Optimization of Standard Reference Cerebral Ventricular Volumes Measuring Protocol using Magnetic Resonance Imaging Techniques for Clinical Applications

Abstract

Introduction Cerebral ventricular morphology and morphometric changes in response to some ensuing cerebral pathological processes. These changes may be use as an indication to detect pathological processes. Thus, it is imperative to establish a measuring procedure to measure very vital organ dimensions to aid clinal diagnoses and to optimize these procedures for clinical application. In addition to establishing baseline data of ventricular volumes to enable research and clinical comparison, and adopting an easy-to-use graphical user interface for the computation of organ dimensions. Aim The study aimed at establishing standard measuring protocol in addition to baseline reference data using normal Magnetic Resonance Images of the brains and developing a user-friendly graphical user interface for the estimation of ventricular volume for clinical applications. Methodology A retrospective study design approach was adopted, and a convenient sampling technique was used to collect a total of 200 magnetic resonance images and 112 of that met the inclusion criteria were analysed at two imaging facilities in Ghana. Data samples were processed using MeVisLab (MVL) DICOM image analysis and processing software application tools for segmentation and measurements of the images. Minitab statistical tool was used for statistical modeling of ventricular volume models, and script writing (coding) was done using Python application programme and was also used to design and developed a GUI for the ventricular volume calculator. Results The overall mean age for the study is 54 years and ranged between 20-89yrs. The overall mean and S.D of the ventricular volume calculated in the study was 33.60 ± 12.07 cm3. The mean Total Ventricular volume (TVV) per each age group, that is, for 20–29 yrs, 30–39 yrs, 40–49 yrs, 50–59 yrs, 60–69 yrs, 70–79 yrs and 80–89 yrs were; (15.49 ± 2.09cm3), (23.47 ± 2.30cm3), (29.41 ± 2.93cm3), (32.14 ± 1.83cm3), (35.54 ± 1.48cm3), (47.54 ± 2.01cm3) and (51.60 ± 3.66cm3) respectively. Regression linear equations modeled per ventricle were as follows: for left lateral ventricle volume [LLVV (cm3) = 0.95 + 1.89L*W*T], for right lateral ventricle volume [RLVV (cm3) = 2.41 + 1.51L*W*T], for the third ventricle volume [3RDVV (cm3) = 0.81 + 0.77L*W*T] and for the fourth ventricle volume [4THVV (cm3) = 0.88 + 0.95L*W*T]. Conclusion Optimized standard reference dataset of ventricular volumes and sizes of normal brains are crucial baseline determinants for early detection of cerebral diseases and mathematical modeling and GUI holds high prospects in the computation of these organ dimensions in medical imaging.