Purpose
To develop a deep learning–based reconstruction framework for ultrafast and robust diffusion tensor imaging and fiber tractography.
Methods
SuperDTI was developed to learn the nonlinear ...relationship between DWIs and the corresponding diffusion tensor parameter maps. It bypasses the tensor fitting procedure, which is highly susceptible to noises and motions in DWIs. The network was trained and tested using data sets from the Human Connectome Project and patients with ischemic stroke. Results from SuperDTI were compared against widely used methods for tensor parameter estimation and fiber tracking.
Results
Using training and testing data acquired using the same protocol and scanner, SuperDTI was shown to generate fractional anisotropy and mean diffusivity maps, as well as fiber tractography, from as few as six raw DWIs, with a quantification error of less than 5% in all white‐matter and gray‐matter regions of interest. It was robust to noises and motions in the testing data. Furthermore, the network trained using healthy volunteer data showed no apparent reduction in lesion detectability when directly applied to stroke patient data.
Conclusions
Our results demonstrate the feasibility of superfast DTI and fiber tractography using deep learning with as few as six DWIs directly, bypassing tensor fitting. Such a significant reduction in scan time may allow the inclusion of DTI into the clinical routine for many potential applications.
Purpose
To address motion in cardiac DWI, stimulated‐echo acquisition mode (STEAM) and second‐order motion‐compensated spin‐echo (SE) sequences have been proposed. Despite applying ...motion‐compensation strategies, residual motion can cause misleading signal attenuation. The purpose of this study is to estimate the motion‐induced error in both sequences by analysis of image phase.
Methods
Diffusion‐weighted motion‐compensated SE sequences and STEAM imaging was applied in vivo with diffusion encoding along 3 orthogonal directions. A b‐value range of 100 to 600 s/mm2 and trigger delays of 25%, 50%, and 75% of end systole and middiastole were used. Eddy‐current contributions were obtained from phantom measurements. After computation of motion‐induced phase maps, the amount of signal dephasing was computed from phase gradients, and the resulting errors in diffusion tensor parameters were calculated.
Results
Motion‐induced dephasing from the STEAM sequence showed less dependency on the b‐value and no dependency on the heart phase, whereas SE imaging performed best at 75% end systole followed by 50% end systole and middiastole. For a typical experimental setting, errors of 3.3%/3.0% mean diffusivity, 4.9%/4.8% fractional anisotropy, 2.9º/3.2º helix angulation, 0.8º/0.7º transverse angulation, and 9.9º/10.0º sheet angulation (SE/STEAM) were calculated.
Conclusion
Image phase contains valuable information regarding uncompensated motion and eddy currents in cardiac DTI. Although the trigger delay window for SE is narrower compared with the STEAM‐based approach, imaging in both systole and diastole is feasible and both sequences perform similarly if the trigger delays are selected carefully with SE.
Abstract
In this study, we introduce an interpretable graph-based deep learning prediction model, AttentionSiteDTI, which utilizes protein binding sites along with a self-attention mechanism to ...address the problem of drug–target interaction prediction. Our proposed model is inspired by sentence classification models in the field of Natural Language Processing, where the drug–target complex is treated as a sentence with relational meaning between its biochemical entities a.k.a. protein pockets and drug molecule. AttentionSiteDTI enables interpretability by identifying the protein binding sites that contribute the most toward the drug–target interaction. Results on three benchmark datasets show improved performance compared with the current state-of-the-art models. More significantly, unlike previous studies, our model shows superior performance, when tested on new proteins (i.e. high generalizability). Through multidisciplinary collaboration, we further experimentally evaluate the practical potential of our proposed approach. To achieve this, we first computationally predict the binding interactions between some candidate compounds and a target protein, then experimentally validate the binding interactions for these pairs in the laboratory. The high agreement between the computationally predicted and experimentally observed (measured) drug–target interactions illustrates the potential of our method as an effective pre-screening tool in drug repurposing applications.
Diffusion imaging holds great potential for the non-invasive assessment of the glymphatic system in humans. One technique, diffusion tensor imaging along the perivascular space (DTI-ALPS), has ...introduced the ALPS-index, a novel metric for evaluating diffusivity within the perivascular space. However, it still needs to be established whether the observed reduction in the ALPS-index reflects axonal changes, a common occurrence in neurodegenerative diseases.
To determine whether axonal alterations can influence change in the ALPS-index.
