Multi-atlas label fusion with random local binary pattern features: Application to hippocampus segmentation

Abstract

Automatic and reliable segmentation of the hippocampus from magnetic resonance (MR) brain images is extremely important in a variety of neuroimage studies. To improve the hippocampus segmentation performance, a local binary pattern based feature extraction method is developed for machine learning based multi-atlas hippocampus segmentation. Under the framework of multi-atlas image segmentation (MAIS), a set of selected atlases are registered to images to be segmented using a non-linear image registration algorithm. The registered atlases are then used as training data to build linear regression models for segmenting the images based on the image features, referred to as random local binary pattern (RLBP), extracted using a novel image feature extraction method. The RLBP based MAIS algorithm has been validated for segmenting hippocampus based on a data set of 135 T1 MR images which are from the Alzheimer’s Disease Neuroimaging Initiative database (adni.loni.usc.edu). By using manual segmentation labels produced by experienced tracers as the standard of truth, six segmentation evaluation metrics were used to evaluate the image segmentation results by comparing automatic segmentation results with the manual segmentation labels. We further computed Cohen’s d effect size to investigate the sensitivity of each segmenting method in detecting volumetric differences of the hippocampus between different groups of subjects. The evaluation results showed that our method was competitive to state-of-the-art label fusion methods in terms of accuracy. Hippocampal volumetric analysis showed that the proposed RLBP method performed well in detecting the volumetric differences of the hippocampus between groups of Alzheimer’s disease patients, mild cognitive impairment subjects, and normal controls. These results have demonstrated that the RLBP based multi-atlas image segmentation method could facilitate efficient and accurate extraction of the hippocampus and may help predict Alzheimer’s disease. The codes of the proposed method is available (https://www.nitrc.org/frs/?group_id=1242).

Hippocampal Segmentation in Brain MRI Images Using Machine Learning Methods: A Survey

“…To reduce the influence of redundant features on the similarity measurement, Pang et al [ 32 ] used dictionary learning to remove redundant information. They used low‐dimensional feature information to construct the Distance field (DF), and the similarity is calculated according to the distance information provided by the DF.…”

“…To improve the performance of image segmentation based on multiple atlases, some researchers have proposed improved image registration strategies [ 28 – 30 ] , while more studies focus on label fusion. For example, by introducing dictionary learning [ 31 ] , metric learning [ 32 ] , manifold learning [ 33 , 34 ] , and other machine learning technologies into label fusion, new label fusion strategies have been improved and proposed. In this paper, this is called the label‐fusion‐based segmentation method.…”

“…Some researchers have transformed the hippocampal segmentation task into a pattern recognition task and performed hippocampal segmentation with feature extraction and classification models [ 35 , 36 ] . By introducing techniques such as multiscale and sparse representation to improve the feature representation or adding richer information such as context information [ 32 , 37 ] , these researchers can obtain a better classifier and improve the effect of hippocampal segmentation. In this paper, this is called the classification‐based segmentation method.…”

Challenges for machine learning in clinical translation of big data imaging studies

“…Other methods to overcome the lack of available labels focus on working with approximations for labels, which are cheaper to acquire (Tajbakhsh et al, 2020). For instance, many methods propose pre-training the network using auxiliary labels generated using automatic tools (e.g., traditional image segmentation methods) and then fine-tuning the model on the small number of manual labels (Guha Roy et al, 2018;Wang et al, 2020), or registration of an atlas to propagate labels from the atlas to the subject space (Zhu et al, 2019). Other approaches are weakly supervised, utilising quick annotations such as image level labels (Feng et al, 2017), bounding box annotations (Rajchl et al, 2017), scribbles (Dorent et al, 2020;Luo et al, 2021) or point labels (McEver and Manjunath, 2020).…”

Combination of convolutional networks with a learning based ensemble strategy for hippocampus subfield segmentation

“…Thus, automatic segmentation of hippocampal subfields is desirable, especially for large-scale disease studies. Among the automatic image segmentation methods, multi-atlas based image segmentation (MAIS) methods have been applied in both the hippocampal segmentation and hippocampal subfield segmentation [4][5][6][7] . In the MAIS methods, all atlas images are first registered to a target image.…”