Real-time estimation of the optimal coil placement in transcranial magnetic stimulation using multi-task deep learning.

Transcranial magnetic stimulation (TMS) has emerged as a promising neuromodulation technique with both therapeutic and diagnostic applications. As accurate coil placement is known to be essential for focal stimulation, computational models have been established to help find the optimal coil positioning by maximizing electric fields at the cortical target. While these numerical simulations provide realistic and subject-specific field distributions, they are computationally demanding, precluding their use in real-time applications. In this paper, we developed a novel multi-task deep neural network which simultaneously predicts the optimal coil placement for a given cortical target as well as the associated TMS-induced electric field. Trained on large amounts of preceding numerical optimizations, the Attention U-Net-based neural surrogate provided accurate coil optimizations in only 35 ms, a fraction of time compared to the state-of-the-art numerical framework. The mean errors on the position estimates were below 2 mm, i.e., smaller than previously reported manual coil positioning errors. The predicted electric fields were also highly correlated (r> 0.97) with their numerical references. In addition to healthy subjects, we validated our approach also in glioblastoma patients. We first statistically underlined the importance of using realistic heterogeneous tumor conductivities instead of simply adopting values from the surrounding healthy tissue. Second, applying the trained neural surrogate to tumor patients yielded similar accurate positioning and electric field estimates as in healthy subjects. Our findings provide a promising framework for future real-time electric field-optimized TMS applications.

Turning the operating room into a mixed-reality environment: a prospective clinical investigation for cerebral aneurysm clipping.

OBJECTIVE: The overall aim of this study was to demonstrate the potential benefit of a novel mixed-reality-head-mounted display (MR-HMD) on the spatial orientation of surgeons. METHODS: In a prospective clinical investigation, the authors applied for the first time a new multicamera navigation technology in an operating room setting that allowed them to directly compare MR-HMD navigation to standard monitor navigation. In the study, which included 14 patients with nonruptured middle cerebral artery aneurysms, the authors investigated how intuitively and effectively surgical instruments could be guided in 5 different visual navigation conditions. RESULTS: The authors demonstrate that multicamera tracking can be reliably integrated in a clinical setting (usability score 1.12 ± 0.31). Moreover, the technology captures large volumes of the operating room, allowing the team to track and integrate different devices and instruments, including MR-HMDs. Directly comparing mixed-reality navigation to standard monitor navigation revealed a significantly improved intuition in mixed reality, leading to navigation times that were twice as fast (2.1×, p ≤ 0.01). Despite the enhanced speed, the same targeting accuracy (approximately 2.5 mm, freehand tool use) in comparison to monitor navigation could be observed. Intraoperative planning strategies with mixed reality clearly outperformed classic preoperative planning: surgeons scored the mixed-reality plan as the best trajectory in 63% of the cases (chance level 33%). CONCLUSIONS: The incorporation of mixed reality in neurosurgical operations marks a significant advancement in the field. The use of mixed reality in brain surgery enhances the spatial awareness of surgeons, enabling more instinctive and precise surgical interventions. This technological integration promises to refine the execution of complex procedures without compromising accuracy.

Using camera-guided electrode microdrive navigation for precise 3D targeting of macaque brain sites.

Spatial accuracy in electrophysiological investigations is paramount, as precise localization and reliable access to specific brain regions help the advancement of our understanding of the brain's complex neural activity. Here, we introduce a novel, multi camera-based, frameless neuronavigation technique for precise, 3-dimensional electrode positioning in awake monkeys. The investigation of neural functions in awake primates often requires stable access to the brain with thin and delicate recording electrodes. This is usually realized by implanting a chronic recording chamber onto the skull of the animal that allows direct access to the dura. Most recording and positioning techniques utilize this implanted recording chamber as a holder of the microdrive or to hold a grid. This in turn reduces the degrees of freedom in positioning. To solve this problem, we require innovative, flexible, but precise tools for neuronal recordings. We instead mount the electrode microdrive above the animal on an arch, equipped with a series of translational and rotational micromanipulators, allowing movements in all axes. Here, the positioning is controlled by infrared cameras tracking the location of the microdrive and the monkey, allowing precise and flexible trajectories. To verify the accuracy of this technique, we created iron deposits in the tissue that could be detected by MRI. Our results demonstrate a remarkable precision with the confirmed physical location of these deposits averaging less than 0.5 mm from their planned position. Pilot electrophysiological recordings additionally demonstrate the accuracy and flexibility of this method. Our innovative approach could significantly enhance the accuracy and flexibility of neural recordings, potentially catalyzing further advancements in neuroscientific research.

How Many People Could Use an SSVEP BCI?

Brain-computer interfaces (BCI) are communication systems that allow people to send messages or commands without movement. BCIs rely on different types of signals in the electroencephalogram (EEG), typically P300s, steady-state visually evoked potentials (SSVEP), or event-related desynchronization. Early BCI systems were often evaluated with a selected group of subjects. Also, many articles do not mention data from subjects who performed poorly. These and other factors have made it difficult to estimate how many people could use different BCIs. The present study explored how many subjects could use an SSVEP BCI. We recorded data from 53 subjects while they participated in 1-4 runs that were each 4 min long. During these runs, the subjects focused on one of four LEDs that each flickered at a different frequency. The eight channel EEG data were analyzed with a minimum energy parameter estimation algorithm and classified with linear discriminant analysis into one of the four classes. Online results showed that SSVEP BCIs could provide effective communication for all 53 subjects, resulting in a grand average accuracy of 95.5%. About 96.2% of the subjects reached an accuracy above 80%, and nobody was below 60%. This study showed that SSVEP based BCI systems can reach very high accuracies after only a very short training period. The SSVEP approach worked for all participating subjects, who attained accuracy well above chance level. This is important because it shows that SSVEP BCIs could provide communication for some users when other approaches might not work for them.

Accuracy of a P300 speller for people with motor impairments: a comparison.

A Brain-Computer Interface (BCI) provides a completely new output pathway that can provide an additional option for a person to express himself/herself if he/she suffers a disorder like amyotrophic lateral sclerosis (ALS), brainstem stroke, brain or spinal cord injury or other diseases which impair the function of the common output pathways which are responsible for the control of muscles. For a P300 based BCI a matrix of randomly flashing characters is presented to the participant. To spell a character the person has to attend to it and to count how many times the character flashes. Although most BCIs are designed to help people with disabilities, they are mainly tested on healthy, young subjects who may achieve better results than people with impairments. In this study we compare measurements, performed on people suffering motor impairments, such as stroke or ALS, to measurements performed on healthy people. The overall accuracy of the persons with motor impairments reached 70.1% in comparison to 91% obtained for the group of healthy subjects. When looking at single subjects, one interesting example shows that under certain circumstances, when it is difficult for a patient to concentrate on one character for a longer period of time, the accuracy is higher when fewer flashes (i.e., stimuli) are presented. Furthermore, the influence of several tuning parameters is discussed as it shows that for some participants adaptations for achieving valuable spelling results are required. Finally, exclusion criteria for people who are not able to use the device are defined.