Retrospective.
100 participants (78 cognitively normal and 22 with mild cognitive impairments) aged 50-90 years old.
3T; diffusion-weighted single-shot spin-echo echo-planar imaging sequence, T1-weighted images (MP-RAGE).
The ratio of two radial diffusivities of the diffusion tensor (i.e., λ2/λ3) across major white matter tracts with distinct venous/perivenous anatomy that fulfill (ALPS-tracts) and do not fulfill (control tracts) ALPS-index anatomical assumptions were analyzed.
To investigate the correlation between λ2/λ3 and age/cognitive function (RAVLT) while accounting for the effect of age, linear regression was implemented to remove the age effect from each variable. Pearson correlation analysis was conducted on the residuals obtained from the linear regression. Statistical significance was set at p < 0.05.
λ2 was ~50% higher than λ3 and demonstrated a consistent pattern across both ALPS and control tracts. Additionally, in both ALPS and control tracts a reduction in the λ2/λ3 ratio was observed with advancing age (r = -0.39, r = -0.29, association and forceps tract, respectively) and decreased memory function (r = 0.24, r = 0.27, association and forceps tract, respectively).
The results unveil a widespread radial asymmetry of white matter tracts that changes with aging and neurodegeration. These findings highlight that the ALPS-index may not solely reflect changes in the diffusivity of the perivascular space but may also incorporate axonal contributions.
3 TECHNICAL EFFICACY: Stage 2.
The recently discovered glymphatic system may support the removal of neurotoxic proteins, mainly during sleep, that are associated with neurodegenerative diseases such as Alzheimer's and Parkinson's ...Disease. Diffusion tensor image analysis along the perivascular space (DTI-ALPS) has been suggested as a method to index the health of glymphatic system (with higher values indicating a more intact glymphatic system). Indeed, in small-scale studies the DTI-ALPS index has been shown to correlate with age, cognitive health, and sleep, and is higher in females than males.
To determine whether these relationships are stable we replicated previous findings associating the DTI-ALPS index with demographic, sleep-related, and cognitive markers in a large sample of participants from the UK Biobank.
We calculated the DTI-ALPS index in UK Biobank participants (n = 17723). Using Bayesian and Frequentist analysis approaches, we replicate previously reported relationships between the DTI-ALPS index.
We found the predicted associations between the DTI-ALPS index and age, longest uninterrupted sleep window (LUSWT) on a typical night, cognitive performance, and sex. However, these effects were substantially smaller than those found in previous studies. Parameter estimates from this study may be used as priors in subsequent studies using a Bayesian approach. These results suggest that the DTI-ALPS index is consistently, and therefore predictably, associated with demographics, LUWST, and cognition.
We propose that the metric, calculated for the first time in a large-scale, population-based cohort, is a stable measure, but one for which stronger links to glymphatic system function are needed before it can be used to understand the relationships between glymphatic system function and health outcomes reported in the UK Biobank.
•We calculated an index of diffusion along the perivascular space (DTI-ALPS index) in 44073 participants from the UK Biobank.•We replicated analyses from previous studies showing a relationship between the DTI-ALPS index, sex, age, sleep, & cognition.•Although these replications were broadly successful, the effect sizes were substantially smaller than previously reported.
Neural mechanisms mediating the transition from acute to chronic pain remain largely unknown. In a longitudinal brain imaging study, we followed up patients with a single sub-acute back pain (SBP) ...episode for more than 1 year as their pain recovered (SBPr), or persisted (SBPp) representing a transition to chronic pain. We discovered brain white matter structural abnormalities (n=24 SBP patients; SBPp=12 and SBPr=12), as measured by diffusion tensor imaging (DTI), at entry into the study in SBPp in comparison to SBPr. These white matter fractional anisotropy (FA) differences accurately predicted pain persistence over the next year, which was validated in a second cohort (n=22 SBP patients; SBPp=11 and SBPr=11), and showed no further alterations over a 1-year period. Tractography analysis indicated that abnormal regional FA was linked to differential structural connectivity to medial vs lateral prefrontal cortex. Local FA was correlated with functional connectivity between medial prefrontal cortex and nucleus accumbens in SBPr. As we have earlier shown that the latter functional connectivity accurately predicts transition to chronic pain, we can conclude that brain structural differences, most likely existing before the back pain-inciting event and independent of the back pain, predispose subjects to pain chronification.