Automatic Cerebral Hemisphere Segmentation in Rat MRI with Ischemic Lesions via Attention-based Convolutional Neural Networks.

We present MedicDeepLabv3+, a convolutional neural network that is the first completely automatic method to segment cerebral hemispheres in magnetic resonance (MR) volumes of rats with ischemic lesions. MedicDeepLabv3+ improves the state-of-the-art DeepLabv3+ with an advanced decoder, incorporating spatial attention layers and additional skip connections that, as we show in our experiments, lead to more precise segmentations. MedicDeepLabv3+ requires no MR image preprocessing, such as bias-field correction or registration to a template, produces segmentations in less than a second, and its GPU memory requirements can be adjusted based on the available resources. We optimized MedicDeepLabv3+ and six other state-of-the-art convolutional neural networks (DeepLabv3+, UNet, HighRes3DNet, V-Net, VoxResNet, Demon) on a heterogeneous training set comprised by MR volumes from 11 cohorts acquired at different lesion stages. Then, we evaluated the trained models and two approaches specifically designed for rodent MRI skull stripping (RATS and RBET) on a large dataset of 655 MR rat brain volumes. In our experiments, MedicDeepLabv3+ outperformed the other methods, yielding an average Dice coefficient of 0.952 and 0.944 in the brain and contralateral hemisphere regions. Additionally, we show that despite limiting the GPU memory and the training data, our MedicDeepLabv3+ also provided satisfactory segmentations. In conclusion, our method, publicly available at https://github.com/jmlipman/MedicDeepLabv3Plus , yielded excellent results in multiple scenarios, demonstrating its capability to reduce human workload in rat neuroimaging studies.

Acute Hippocampal Damage as a Prognostic Biomarker for Cognitive Decline but Not for Epileptogenesis after Experimental Traumatic Brain Injury.

It is necessary to develop reliable biomarkers for epileptogenesis and cognitive impairment after traumatic brain injury when searching for novel antiepileptogenic and cognition-enhancing treatments. We hypothesized that a multiparametric magnetic resonance imaging (MRI) analysis along the septotemporal hippocampal axis could predict the development of post-traumatic epilepsy and cognitive impairment. We performed quantitative T2 and T2* MRIs at 2, 7 and 21 days, and diffusion tensor imaging at 7 and 21 days after lateral fluid-percussion injury in male rats. Morris water maze tests conducted between 35-39 days post-injury were used to diagnose cognitive impairment. One-month-long continuous video-electroencephalography monitoring during the 6th post-injury month was used to diagnose epilepsy. Single-parameter and regularized multiple linear regression models were able to differentiate between sham-operated and brain-injured rats. In the ipsilateral hippocampus, differentiation between the groups was achieved at most septotemporal locations (cross-validated area under the receiver operating characteristic curve (AUC) 1.0, 95% confidence interval 1.0-1.0). In the contralateral hippocampus, the highest differentiation was evident in the septal pole (AUC 0.92, 95% confidence interval 0.82-0.97). Logistic regression analysis of parameters imaged at 3.4 mm from the contralateral hippocampus's temporal end differentiated between the cognitively impaired rats and normal rats (AUC 0.72, 95% confidence interval 0.55-0.84). Neither single nor multiparametric approaches could identify the rats that would develop post-traumatic epilepsy. Multiparametric MRI analysis of the hippocampus can be used to identify cognitive impairment after an experimental traumatic brain injury. This information can be used to select subjects for preclinical trials of cognition-improving interventions.

Hippocampal position and orientation as prognostic biomarkers for posttraumatic epileptogenesis: An experimental study in a rat lateral fluid percussion model.

