Brain MR Image Enhancement for Tumor Segmentation Using 3D U-Net

MRI images are visually inspected by domain experts for the analysis and quantification of the tumorous tissues. Due to the large volumetric data, manual reporting on the images is subjective, cumbersome, and error prone. To address these problems, automatic image analysis tools are employed for tumor segmentation and other subsequent statistical analysis. However, prior to the tumor analysis and quantification, an important challenge lies in the pre-processing. In the present study, permutations of different pre-processing methods are comprehensively investigated. In particular, the study focused on Gibbs ringing artifact removal, bias field correction, intensity normalization, and adaptive histogram equalization (AHE). The pre-processed MRI data is then passed onto 3D U-Net for automatic segmentation of brain tumors. The segmentation results demonstrated the best performance with the combination of two techniques, i.e., Gibbs ringing artifact removal and bias-field correction. The proposed technique achieved mean dice score metrics of 0.91, 0.86, and 0.70 for the whole tumor, tumor core, and enhancing tumor, respectively. The testing mean dice scores achieved by the system are 0.90, 0.83, and 0.71 for the whole tumor, core tumor, and enhancing tumor, respectively. The novelty of this work concerns a robust pre-processing sequence for improving the segmentation accuracy of MR images. The proposed method overcame the testing dice scores of the state-of-the-art methods. The results are benchmarked with the existing techniques used in the Brain Tumor Segmentation Challenge (BraTS) 2018 challenge.

Impact of Preprocessing and Harmonization Methods on the Removal of Scanner Effects in Brain MRI Radiomic Features

In brain MRI radiomics studies, the non-biological variations introduced by different image acquisition settings, namely scanner effects, affect the reliability and reproducibility of the radiomics results. This paper assesses how the preprocessing methods (including N4 bias field correction and image resampling) and the harmonization methods (either the six intensity normalization methods working on brain MRI images or the ComBat method working on radiomic features) help to remove the scanner effects and improve the radiomic feature reproducibility in brain MRI radiomics. The analyses were based on in vitro datasets (homogeneous and heterogeneous phantom data) and in vivo datasets (brain MRI images collected from healthy volunteers and clinical patients with brain tumors). The results show that the ComBat method is essential and vital to remove scanner effects in brain MRI radiomic studies. Moreover, the intensity normalization methods, while not able to remove scanner effects at the radiomic feature level, still yield more comparable MRI images and improve the robustness of the harmonized features to the choice among ComBat implementations.

DeepLab and Bias Field Correction Based Automatic Cone Photoreceptor Cell Identification with Adaptive Optics Scanning Laser Ophthalmoscope Images

The identification of cone photoreceptor cells is important for early diagnosing of eye diseases. We proposed automatic deep-learning cone photoreceptor cell identification on adaptive optics scanning laser ophthalmoscope images. The proposed algorithm is based on DeepLab and bias field correction. Considering manual identification as reference, our algorithm is highly effective, achieving precision, recall, and F 1 score of 96.7%, 94.6%, and 95.7%, respectively. To illustrate the performance of our algorithm, we present identification results for images with different cone photoreceptor cell distributions. The experimental results show that our algorithm can achieve accurate photoreceptor cell identification on images of human retinas, which is comparable to manual identification.

Research on ARKFCM Algorithm Based on Membership Constraint and Bias Field Correction in Neonatal HIE Image Segmentation Method

First, this paper presents the algorithm of adaptively regularized kernel-based fuzzy C-means based on membership constraint (G-ARKFCM). Under the idea of competitive learning based on penalizing opponents, a new membership constraint function penalty item is introduced for each sample point in the segmented image, so that the ARKFCM algorithm is no longer limited to the fuzzy index m = 2. Secondly, the multiplicative intrinsic component optimization (MICO) is introduced into G-ARKFCM to obtain the GM-ARKFCM algorithm, which can correct the bias field when segmenting neonatal HIE images. Compared with other algorithms, the GM-ARKFCM algorithm has better segmentation quality and robustness. The GM-ARKFCM algorithm can more completely segment the neonatal ventricles and surrounding white matter and can retain more information of the original image.

Robust Finger Vein Recognition Based on Deep CNN with Spatial Attention and Bias Field Correction

As one of the biometric information based authentication technologies, finger vein recognition has received increasing attention due to its safety and convenience. However, it is still a challenging task to design an efficient and robust finger vein recognition system because of the low quality of the finger vein images, lack of sufficient number of training samples with image-level annotated information and no pixel-level finger vein texture labels in the public available finger vein databases. In this paper, we propose a novel CNN-based finger vein recognition approach with bias field correction, spatial attention mechanism and a multistage transfer learning strategy to cope with the difficulties mentioned above. In the proposed method, the bias field correction module is to remove the unbalanced bias field of the original images by using a two-dimensional polynomial fitting algorithm, the spatial attention module is to enhance the informative vein texture regions while suppressing the other less informative regions, and the multistage transfer learning strategy is to solve the problem caused by insufficient training for CNN-based model due to lack of labeled training samples in the public finger vein databases. Moreover, several measures, including a label smoothing scheme and data augmentation, are exploited to improve the performance of the proposed method. Extensive experiments have been conducted in the work on three public databases, and the results show that the proposed approach outperforms the existing state-of-the-art methods.

Quantitative Multi-Parameter Mapping Optimized for the Clinical Routine

Using quantitative multi-parameter mapping (MPM), studies can investigate clinically relevant microstructural changes with high reliability over time and across subjects and sites. However, long acquisition times (20 min for the standard 1-mm isotropic protocol) limit its translational potential. This study aimed to evaluate the sensitivity gain of a fast 1.6-mm isotropic MPM protocol including post-processing optimized for longitudinal clinical studies. 6 healthy volunteers (35±7 years old; 3 female) were scanned at 3T to acquire the following whole-brain MPM maps with 1.6 mm isotropic resolution: proton density (PD), magnetization transfer saturation (MT), longitudinal relaxation rate (R1), and transverse relaxation rate (R2*). MPM maps were generated using two RF transmit field (B1+) correction methods: (1) using an acquired B1+ map and (2) using a data-driven approach. Maps were generated with and without Gibb's ringing correction. The intra-/inter-subject coefficient of variation (CoV) of all maps in the gray and white matter, as well as in all anatomical regions of a fine-grained brain atlas, were compared between the different post-processing methods using Student's t-test. The intra-subject stability of the 1.6-mm MPM protocol is 2–3 times higher than for the standard 1-mm sequence and can be achieved in less than half the scan duration. Intra-subject variability for all four maps in white matter ranged from 1.2–5.3% and in gray matter from 1.8 to 9.2%. Bias-field correction using an acquired B1+ map significantly improved intra-subject variability of PD and R1 in the gray (42%) and white matter (54%) and correcting the raw images for the effect of Gibb's ringing further improved intra-subject variability in all maps in the gray (11%) and white matter (10%). Combining Gibb's ringing correction and bias field correction using acquired B1+ maps provides excellent stability of the 7-min MPM sequence with 1.6 mm resolution suitable for the clinical routine.