Introduction – Deep Brain Stimulation is an established treatment for movement disorders. Data-driven approaches like probabilistic mapping and predictive modeling are being increasingly used to optimize stimulation parameter selection. However, it remains unclear whether intra-operative test data alone can reliably inform such models. Clarifying the predictive value of intra-operative data is essential, as it may influence data collection strategies and surgical and programming protocols. Objective – This study examined whether post-operative DBS effects can be accurately predicted using intra-operative data alone or if post-operative information is required. Methods – A dataset comprising 1117 intra-operative and 1553 post-operative stimulation tests from 35 patients (14 Essential Tremor, 21 Parkinson’s Disease) was analyzed. Volumes of tissue activated (VTAs) were simulated to generate probabilistic maps, from which mapping and target anatomy-related features were extracted to train predictive classification models (low and high improvement, side effects). Model performance was assessed across scenarios linking intra- and post-operative data. Results – Intra-operative data effectively identified regions associated with optimal stimulation, highlighting their utility in guiding parameter selection (predictive accuracy ~60%). However, the predictive relationship between VTAs, probabilistic maps, and clinical outcomes differed between intra- and post-operative contexts. When trained solely on intra-operative VTAs, the model performed at a near chance level (predictive accuracy ~35%). Discussion – The study demonstrates that while intra-operative data are useful for identifying optimal target regions, they are insufficient for accurately predicting post-operative DBS outcomes. Incorporating post-operative data remains crucial for reliable and clinically meaningful DBS effect prediction.

Introduction:Probabilistic Stimulation Maps (PSMs) are increasingly employed to identify brain regions associated with optimal therapeutic outcomes in Deep Brain Stimulation (DBS). However, their reliability and generalizability are challenged by the limited size of most patient cohorts and the inherent variability introduced by different statistical methods and input data configurations. This study aimed to investigate the geometrical variability of Probabilistic Sweet Spots (PSS) as a function of both the number of patients (nPat) and the number of stimulations per patient (nStim), and to model a stability boundary defining the minimum data requirements for obtaining geometrically stable PSS. Methods:Three statistical approaches-Bayesian t-test, Wilcoxon test with False Discovery Rate (FDR) correction, and Wilcoxon test with nonparametric permutation correction-were applied to two patient cohorts: a primary cohort of 36 patients undergoing DBS for Parkinson's Disease (PD), and a secondary cohort of 61 patients treated for Essential Tremor (ET), used to assess generalizability. Stimulation test data was collected intra-operatively for the first cohort and post-operatively for the second one. Geometric stability was evaluated based on variability in PSS volume extent and centroid location. Results:The analysis revealed a non-linear trade-off between nPat and nStim to yield stable PSS. A stability boundary was defined, representing the minimum combinations of nPat-nStim required for anatomically robust PSS. Among the tested methods, the Bayesian t-test achieved stability with smaller sample sizes (∼15 patients) and demonstrated a consistent performance across both cohorts. In contrast, the Wilcoxon-based methods showed variable behavior between cohorts, which differed in symptom type and testing phase (intra-operative testing vs. post-operative screening). Discussion:The proposed PSS stability boundary provides a practical reference for designing DBS studies and stimulation screening protocols aimed at probabilistic mapping. The Bayesian t-test emerged as a reliable method across both cohorts, supporting its potential in studies with limited sample sizes and scenarios where the method needs to be readily generalized to varying symptoms. These findings underscore the importance of considering both cohort size and stimulation count in probabilistic DBS mapping and call for further investigation into method-specific sensitivities to clinical and procedural factors.

Background:The Movement Disorder Society Unified Parkinson's Disease Rating Scale (MDS-UPDRS) is used as a standardized approach to assess motor function in Parkinson's disease (PD). This assessment is based on the examiner's subjective judgement and is therefore variable. While quantitative approaches have been evaluated for upper-limb movements, data is scarce on lower-limb movements. Thus, our aim was to implement and assess a setup to quantify lower-limb movements as defined in the MDS-UPDRS in healthy participants introducing new parameters for smoothness and acceleration patterns. Methods:Twenty-three participants (age-range = 21-31 years) performed five 20-s trials for both lower-limb movement tasks from the MDS-UPDRS-III, i.e., toe tapping (item 3.7) and leg agility (item 3.8). Foot and leg movements were recorded using four inertial measurement units (two per leg: one mounted on the foot and one on the ankle). Biomarkers such as kinematic parameters (e.g., frequency, angular amplitude, movement smoothness, acceleration-based parameters) were extracted to characterize foot-movement dynamics (dominant vs. non-dominant leg), with statistical analyses including linear mixed-effects models applied to four consecutive, non-overlapping 5-s intervals. Results:A paired Wilcoxon test showed no significant differences in parameters for toe tapping and leg agility based on leg dominance. For toe tapping, the relationship between frequency and angle displayed non-linearity, with a clear decrease in angle with 17.41° -0.83°/interval*t (t = 1-4 time intervals) and no clear decrease in frequency with 2.75 Hz - 0.02 Hz/interval*t. The median frequency and angle for toe tapping were 2.8 Hz and 16° respectively. The median frequency for leg agility was 2.6 Hz. Conclusion:Reference values could be determined for all parameters including smoothness and acceleration patterns. The quantitative assessment of two MDS-UPDRS-III items shows that temporal changes and adaptation-mechanisms significantly influenced leg-movement dynamics. Reducing exercise duration to 10 s and implementing a metronome with defined frequency could enhance measurement accuracy and reliability, offering more precise parameters for future applications in PD-patients.

