The neuroradiologist correctly predicted the tu mor type in 73% of the cases, whereas four neural networks using diffe rent datasets as inputs were 58 to 95% correct. The neural network tha t used only the three spectroscopy ratios had the least predictive abi lity. With the addition of data including MR imaging characteristics, age, sex, and tumor size, the network's accuracy improved to 72%, cons istent with the predictions of the neuroradiologist who was using the same information. Use of only the analog data (leaving out information obtained from MR imaging), resulted in 88% accuracy. A network that u sed all of the data was able to identify 95% of the tumors correctly. It is concluded that a neural network provided with imaging data, spec troscopic data, and a limited amount of clinical information can predi ct pediatric posterior fossa tumor type with remarkable accuracy.