پیش بینی هیدروسفالی با استفاده از جنگلهای تصادفی
Abstract
1- Introduction
2- Methodology
3- Experiments and results
4- Discussion
5- Conclusion and future work
References
Abstract
Ventricles of the human brain enlarge with aging, neurodegenerative diseases, intrinsic, and extrinsic pathologies. The morphometric examination of neuroimages is an effective approach to assess structural changes occurring due to diseases such as hydrocephalus. In this study, we explored the effectiveness of commonly used morphological parameters in hydrocephalus diagnosis. For this purpose, the effect of six common morphometric parameters; Frontal Horns' Length (FHL), Maximum Lateral Length (MLL), Biparietal Diameter (BPD), Evans' Ratio (ER), Cella Media Ratio (CMR), and Frontal Horns’ Ratio (FHR) were compared in terms of their importance in predicting hydrocephalus using a Random Forest classifier. The experimental results demonstrated that hydrocephalus can be detected with 91.46 % accuracy using all of these measurements. The accuracy of classification using only CMR and FHL reached up to 93.33 %. In terms of individual performances, CMR and FHL were the top performers whereas BPD and FHR did not contribute as much to the overall accuracy.
Introduction
Brain ventricles are dilated with the accumulation of excessive cerebrospinal fluid which leads to a condition known as hydrocephalus. Hydrocephalus affects a wide range of people, from infants to elderly adults. Generally, the ventricular enlargement is measured using parameters derived from the dimensions of the ventricles instead of their actual volumes. Different morphological parameters are used in the literature for the diagnosis of hydrocephalus such as Bicaudate Ratio (BCR), Bifrontal Index (BFI), Bioccipital Index (BOI), Biparietal Diameter (BPD), Cella Media Ratio (CMR), Evans’ Ratio (ER), Minimum Lateral Length (MLL), Third Ventricle Width (TVW), Third Ventricle Sylvian Fissure Ratio Index (TSFI), and Third Ventricle Ratio (TVR) [1, 2]. These parameters are useful not only for the diagnosis and classification of hydrocephalus but also for the follow-up and evaluation of the expansion of the ventricular system after operations such as ventricular shunts [3, 4]. Diagnostic methods for hydrocephalus involve a mixture of clinical and imaging approaches. Accurate and effective evaluation of many CSF-related diseases, especially hydrocephalus, can be performed much faster using new sequences and techniques developed in parallel with the progress in MRI technology. MRI is mostly preferred over CT since it may provide better detail of the borders of ventricles [5, 6]. MRI is helpful in the diagnosis of hydrocephalus and helps in the management and postoperative follow-up of the patients [7, 8]. In this paper, we aim to compare the performances of the above-mentioned parameters in hydrocephalus detection. For this purpose, we trained a random forest classifier to predict hydrocephalus and measure the importance of each parameter. To our knowledge, there is no other study in the literature that compares the performance of these parameters.