Hyperkinetic movement disorders (HMDs) such as dystonia, tremor, chorea, myoclonus, and tics are disabling motor manifestations across childhood and adulthood. Their fluctuating, intermittent, and frequently co-occurring expressions hinder clinical recognition and longitudinal monitoring, which remain largely subjective and vulnerable to inter-rater variability. Objective and scalable methods to distinguish overlapping HMD phenotypes from routine clinical videos are still lacking. Here, we developed a pose-based machine-learning framework that converts standard outpatient videos into anatomically meaningful keypoint time series and computes kinematic descriptors spanning statistical, temporal, spectral, and higher-order irregularity-complexity features.

OBJECTIVE:To explore whether routine outpatient video combined with deep learning-based pose estimation and clinically interpretable kinematic features can support multi-label phenotyping of co-occurring hyperkinetic movement disorders (HMDs). METHODS:In this exploratory single-centre proof-of-concept study, videos from 21 patients with HMDs and 4 healthy controls were processed with markerless pose estimation (YOLOv8) and 2-dimensional keypoint trajectories transformed into kinematic descriptors spanning statistical, temporal, spectral, and complexity domains. Ten-second windows were aligned to expert annotations for eight hyperkinetic phenomenologies. Conventional supervised classifiers were trained on these tabular descriptors. Window-level predictions were aggregated to the patient level, and label-specific thresholds were tuned on training participants only. RESULTS:In patient-level multi-label performance reporting, (i) the best single pipeline selected by discrimination (StandardScaler + MLP) achieved a macro-AUPRC of 0.821 ± 0.019 and a macro-receiver operating characteristic area under the curve of 0.830 ± 0.029. (ii) The best single pipeline selected by Hamming accuracy (MinMaxScaler + SVM) reached 0.764 ± 0.041. (iii) Under prespecified nested cross-validation with per-label model selection within training folds (primary analysis), macro-AUPRC was 0.717 ± 0.030, macro-AUROC was 0.767 ± 0.069, Hamming accuracy was 0.764 ± 0.014 and patient-label agreement was 153/200 (76.5%). (iv) Post hoc per-label selection of the best-performing pipeline defined an exploratory upper bound of 172/200 (86.0%). INTERPRETATION:In this exploratory study, a hybrid pipeline combining deep learning pose estimation with feature-engineered supervised classification produced encouraging patient-level multi-label performance for co-occurring HMDs. These findings are proof-of-concept; external, multicentre, prospective validation is required before clinical or trial use.

Dual deep brain stimulation of the ventrointermediate and ventrooral lateral thalamic nuclei was proposed to restore control of delayed-escape essential tremor. We report an illustrative case that includes a new hardware setting, named uncoupled, asynchronous dual deep brain stimulation, and an advanced topographic analysis of the volume of tissue activated. A 61-year-old patient with a 20-year history of essential tremor underwent bilateral implantation in the ventrointermediate nucleus and the immediately underlying tegmental field. This intervention alleviated the tremor but, over time, resulted in recurrence in the right upper limb and additional symptoms. Six years later, a second surgery was performed, implanting a new left electrode slightly anteriorly in the ventrooral lateral nucleus and the immediately underlying structures, along with a new neurostimulator. This approach successfully controlled the tremor. The second system was independent of the first, leading to an uncoupled, asynchronous dual deep brain stimulation. The analysis of the volume of tissue activated confirmed the involvement of the ventrointermediate and ventrooral lateral thalamic nuclei. It highlighted the role of the prelemniscal-radiation hub within the subthalamic tegmental fields, and the implication of the subthalamic nucleus and the zona incerta. The review of literature shows the specificity of uncoupled dual stimulation in the field, and the utility of advanced topographic analysis in disentangling neural correlates of anti-tremor effect.

Picture naming tasks are widely used to evaluate language processing in aphasia. Automatically detected naming latencies (aNL) often differ from the gold standard, manually detected latencies (mNL). Initial phonemes (InitP) influence aNL detections. The study aimed to develop and explore an InitP-based optimization approach to enhance automatic naming latency detection. An algorithm was developed to optimize several parameters of the word speech envelope for 16 InitPs using German training data from 134 healthy subjects and 31 persons with aphasia. These parameters were applied to test data to assess InitP influence. The resulting naming latencies (FHNW_NL), detected by the newly developed algorithm, were compared to manually detected latencies (mNL) and to speech onsets identified by Kaldi and Chronset. The mean signed difference between FHNW_NL and mNL falls within the +/- 15 ms inter-rater variability interval. Differences from Kaldi and Chronset to mNL were over twice and four times higher, respectively, than FHNW_NL. Persons with aphasia showed higher variances than healthy subjects. This study shows that InitPbased optimization of naming latency detection parameters improves aNL detection accuracy. This method could also benefit other research and clinical fields.

