Tractography-based targeting of ventral intermediate nucleus: A comparison between conventional stereotactic targeting and diffusion tensor imaging-based targeting

BackgroundThe ventral intermediate (VIM) nucleus of the thalamus is the main target for lesioning using magnetic resonance imaging (MRI) guided focused ultrasound (MRgFUS) or deep brain stimulation (DBS). Targeting of VIM still depends on standard stereotactic coordinates, which do not account for inter-individual variability. Several approaches have been proposed including visualization of dentato-rubro-thalamic tract (DRTT) using diffusion tensor imaging tractography.ObjectiveTo compare probabilistic tracking of DRTT with deterministic tracking of DRTT and stereotactic coordinates to identify the most appropriate approach to target VIM.MethodsIn this retrospective study, we assessed the VIM targeted using stereotactic coordinates, deterministic and probabilistic tracking of DRTT in 19 patients with essential tremor who underwent DBS with VIM targeted using microelectrode recordings. We subsequently determined the positions of VIM derived from these three approaches and compared with that of DBS lead position using paired sample t-tests.ResultsThe probabilistic tracking of DRTT was significantly anterior to the lead (1.45 ± 1.61 mm (P< 0.0001)), but not in the medial/lateral position (−0.29±2.42 mm (P=0.50)). Deterministic tracking of DRTT was significantly lateral (2.16 ± 1.94 mm (P< 0.0001)) and anterior to the lead (1.66 ± 2.1 mm (P< 0.0001)). The stereotactic coordinates were significantly lateral (2.41 ± 1.41 mm (P< 0.0001)) and anterior (1.23 ± 0.89 mm (P< 0.0001)) to the lead.ConclusionProbabilistic tracking of DRTT was found to be superior in targeting VIM compared to deterministic tracking and stereotactic coordinates.

Multishell Diffusion MRI Reflects Improved Physical Fitness Induced by Dance Intervention

Using multishell diffusion MRI and both tract-based spatial statistics (TBSS) and probabilistic tracking of specific tracts of interest, we evaluated the neural underpinnings of the impact of a six-month dance intervention (DI) on physical fitness and cognitive outcomes in nondemented seniors. The final cohort had 76 nondemented seniors, randomized into DI and control (life as usual) groups. Significant effects were observed between the DI and control groups in physical fitness measures and in attention. We detected associations between improved physical fitness and changes in diffusion tensor imagining (DTI) measures in the whole white matter (WM) skeleton and in the corticospinal tract and the superior longitudinal fascicle despite the fact that no significant differences in changes to the WM microstructure were found between the two groups.

Long-Term Tracking of Group-Housed Livestock Using Keypoint Detection and MAP Estimation for Individual Animal Identification

Tracking individual animals in a group setting is a exigent task for computer vision and animal science researchers. When the objective is months of uninterrupted tracking and the targeted animals lack discernible differences in their physical characteristics, this task introduces significant challenges. To address these challenges, a probabilistic tracking-by-detection method is proposed. The tracking method uses, as input, visible keypoints of individual animals provided by a fully-convolutional detector. Individual animals are also equipped with ear tags that are used by a classification network to assign unique identification to instances. The fixed cardinality of the targets is leveraged to create a continuous set of tracks and the forward-backward algorithm is used to assign ear-tag identification probabilities to each detected instance. Tracking achieves real-time performance on consumer-grade hardware, in part because it does not rely on complex, costly, graph-based optimizations. A publicly available, human-annotated dataset is introduced to evaluate tracking performance. This dataset contains 15 half-hour long videos of pigs with various ages/sizes, facility environments, and activity levels. Results demonstrate that the proposed method achieves an average precision and recall greater than 95% across the entire dataset. Analysis of the error events reveals environmental conditions and social interactions that are most likely to cause errors in real-world deployments.

Combining Denoising Autoencoders and Dynamic Programming for Acoustic Detection and Tracking of Underwater Moving Targets

Accurate detection and tracking of moving targets in underwater environments pose significant challenges, because noise in acoustic measurements (e.g., SONAR) makes the signal highly stochastic. In continuous marine monitoring a further challenge is related to the computational complexity of the signal processing pipeline—due to energy constraints, in off-shore monitoring platforms algorithms should operate in real time with limited power consumption. In this paper, we present an innovative method that allows to accurately detect and track underwater moving targets from the reflections of an active acoustic emitter. Our system is based on a computationally- and energy-efficient pre-processing stage carried out using a deep convolutional denoising autoencoder (CDA), whose output is then fed to a probabilistic tracking method based on the Viterbi algorithm. The CDA is trained on a large database of more than 20,000 reflection patterns collected during 50 designated sea experiments. System performance is then evaluated on a controlled dataset, for which ground truth information is known, as well as on recordings collected during different sea experiments. Results show that, compared to the benchmark, our method achieves a favorable trade-off between detection and false alarm rate, as well as improved tracking accuracy.