OBJECTIVE: This study was undertaken to identify prognostic biomarkers for posttraumatic epileptogenesis derived from parameters related to the hippocampal position and orientation. METHODS: Data were derived from two preclinical magnetic resonance imaging (MRI) follow-up studies: EPITARGET (156 rats) and Epilepsy Bioinformatics Study for Antiepileptogenic Therapy (EpiBioS4Rx; University of Eastern Finland cohort, 43 rats). Epileptogenesis was induced with lateral fluid percussion-induced traumatic brain injury (TBI) in adult male Sprague Dawley rats. In the EPITARGET cohort, T 2 ∗ -weighted MRI was performed at 2, 7, and 21 days and in the EpiBioS4Rx cohort at 2, 9, and 30 days and 5 months post-TBI. Both hippocampi were segmented using convolutional neural networks. The extracted segmentation mask was used for a geometric construction, extracting 39 parameters that described the position and orientation of the left and right hippocampus. In each cohort, we assessed the parameters as prognostic biomarkers for posttraumatic epilepsy (PTE) both individually, using repeated measures analysis of variance, and in combination, using random forest classifiers. RESULTS: The extracted parameters were highly effective in discriminating between sham-operated and TBI rats in both the EPITARGET and EpiBioS4Rx cohorts at all timepoints (t; balanced accuracy > .9). The most discriminating parameter was the inclination of the hippocampus ipsilateral to the lesion at t = 2 days and the volumes at t ≥ 7 days after TBI. Furthermore, in the EpiBioS4Rx cohort, we could effectively discriminate epileptogenic from nonepileptogenic animals with a longer MRI follow-up, at t = 150 days (area under the curve = .78, balanced accuracy = .80, p = .0050), based on the orientation of both hippocampi. We found that the ipsilateral hippocampus rotated outward on the horizontal plane, whereas the contralateral hippocampus rotated away from the vertical direction. SIGNIFICANCE: We demonstrate that assessment of TBI-induced hippocampal deformation by clinically translatable MRI methodologies detects subjects with prior TBI as well as those at high risk of PTE, paving the way toward subject stratification for antiepileptogenesis studies.

Convolutional Neural Networks Enable Robust Automatic Segmentation of the Rat Hippocampus in MRI After Traumatic Brain Injury.

Registration-based methods are commonly used in the automatic segmentation of magnetic resonance (MR) brain images. However, these methods are not robust to the presence of gross pathologies that can alter the brain anatomy and affect the alignment of the atlas image with the target image. In this work, we develop a robust algorithm, MU-Net-R, for automatic segmentation of the normal and injured rat hippocampus based on an ensemble of U-net-like Convolutional Neural Networks (CNNs). MU-Net-R was trained on manually segmented MR images of sham-operated rats and rats with traumatic brain injury (TBI) by lateral fluid percussion. The performance of MU-Net-R was quantitatively compared with methods based on single and multi-atlas registration using MR images from two large preclinical cohorts. Automatic segmentations using MU-Net-R and multi-atlas registration were of excellent quality, achieving cross-validated Dice scores above 0.90 despite the presence of brain lesions, atrophy, and ventricular enlargement. In contrast, the performance of single-atlas segmentation was unsatisfactory (cross-validated Dice scores below 0.85). Interestingly, the registration-based methods were better at segmenting the contralateral than the ipsilateral hippocampus, whereas MU-Net-R segmented the contralateral and ipsilateral hippocampus equally well. We assessed the progression of hippocampal damage after TBI by using our automatic segmentation tool. Our data show that the presence of TBI, time after TBI, and whether the hippocampus was ipsilateral or contralateral to the injury were the parameters that explained hippocampal volume.

Transfer Learning in Magnetic Resonance Brain Imaging: A Systematic Review.

