Parkinson’s disease (PD) is an incurable neurological disorder that often begins insidiously with sleep disturbances and somatic symptoms, progressing to whole-body motor and cognitive symptoms^1,2,3,4,5. Dysfunction of the somato-cognitive action network (SCAN)—which is thought to control action execution^6,7 by coordinating arousal, organ physiology and whole-body motor plans with behavioural motivation—is a potential contributor to the diverse clinical manifestations of PD. To investigate the role of the SCAN in PD pathophysiology and treatments (medications, deep-brain stimulation (DBS), transcranial magnetic stimulation (TMS) and MRI-guided focused ultrasound stimulation (MRgFUS)), we built a large (n = 863), multimodal, multi-intervention clinical imaging dataset. Resting-state functional connectivity revealed that the substantia nigra and all PD DBS targets (subthalamic nucleus, globus pallidus and ventral intermediate thalamus) are selectively connected to the SCAN rather than to effector-specific motor regions. Importantly, PD was characterized by specific hyperconnectivity between the SCAN and the subcortex. We therefore followed six PD cohorts undergoing DBS, TMS, MRgFUS and levodopa therapy using precision resting-state functional connectivity and electrocorticography recording. Efficacious treatments reduced SCAN-to-subcortex hyperconnectivity. Targeting the SCAN instead of effector regions doubled the efficacy of TMS treatments. Focused ultrasound treatment benefits increased when the target was closer to the thalamic SCAN sweet spot. Thus, SCAN hyperconnectivity is central to PD pathophysiology and its alleviation is a hallmark of successful neuromodulation. Targeting functionally defined subcortical SCAN nodes may improve existing therapies (DBS, MRgFUS), whereas cortical SCAN targets offer effective non-invasive or minimally invasive neuromodulation for PD.

Signal loss in blood oxygen level-dependent (BOLD) functional neuroimaging is common and can lead to misinterpretation of findings. Here, we reconstructed compromised fMRI signal using deep machine learning. We trained a model to learn principles governing BOLD activity in one dataset and reconstruct artificially compromised regions in an independent dataset, frame by frame. Intriguingly, BOLD time series extracted from reconstructed frames are correlated with the original time series, even though the frames do not independently carry any temporal information. Moreover, reconstructed functional connectivity maps exhibit good correspondence with the original connectivity maps, indicating that the model recovers functional relationships among brain regions. We replicated this result in two healthy datasets and in patients whose scans suffered signal loss due to intracortical electrodes. Critically, the reconstructions capture individual-specific information. Deep machine learning thus presents a unique opportunity to reconstruct compromised BOLD signal while capturing features of an individual’s own functional brain organization.

Neural electrode interfaces are essential to the stimulation safety and recording quality of various bioelectronic therapies. The recently proposed hierarchical platinum-iridium (Pt-Ir) electrodes produced by femtosecond lasers have exhibited superior electrochemical performance in vitro, but their in vivo performance is still unclear. In this study, we explored the electrochemical performance, biological response, and tissue adhesion of hierarchical Pt-Ir electrodes by implantation in adult rat brains for 1, 8, and 16 weeks. Regular smooth Pt-Ir electrodes were used as a control. The results showed that the electrochemical performance of both electrodes decreased and leveled off during implantation. However, after 16 weeks, the charge storage capacity of hierarchical electrodes stabilized at ~16.8 mC/cm², which was 15 times that of the smooth control electrodes (1.1 mC/cm²). Moreover, the highly structured electrodes had lower impedance amplitude and cutoff frequency values. The similar histological response to smooth electrodes indicated good biocompatibility of the hierarchically structured Pt-Ir electrodes. Given their superior in vivo performance, the femtosecond laser-treated Pt-Ir electrode showed great potential for neuromodulation applications.

Radio frequency (RF)-induced heating associated with medical implants is a critical concern when performing a magnetic resonance imaging (MRI) scan. The purpose of this study is to construct a “digital twin” of RF coil integrated in an MRI scanner by $B_{1}^{+}$ map matching, reproducing the physical electromagnetic (EM) field, allowing it to be a qualified test environment for heating evaluations. $B_{1}^{+}$ map inside an American Society for Testing Material (ASTM) phantom with various positions was used to characterize the physical EM field. A loaded tuned two-port birdcage body coil model was established to generate the numerical EM field and optimized to minimize the difference between the measured and simulated fields, which was quantified by “mean symmetric mean absolute percentage error” (mSMAPE). The amplitude of the simulated field was normalized to MR sequence labeling. Finally, the average specific absorption rate (SAR) inside the ASTM phantom was measured in a calorimetry experiment and compared to simulation one heated by the pre-established coil. A heating experiment was performed on an insulated cable to further validate the established test environment. The impact of excitation parameters and the port location was investigated. mSMAPE of the optimized field was less than 10%. The difference between the simulated and measured average SAR was 8.5%. $R^{2}$ for the linear regression of the temperature rise and predicted power deposition using the optimized field was 0.993. The digital twin was successfully constructed and verified a sufficiently accurate description of the physical field. The established construction method of the digital twin provides a handy choice for building RF test environment and may create more possibilities for the development of this area.