Contributions
Abstract: EP1545
Type: ePoster
Abstract Category: Pathology and pathogenesis of MS - 21 Imaging
Introduction: This study used the commonly available automated tools of SIENAX from FSL, Lesion Segmentation tool (LST) for SPM, and NeuroQuant including its lesion segmentation module LesionQuant, from CorTechs Labs, Inc. to determine lesion and brain volumes in a group of MS patients. These results were compared to expert manual lesion segmentations. SIENAX and LST are research tools, widely used in the research community. NeuroQuant is an FDA cleared, fully automated brain segmentation and volume measurement tool for clinical use.
Methods: T1-weighted and FLAIR images from 24 MS (19F, 5M, 42 ± 12 yo) patients were included in the study. Total brain, gray matter and white matter volumes were measured using SIENAX and NeuroQuant. The lesion volumes were calculated using LST and the LesionQuant module of NeuroQuant. Lesions were also manually segmented for each patient. Manually determined lesion volumes were compared with the results of the automated tools. The Pearson correlation coefficient was used in comparing brain volume measurements of SIENAX and NeuroQuant, as well as comparing the lesion volumes from LST, LesionQuant and manual segmentation.
Results: The total brain volumes calculated with SIENAX and NeuroQuant were correlated (R=0.83). LST and LesionQuant lesion volume determinations were highly correlated with each other and with manually determined lesion volumes (all R=0.97).
Conclusions: The two automated lesion determination tools, LST and LesionQuant, have comparable performances and their results are highly correlated with the manually segmented lesion volumes. The whole brain volume determinations using SIENAX and NeuroQuant also produced results similar to each other and can be used to quantify brain atrophy in MS patients.
Disclosure: Authors are the employees of CorTechs Labs, Inc.
Abstract: EP1545
Type: ePoster
Abstract Category: Pathology and pathogenesis of MS - 21 Imaging
Introduction: This study used the commonly available automated tools of SIENAX from FSL, Lesion Segmentation tool (LST) for SPM, and NeuroQuant including its lesion segmentation module LesionQuant, from CorTechs Labs, Inc. to determine lesion and brain volumes in a group of MS patients. These results were compared to expert manual lesion segmentations. SIENAX and LST are research tools, widely used in the research community. NeuroQuant is an FDA cleared, fully automated brain segmentation and volume measurement tool for clinical use.
Methods: T1-weighted and FLAIR images from 24 MS (19F, 5M, 42 ± 12 yo) patients were included in the study. Total brain, gray matter and white matter volumes were measured using SIENAX and NeuroQuant. The lesion volumes were calculated using LST and the LesionQuant module of NeuroQuant. Lesions were also manually segmented for each patient. Manually determined lesion volumes were compared with the results of the automated tools. The Pearson correlation coefficient was used in comparing brain volume measurements of SIENAX and NeuroQuant, as well as comparing the lesion volumes from LST, LesionQuant and manual segmentation.
Results: The total brain volumes calculated with SIENAX and NeuroQuant were correlated (R=0.83). LST and LesionQuant lesion volume determinations were highly correlated with each other and with manually determined lesion volumes (all R=0.97).
Conclusions: The two automated lesion determination tools, LST and LesionQuant, have comparable performances and their results are highly correlated with the manually segmented lesion volumes. The whole brain volume determinations using SIENAX and NeuroQuant also produced results similar to each other and can be used to quantify brain atrophy in MS patients.
Disclosure: Authors are the employees of CorTechs Labs, Inc.