(1) Background: Transfer learning refers to machine learning techniques that focus on acquiring knowledge from related tasks to improve generalization in the tasks of interest. In magnetic resonance imaging (MRI), transfer learning is important for developing strategies that address the variation in MR images from different imaging protocols or scanners. Additionally, transfer learning is beneficial for reutilizing machine learning models that were trained to solve different (but related) tasks to the task of interest. The aim of this review is to identify research directions, gaps in knowledge, applications, and widely used strategies among the transfer learning approaches applied in MR brain imaging; (2) Methods: We performed a systematic literature search for articles that applied transfer learning to MR brain imaging tasks. We screened 433 studies for their relevance, and we categorized and extracted relevant information, including task type, application, availability of labels, and machine learning methods. Furthermore, we closely examined brain MRI-specific transfer learning approaches and other methods that tackled issues relevant to medical imaging, including privacy, unseen target domains, and unlabeled data; (3) Results: We found 129 articles that applied transfer learning to MR brain imaging tasks. The most frequent applications were dementia-related classification tasks and brain tumor segmentation. The majority of articles utilized transfer learning techniques based on convolutional neural networks (CNNs). Only a few approaches utilized clearly brain MRI-specific methodology, and considered privacy issues, unseen target domains, or unlabeled data. We proposed a new categorization to group specific, widely-used approaches such as pretraining and fine-tuning CNNs; (4) Discussion: There is increasing interest in transfer learning for brain MRI. Well-known public datasets have clearly contributed to the popularity of Alzheimer's diagnostics/prognostics and tumor segmentation as applications. Likewise, the availability of pretrained CNNs has promoted their utilization. Finally, the majority of the surveyed studies did not examine in detail the interpretation of their strategies after applying transfer learning, and did not compare their approach with other transfer learning approaches.

Comparing methods of detecting and segmenting unruptured intracranial aneurysms on TOF-MRAS: The ADAM challenge.

Accurate detection and quantification of unruptured intracranial aneurysms (UIAs) is important for rupture risk assessment and to allow an informed treatment decision to be made. Currently, 2D manual measures used to assess UIAs on Time-of-Flight magnetic resonance angiographies (TOF-MRAs) lack 3D information and there is substantial inter-observer variability for both aneurysm detection and assessment of aneurysm size and growth. 3D measures could be helpful to improve aneurysm detection and quantification but are time-consuming and would therefore benefit from a reliable automatic UIA detection and segmentation method. The Aneurysm Detection and segMentation (ADAM) challenge was organised in which methods for automatic UIA detection and segmentation were developed and submitted to be evaluated on a diverse clinical TOF-MRA dataset. A training set (113 cases with a total of 129 UIAs) was released, each case including a TOF-MRA, a structural MR image (T1, T2 or FLAIR), annotation of any present UIA(s) and the centre voxel of the UIA(s). A test set of 141 cases (with 153 UIAs) was used for evaluation. Two tasks were proposed: (1) detection and (2) segmentation of UIAs on TOF-MRAs. Teams developed and submitted containerised methods to be evaluated on the test set. Task 1 was evaluated using metrics of sensitivity and false positive count. Task 2 was evaluated using dice similarity coefficient, modified hausdorff distance (95th percentile) and volumetric similarity. For each task, a ranking was made based on the average of the metrics. In total, eleven teams participated in task 1 and nine of those teams participated in task 2. Task 1 was won by a method specifically designed for the detection task (i.e. not participating in task 2). Based on segmentation metrics, the top two methods for task 2 performed statistically significantly better than all other methods. The detection performance of the top-ranking methods was comparable to visual inspection for larger aneurysms. Segmentation performance of the top ranking method, after selection of true UIAs, was similar to interobserver performance. The ADAM challenge remains open for future submissions and improved submissions, with a live leaderboard to provide benchmarking for method developments at https://adam.isi.uu.nl/.

Automated joint skull-stripping and segmentation with Multi-Task U-Net in large mouse brain MRI databases.

