RESUMO
We applied machine learning algorithms for differentiation of posterior fossa tumors using apparent diffusion coefficient (ADC) histogram analysis and structural MRI findings. A total of 256 patients with intra-axial posterior fossa tumors were identified, of whom 248 were included in machine learning analysis, with at least 6 representative subjects per each tumor pathology. The ADC histograms of solid components of tumors, structural MRI findings, and patients' age were applied to construct decision models using Classification and Regression Tree analysis. We also compared different machine learning classification algorithms (i.e., naïve Bayes, random forest, neural networks, support vector machine with linear and polynomial kernel) for dichotomized differentiation of the 5 most common tumors in our cohort: metastasis (n = 65), hemangioblastoma (n = 44), pilocytic astrocytoma (n = 43), ependymoma (n = 27), and medulloblastoma (n = 26). The decision tree model could differentiate seven tumor histopathologies with terminal nodes yielding up to 90% accurate classification rates. In receiver operating characteristics (ROC) analysis, the decision tree model achieved greater area under the curve (AUC) for differentiation of pilocytic astrocytoma (p = 0.020); and atypical teratoid/rhabdoid tumor ATRT (p = 0.001) from other types of neoplasms compared to the official clinical report. However, neuroradiologists' interpretations had greater accuracy in differentiating metastases (p = 0.001). Among different machine learning algorithms, random forest models yielded the highest accuracy in dichotomized classification of the 5 most common tumor types; and in multiclass differentiation of all tumor types random forest yielded an averaged AUC of 0.961 in training datasets, and 0.873 in validation samples. Our study demonstrates the potential application of machine learning algorithms and decision trees for accurate differentiation of brain tumors based on pretreatment MRI. Using easy to apply and understandable imaging metrics, the proposed decision tree model can help radiologists with differentiation of posterior fossa tumors, especially in tumors with similar qualitative imaging characteristics. In particular, our decision tree model provided more accurate differentiation of pilocytic astrocytomas from ATRT than by neuroradiologists in clinical reads.
RESUMO
RATIONALE AND OBJECTIVES: Pituitary macroadenomas are predominantly benign intracranial neoplasms that can be locally aggressive with invasion of adjacent structures. Biomarkers of aggressive behavior have been identified in the pathology literature, including the proliferative marker MIB-1. In the radiology literature, diffusion weighted imaging and low ADC values provide similar markers of aggressive behavior in brain tumors. The purpose of this study was to determine if there is a correlation between ADC and MIB-1 in pituitary macroadenomas. MATERIALS AND METHODS: A retrospective review of diffusion imaging and immunohistochemical characteristics of pituitary macroadenomas was performed. The ADC ratio and specimen Ki-67 (MIB-1) indices were measured. Linear regression analysis of normalized ADC values and MIB-1 indices was used to compare these parameters. RESULTS: There were 17 patients with available ADC maps and MIB-1 indices. Local invasion was confirmed by imaging and intraoperative visualization in 11 patients. The mean ADC ratio for the invasive group was 0.68, with a mean MIB-1 index of 2.21 %. In the noninvasive group, the mean ADC ratio was 1.05, with a mean MIB-1 index of 0.9 %. Linear regression analysis of normalized ADC values versus MIB-1 demonstrates a negative correlation, with a linear slope significantly different from zero (p = 0.003, correlation coefficient of 0.77, and r squared = 0.59). CONCLUSION: We determine a strong correlation of low ADC values and MIB-1, demonstrating the potential of diffusion imaging as a possible biomarker for atypical, proliferative adenomas, which may ultimately affect the surgical approach and postoperative management.