RESUMEN
BACKGROUND: Multiple sclerosis (MS) is a neurologic disease of the central nervous system which affects almost three million people worldwide. MS is characterized by a demyelination process that leads to brain lesions, allowing these affected areas to be visualized with magnetic resonance imaging (MRI). Deep learning techniques, especially computational algorithms based on convolutional neural networks (CNNs), have become a frequently used algorithm that performs feature self-learning and enables segmentation of structures in the image useful for quantitative analysis of MRIs, including quantitative analysis of MS. To obtain quantitative information about lesion volume, it is important to perform proper image preprocessing and accurate segmentation. Therefore, we propose a method for volumetric quantification of lesions on MRIs of MS patients using automatic segmentation of the brain and lesions by two CNNs. METHODS: We used CNNs at two different moments: the first to perform brain extraction, and the second for lesion segmentation. This study includes four independent MRI datasets: one for training the brain segmentation models, two for training the lesion segmentation model, and one for testing. RESULTS: The proposed brain detection architecture using binary cross-entropy as the loss function achieved a 0.9786 Dice coefficient, 0.9969 accuracy, 0.9851 precision, 0.9851 sensitivity, and 0.9985 specificity. In the second proposed framework for brain lesion segmentation, we obtained a 0.8893 Dice coefficient, 0.9996 accuracy, 0.9376 precision, 0.8609 sensitivity, and 0.9999 specificity. After quantifying the lesion volume of all patients from the test group using our proposed method, we obtained a mean value of 17,582 mm3. CONCLUSIONS: We concluded that the proposed algorithm achieved accurate lesion detection and segmentation with reproducibility corresponding to state-of-the-art software tools and manual segmentation. We believe that this quantification method can add value to treatment monitoring and routine clinical evaluation of MS patients.
RESUMEN
It is estimated that multiple sclerosis (MS) affects 35,000 Brazilians and 2.5 million individuals worldwide. Many studies have suggested a possible role of metallic elements in the etiology of MS, but their concentration in the blood of MS patients is nonetheless little investigated in Brazil. In this work, these elements were studied through Inductively Coupled Plasma Mass Spectrometry (ICP-MS), whose analysis provides a tool to quantify the concentrations of metal elements in the blood samples of individuals with neurodegenerative disorders. This study aimed to compare the concentration of metallic elements in blood samples from patients with MS and healthy individuals. Blood was collected from 30 patients with multiple sclerosis and compared with the control group. Blood samples were digested in closed vessels using a microwave and ICP-MS was used to determine the concentrations of 12 metallic elements (Ba, Be, Ca, Co, Cr, Cu, Fe, Mg, Mo, Ni, Pb and Zn). In MS patients, we observed a reduction in the concentrations of beryllium, copper, chromium, cobalt, nickel, magnesium and iron. The mean concentration of lead in blood was significantly elevated in the MS group. However, no difference was observed in the concentrations of Mo, Ba, Ca and Zn in blood samples from MS patients and the control group. According to our data, there is a possible role for the concentrations of 8 of the 12 evaluated metallic elements in multiple sclerosis. Abnormalities in transition metals levels in biological matrices have been reported in several neurological diseases.
Asunto(s)
Espectrometría de Masas , Metales/sangre , Esclerosis Múltiple/sangre , Adulto , Ambiente , Femenino , Humanos , Masculino , Persona de Mediana EdadRESUMEN
Brain volume measurements are becoming an important tool for assessing success in controlling multiple sclerosis (MS) activity. MSmetrix (icometrix) is an easy-to-use platform, specific for MS magnetic resonance imaging (MRI) of the brain. It provides data on total brain volume, grey matter volume and lesion load volume. The objective of the present study was to assess whether disability and the number of relapses during the previous year correlated with brain volume measurements from MSmetrix. Data on 185 icometrix reports from patients with MS were used to evaluate the potential correlation between brain volume measurements and clinical parameters. There was a significant correlation between higher disability and decreased brain volume (total and grey matter). Increased lesion load in the brain and higher number of relapses in the previous year were also independently correlated with decreased brain tissue volume and with increased disability. This is the first study with real-world data to show that icometrix is a relevant tool for the study of brain volume loss in MS.