Reducing annotation burden in MR: A novel MR-contrast guided contrastive learning approach for image segmentation.

BACKGROUND: Contrastive learning, a successful form of representational learning, has shown promising results in pretraining deep learning (DL) models for downstream tasks. When working with limited annotation data, as in medical image segmentation tasks, learning domain-specific local representations can further improve the performance of DL models. PURPOSE: In this work, we extend the contrastive learning framework to utilize domain-specific contrast information from unlabeled Magnetic Resonance (MR) images to improve the performance of downstream MR image segmentation tasks in the presence of limited labeled data. METHODS: The contrast in MR images is controlled by underlying tissue properties (e.g., T1 or T2) and image acquisition parameters. We hypothesize that learning to discriminate local representations based on underlying tissue properties should improve subsequent segmentation tasks on MR images. We propose a novel constrained contrastive learning (CCL) strategy that uses tissue-specific information via a constraint map to define positive and negative local neighborhoods for contrastive learning, embedding this information in the representational space during pretraining. For a given MR contrast image, the proposed strategy uses local signal characteristics (constraint map) across a set of related multi-contrast MR images as a surrogate for underlying tissue information. We demonstrate the utility of the approach for downstream: (1) multi-organ segmentation tasks in T2-weighted images where a DL model learns T2 information with constraint maps from a set of 2D multi-echo T2-weighted images (n = 101) and (2) tumor segmentation tasks in multi-parametric images from the public brain tumor segmentation (BraTS) (n = 80) dataset where DL models learn T1 and T2 information from multi-parametric BraTS images. Performance is evaluated on downstream multi-label segmentation tasks with limited data in (1) T2-weighted images of the abdomen from an in-house Radial-T2 (Train/Test = 30/20), (2) public Cartesian-T2 (Train/Test = 6/12) dataset, and (3) multi-parametric MR images from the public brain tumor segmentation dataset (BraTS) (Train/Test = 40/50). The performance of the proposed CCL strategy is compared to state-of-the-art self-supervised contrastive learning techniques. In each task, a model is also trained using all available labeled data for supervised baseline performance. RESULTS: The proposed CCL strategy consistently yielded improved Dice scores, Precision, and Recall metrics, and reduced HD95 values across all segmentation tasks. We also observed performance comparable to the baseline with reduced annotation effort. The t-SNE visualization of features for T2-weighted images demonstrates its ability to embed T2 information in the representational space. On the BraTS dataset, we also observed that using an appropriate multi-contrast space to learn T1+T2, T1, or T2 information during pretraining further improved the performance of tumor segmentation tasks. CONCLUSIONS: Learning to embed tissue-specific information that controls MR image contrast with the proposed constrained contrastive learning improved the performance of DL models on subsequent segmentation tasks compared to conventional self-supervised contrastive learning techniques. The use of such domain-specific local representations could help understand, improve performance, and mitigate the scarcity of labeled data in MR image segmentation tasks.

Abdominal T2-Weighted Imaging and T2 Mapping Using a Variable Flip Angle Radial Turbo Spin-Echo Technique.

BACKGROUND: T2 mapping is of great interest in abdominal imaging but current methods are limited by low resolution, slice coverage, motion sensitivity, or lengthy acquisitions. PURPOSE: Develop a radial turbo spin-echo technique with refocusing variable flip angles (RADTSE-VFA) for high spatiotemporal T2 mapping and efficient slice coverage within a breath-hold and compare to the constant flip angle counterpart (RADTSE-CFA). STUDY TYPE: Prospective technical efficacy. SUBJECTS: Testing performed on agarose phantoms and 12 patients. Focal liver lesion classification tested on malignant (N = 24) and benign (N = 11) lesions. FIELD STRENGTH/SEQUENCE: 1.5 T/RADTSE-VFA, RADTSE-CFA. ASSESSMENT: A constrained objective function was used to optimize the refocusing flip angles. Phantom and/or in vivo data were used to assess relative contrast, T2 estimation, specific absorption rate (SAR), and focal liver lesion classification. STATISTICAL TESTS: t-Tests or Mann-Whitney Rank Sum tests were used. RESULTS: Phantom data did not show significant differences in mean relative contrast (P = 0.10) and T2 accuracy (P = 0.99) between RADTSE-VFA and RADTSE-CFA. Adding noise caused T2 overestimation predominantly for RADTSE-CFA and low T2 values. In vivo results did not show significant differences in mean spleen-to-liver (P = 0.62) and kidney-to-liver (P = 0.49) relative contrast between RADTSE-VFA and RADTSE-CFA. Mean T2 values were not significantly different between the two techniques for spleen (T2VFA  = 109.2 ± 12.3 msec; T2CFA  = 110.7 ± 11.1 msec; P = 0.78) and kidney-medulla (T2VFA  = 113.0 ± 8.7 msec; T2CFA  = 114.0 ± 8.6 msec; P = 0.79). Liver T2 was significantly higher for RADTSE-CFA (T2VFA  = 52.6 ± 6.6 msec; T2CFA  = 60.4 ± 8.0 msec) consistent with T2 overestimation in the phantom study. Focal liver lesion classification had comparable T2 distributions for RADTSE-VFA and RADTSE-CFA for malignancies (P = 1.0) and benign lesions (P = 0.39). RADTSE-VFA had significantly lower SAR than RADTSE-CFA increasing slice coverage by 1.5. DATA CONCLUSION: RADTSE-VFA provided noise-robust T2 estimation compared to the constant flip angle counterpart while generating T2-weighted images with comparable contrast. The VFA scheme minimized SAR improving slice efficiency for breath-hold imaging. LEVEL OF EVIDENCE: 2 TECHNICAL EFFICACY STAGE: 1.