Skull-stripping and region segmentation are fundamental steps in preclinical magnetic resonance imaging (MRI) studies, and these common procedures are usually performed manually. We present Multi-task U-Net (MU-Net), a convolutional neural network designed to accomplish both tasks simultaneously. MU-Net achieved higher segmentation accuracy than state-of-the-art multi-atlas segmentation methods with an inference time of 0.35 s and no pre-processing requirements. We trained and validated MU-Net on 128 T2-weighted mouse MRI volumes as well as on the publicly available MRM NeAT dataset of 10 MRI volumes. We tested MU-Net with an unusually large dataset combining several independent studies consisting of 1782 mouse brain MRI volumes of both healthy and Huntington animals, and measured average Dice scores of 0.906 (striati), 0.937 (cortex), and 0.978 (brain mask). Further, we explored the effectiveness of our network in the presence of different architectural features, including skip connections and recently proposed framing connections, and the effects of the age range of the training set animals. These high evaluation scores demonstrate that MU-Net is a powerful tool for segmentation and skull-stripping, decreasing inter and intra-rater variability of manual segmentation. The MU-Net code and the trained model are publicly available at https://github.com/Hierakonpolis/MU-Net.

RatLesNetv2: A Fully Convolutional Network for Rodent Brain Lesion Segmentation.

We present a fully convolutional neural network (ConvNet), named RatLesNetv2, for segmenting lesions in rodent magnetic resonance (MR) brain images. RatLesNetv2 architecture resembles an autoencoder and it incorporates residual blocks that facilitate its optimization. RatLesNetv2 is trained end to end on three-dimensional images and it requires no preprocessing. We evaluated RatLesNetv2 on an exceptionally large dataset composed of 916 T2-weighted rat brain MRI scans of 671 rats at nine different lesion stages that were used to study focal cerebral ischemia for drug development. In addition, we compared its performance with three other ConvNets specifically designed for medical image segmentation. RatLesNetv2 obtained similar to higher Dice coefficient values than the other ConvNets and it produced much more realistic and compact segmentations with notably fewer holes and lower Hausdorff distance. The Dice scores of RatLesNetv2 segmentations also exceeded inter-rater agreement of manual segmentations. In conclusion, RatLesNetv2 could be used for automated lesion segmentation, reducing human workload and improving reproducibility. RatLesNetv2 is publicly available at https://github.com/jmlipman/RatLesNetv2.

Performance of Diffusion Kurtosis Imaging Versus Diffusion Tensor Imaging in Discriminating Between Benign Tissue, Low and High Gleason Grade Prostate Cancer.

RATIONALE AND OBJECTIVES: To investigate the performance of diffusion kurtosis imaging (DKI) and diffusion tensor imaging (DTI) in discriminating benign tissue, low- and high-grade prostate adenocarcinoma (PCa). MATERIALS AND METHODS: Forty-eight patients with biopsy-proven PCa of different Gleason grade (GG), who provided written informed consent, were enrolled. All subjects underwent 3T DWI examinations by using b values 0, 500, 1000, 1500, 2000, and 2500 s/mm2 and six gradient directions. Mean diffusivity, fractional anisotropy (FA), apparent kurtosis (K), apparent kurtosis-derived diffusivity (D), and proxy fractional kurtosis anisotropy (KFA) maps were obtained. Regions of interest were selected in PCa, in the contralateral benign zone, and in the peritumoral area. Histogram analysis was performed by measuring mean, 10th, 25th, and 90th (p90) percentile of the whole-lesion volume. Kruskal-Wallis test with Bonferroni correction was used to assess significant differences between different regions of interest. The correlation between diffusion metrics and GG and between DKI and DTI parameters was evaluated with Pearson's test. ROC curve analysis was carried out to analyze the ability of histogram variables to differentiate low- and high-GG PCa. RESULTS: All metrics significantly discriminated PCa from benign and from peritumoral tissue (except for K, KFAp90, and FA). Kp90 showed the highest correlation with GG and the best diagnostic ability (area under the curve = 0.84) in discriminating low- from high-risk PCa. CONCLUSION: Compared to DTI, DKI provides complementary and additional information about prostate cancer tissue, resulting more sensitive to PCa-derived modifications and more accurate in discriminating low- and high-risk PCa.