Automated and manual hippocampal segmentation techniques: Comparison of results, reproducibility and clinical applicability

Abstract

BackgroundImaging techniques used to measure hippocampal atrophy are key to understanding the clinical progression of Alzheimer's disease (AD). Various semi-automated hippocampal segmentation techniques are available and require human expert input to learn how to accurately segment new data. Our goal was to compare 1) the performance of our automated hippocampal segmentation technique relative to manual segmentations, and 2) the performance of our automated technique when provided with a training set from two different raters. We also explored the ability of hippocampal volumes obtained using manual and automated hippocampal segmentations to predict conversion from MCI to AD.MethodsWe analyzed 161 1.5 T T1-weighted brain magnetic resonance images (MRI) from the ADCS Donepezil/Vitamin E clinical study. All subjects carried a diagnosis of mild cognitive impairment (MCI). Three different segmentation outputs (one produced by manual tracing and two produced by a semi-automated algorithm trained with training sets developed by two raters) were compared using single measure intraclass correlation statistics (smICC). The radial distance method was used to assess each segmentation technique's ability to detect hippocampal atrophy in 3D. We then compared how well each segmentation method detected baseline hippocampal differences between MCI subjects who remained stable (MCInc) and those who converted to AD (MCIc) during the trial. Our statistical maps were corrected for multiple comparisons using permutation-based statistics with a threshold of p < .01.ResultsOur smICC analyses showed significant agreement between the manual and automated hippocampal segmentations from rater 1 [right smICC = 0.78 (95%CI 0.72–0.84); left smICC = 0.79 (95%CI 0.72–0.85)], the manual segmentations from rater 1 versus the automated segmentations from rater 2 [right smICC = 0.78 (95%CI 0.7–0.84); left smICC = 0.78 (95%CI 0.71–0.84)], and the automated segmentations of rater 1 versus rater 2 [right smICC = 0.97 (95%CI 0.96–0.98); left smICC = 0.97 (95%CI 0.96–0.98)]. All three segmentation methods detected significant CA1 and subicular atrophy in MCIc compared to MCInc at baseline (manual: right pcorrected = 0.0112, left pcorrected = 0.0006; automated rater 1: right pcorrected = 0.0318, left pcorrected = 0.0302; automated rater 2: right pcorrected = 0.0029, left pcorrected = 0.0166).ConclusionsThe hippocampal volumes obtained with a fast semi-automated segmentation method were highly comparable to the ones obtained with the labor-intensive manual segmentation method. The AdaBoost automated hippocampal segmentation technique is highly reliable allowing the efficient analysis of large data sets.

Imaging biomarkers in neurodegeneration: current and future practices

“…The scatter plot (lower left) shows the correlation of regional Flortaucipir SUVR and object activation in posterior-medial regions. See Maass et al [158] for study details longitudinal studies analysed the progression of AD [188][189][190] using structural MRI and deep learning (DL) algorithms such as recurrent neural networks (RNNs) and variations of long short-term memory networks (LSTMN). The most common feature in order to study disease progression is hippocampal volume.…”

Deep Learning Segmentation of Triple-Negative Breast Cancer (TNBC) Patient Derived Tumor Xenograft (PDX) and Sensitivity of Radiomic Pipeline to Tumor Probability Boundary

“…Our approach mimics the clinical scanning protocols where multi-contrast MR are used by radiologists to assess tumor boundaries. One principal issue with manual tumor delineation is the variability of delineation among multiple experts, which leads to lack of reproducibility [46,47]. In this study, as expected, the tumor volume differed substantially between the experts, demonstrating the need to develop a reproducible pipeline.…”

Imaging Techniques in Alzheimer’s Disease: A Review of Applications in Early Diagnosis and Longitudinal Monitoring

“…Lastly, manual segmentation of T1-weighted images is time-consuming [ 53 ], requires specialistic training and can result in high levels of variability in the measurements [ 54 ], due to different protocols for assessing the measurements [ 33 ]. Fortunately, in the last decade, much effort has been put into establishing methods for automated segmentation, resulting in more accurate data from MRI images in less time [ 53 , 55 , 56 , 57 ]. One major drawback of structural MRI in general is the impossibility to directly observe the effect of amyloid plaques or NFTs in the brain.…”