Convolutional Neural Network Based Frameworks for Fast Automatic Segmentation of Thalamic Nuclei from Native and Synthesized Contrast Structural MRI.

Thalamic nuclei have been implicated in several neurological diseases. Thalamic nuclei parcellation from structural MRI is challenging due to poor intra-thalamic nuclear contrast while methods based on diffusion and functional MRI are affected by limited spatial resolution and image distortion. Existing multi-atlas based techniques are often computationally intensive and time-consuming. In this work, we propose a 3D convolutional neural network (CNN) based framework for thalamic nuclei parcellation using T1-weighted Magnetization Prepared Rapid Gradient Echo (MPRAGE) images. Transformation of images to an efficient representation has been proposed to improve the performance of subsequent classification tasks especially when working with limited labeled data. We investigate this by transforming the MPRAGE images to White-Matter-nulled MPRAGE (WMn-MPRAGE) contrast, previously shown to exhibit good intra-thalamic nuclear contrast, prior to the segmentation step. We trained two 3D segmentation frameworks using MPRAGE images (n = 35 subjects): (a) a native contrast segmentation (NCS) on MPRAGE images and (b) a synthesized contrast segmentation (SCS) where synthesized WMn-MPRAGE representation generated by a contrast synthesis CNN were used. Thalamic nuclei labels were generated using THOMAS, a multi-atlas segmentation technique proposed for WMn-MPRAGE images. The segmentation accuracy and clinical utility were evaluated on a healthy cohort (n = 12) and a cohort (n = 45) comprising of healthy subjects and patients with alcohol use disorder (AUD), respectively. Both the segmentation CNNs yielded comparable performances on most thalamic nuclei with Dice scores greater than 0.84 for larger nuclei and at least 0.7 for smaller nuclei. However, for some nuclei, the SCS CNN yielded significant improvements in Dice scores (medial geniculate nucleus, P = 0.003, centromedian nucleus, P = 0.01) and percent volume difference (ventral anterior, P = 0.001, ventral posterior lateral, P = 0.01) over NCS. In the AUD cohort, the SCS CNN demonstrated a significant atrophy in ventral lateral posterior nucleus in AUD patients compared to healthy age-matched controls (P = 0.01), agreeing with previous studies on thalamic atrophy in alcoholism, whereas the NCS CNN showed spurious atrophy of the ventral posterior lateral nucleus. CNN-based segmentation of thalamic nuclei provides a fast and automated technique for thalamic nuclei prediction in MPRAGE images. The transformation of images to an efficient representation, such as WMn-MPRAGE, can provide further improvements in segmentation performance.

Auditory and Visual System White Matter Is Differentially Impacted by Normative Aging in Macaques.

Deficits in auditory and visual processing are commonly encountered by older individuals. In addition to the relatively well described age-associated pathologies that reduce sensory processing at the level of the cochlea and eye, multiple changes occur along the ascending auditory and visual pathways that further reduce sensory function in each domain. One fundamental question that remains to be directly addressed is whether the structure and function of the central auditory and visual systems follow similar trajectories across the lifespan or sustain the impacts of brain aging independently. The present study used diffusion magnetic resonance imaging and electrophysiological assessments of auditory and visual system function in adult and aged macaques to better understand how age-related changes in white matter connectivity at multiple levels of each sensory system might impact auditory and visual function. In particular, the fractional anisotropy (FA) of auditory and visual system thalamocortical and interhemispheric corticocortical connections was estimated using probabilistic tractography analyses. Sensory processing and sensory system FA were both reduced in older animals compared with younger adults. Corticocortical FA was significantly reduced only in white matter of the auditory system of aged monkeys, while thalamocortical FA was lower only in visual system white matter of the same animals. Importantly, these structural alterations were significantly associated with sensory function within each domain. Together, these results indicate that age-associated deficits in auditory and visual processing emerge in part from microstructural alterations to specific sensory white matter tracts, and not from general differences in white matter condition across the aging brain.SIGNIFICANCE STATEMENT Age-associated deficits in sensory processing arise from structural and functional alterations to both peripheral sensory organs and central brain regions. It remains unclear whether different sensory systems undergo similar or distinct trajectories in function across the lifespan. To provide novel insights into this question, this study combines electrophysiological assessments of auditory and visual function with diffusion MRI in aged macaques. The results suggest that age-related sensory processing deficits in part result from factors that impact the condition of specific white matter tracts, and not from general decreases in connectivity between sensory brain regions. Such anatomic specificity argues for a framework aimed at understanding vulnerabilities with relatively local influence and brain region specificity.

Auditory Processing Deficits Are Selectively Associated with Medial Temporal Lobe Mnemonic Function and White Matter Integrity in Aging Macaques.

Deficits in auditory function and cognition are hallmarks of normative aging. Recent evidence suggests that hearing-impaired individuals have greater risks of developing cognitive impairment and dementia compared to people with intact auditory function, although the neurobiological bases underlying these associations are poorly understood. Here, a colony of aging macaques completed a battery of behavioral tests designed to probe frontal and temporal lobe-dependent cognition. Auditory brainstem responses (ABRs) and visual evoked potentials were measured to assess auditory and visual system function. Structural and diffusion magnetic resonance imaging were then performed to evaluate the microstructural condition of multiple white matter tracts associated with cognition. Animals showing higher cognitive function had significantly better auditory processing capacities, and these associations were selectively observed with tasks that primarily depend on temporal lobe brain structures. Tractography analyses revealed that the fractional anisotropy (FA) of the fimbria-fornix and hippocampal commissure were associated with temporal lobe-dependent visual discrimination performance and auditory sensory function. Conversely, FA of frontal cortex-associated white matter was not associated with auditory processing. Visual sensory function was not associated with frontal or temporal lobe FA, nor with behavior. This study demonstrates significant and selective relationships between ABRs, white matter connectivity, and higher-order cognitive ability.

Tract-Specific White Matter Correlates of Age-Related Reward Devaluation Deficits in Macaque Monkeys.

AIM: Cognitive aging is known to alter reward-guided behaviors that require interactions between the orbitofrontal cortex (OFC) and amygdala. In macaques, OFC, but not amygdala volumes decline with age and correlate with performance on a reward devaluation (RD) task. The present study used diffusion magnetic resonance imaging (dMRI) methods to investigate whether the condition of the white matter associated with amygdala-OFC connectivity changes with age and relates to reward devaluation. METHODS: Diffusion-, T1- and T2-weighted MRIs were acquired from adult and aged bonnet macaques. Using probabilistic tractography, fractional anisotropy (FA) estimates from two separate white matter tracts associated with amygdala-OFC connectivity, the uncinate fasciculus (UF) and amygdalofugal (AF) pathways, were obtained. Performance measures on RD and reversal learning (RL) tasks were also acquired and related to FA indices from each anatomical tract. RESULTS: Aged monkeys were impaired on both the RD and RL tasks and had lower FA indices in the AF pathway. Higher FA indices from the right hemisphere UF pathway correlated with better performance on an object-based RD task, whereas higher FA indices from the right hemisphere AF were associated with better performance on an object-free version of the task. FA measures from neither tract correlated with RL performance. CONCLUSIONS: These results suggest that the condition of the white matter connecting the amygdala and OFC may impact reward devaluation behaviors. Furthermore, the observation that FA indices from the UF and AF differentially relate to reward devaluation suggests that the amygdala-OFC interactions that occur via these separate tracts are partially independent.