Background: With the increasing prevalence of multimorbidity, the demand for palliative and end-of-life care is expected to rise substantially in the coming decades. Digital health technologies may enable automated detection of clinically relevant distress, including symptoms such as pain, breathlessness (dyspnea), anxiety/panic, nausea, and agitation. Remote detection of such distress in cognitively impaired patients who are unable to reliably call for help could enable timely intervention when patients are unattended. Methods: This observational feasibility study will collect multimodal data from a non-invasive sensor system consisting of 3D radar, a photoplethysmographic sensor (wearable), and a microphone. Sensor data will be linked to distress events identified by nurses or physicians during routine clinical care using structured proxy assessments. Adults (≥18 years) admitted to the Palliative Care Center Basel who are unable to reliably call for help due to cognitive impairment will be included based on written informed consent provided by a legal proxy. Aim: The aim of this study is to evaluate the feasibility of multimodal sensor-based monitoring for detecting clinician-identified distress events and to explore associations between sensor-derived variables and distress, informing future validation studies and the development of automated detection approaches in palliative care.

OBJECTIVE:Probabilistic mapping is increasingly used to identify optimal stimulation regions (Probabilistic Sweet Spots, PSS) in Deep Brain Stimulation (DBS). Outcomes, however, depend on workflow parameters. This study examined how methodological and data handling choices affect PSS stability and spatial consistency across varying sample sizes. METHODS:Intraoperative stimulation test data from 36 Parkinson's Disease patients were analyzed. PSS were computed across increasing sample sizes using four statistical approaches: Bayesian t-test (BAYES), Logistic Regression Model (LRM), Wilcoxon test with FDR correction (WFDR), and Wilcoxon test with permutation correction (WPERM). We assessed the effects of statistical tests, hemispheric data handling, and masking parameters (i.e., minimum number of patients and stimulations per voxel) on PSS stability and consistency, evaluated in terms of size and spatial location. RESULTS:BAYES was more robust at small to intermediate sample sizes, while WFDR and LRM stabilized only in larger cohorts (∼25-30 patients). WPERM consistently underperformed. Stability was higher in the left hemisphere. Combining hemispheres did not improve stability, suggesting asymmetries in stimulation effects. Masking parameters mainly affected PSS volume, with stricter thresholds reducing absolute size, but did not alter stability patterns. CONCLUSION:Statistical test choice, hemispheric analysis, and masking parameters strongly influence PSS outcomes. The Bayesian t-test is recommended for small to intermediate cohorts, and hemispheres should be analyzed separately to avoid masking clinically relevant asymmetries. SIGNIFICANCE:By highlighting the interplay between sample size, statistical methods, hemispheric data, and masking strategies, this work contributes to standardizing probabilistic mapping practices and improving their reliability for clinical translation.

This paper describes a 3D electromagnetic tracking (EMT) system, based on quasi-static magnetic fields and a sub-millimeter 3D magnetometer, providing complete localization - both spatial and angular positions - during surgical procedures. By integrating miniaturized sensors into surgical tools, such as deep brain stimulation (DBS) electrodes, this tracking system offers complementary or alternative solutions for X-ray imaging. Each spatial position in the measurement volume (MV) is uniquely encoded by a vector of four magnetic field amplitudes using the multilateration principle. The orientation is derived from the three orthogonal components associated with this vector. The field generator (FG) was manufactured on printed circuit boards ensuring high reproducibility and accurate magnetic fields. Position localization was evaluated using a custom magnetic field camera placed at various positions in the MV while the orientation was evaluating using a stereotactic system used in DBS surgery. Finally, DBS implantations were simulated to conclude on the validity of the tracking system for DBS surgery. The system achieved spatial and angular errors of 1.72 mm and 0.89 degrees within a MV of 15 x 15 x 15 cm3 located at 18 cm from the FG and an update rate of the position of 0.4 Hz. Better performances - mean spatial and angular errors of 0.87 mm and 0.52 degrees - were achieved when simulating DBS implantations. With its large distance to the FG, this quasi-static EMT system is particularly well-suited for DBS surgery, offering regular feedback to neurosurgeons. The tracking system could also be adapted to other functional neurosurgeries.

Most studies conducting probabilistic mapping of the effect of deep brain stimulation (DBS) use an external anatomical reference such as those of the MNI (Montreal Neurological Institute, [1]), alternatively, group-specific templates can be generated to avoid external anatomical bias. This study investigates the effect of cohort size on the creation of such anatomical references for movement disorders. Pre-operative MRI data from 70 patients implanted with DBS systems were used to generate anatomical templates with varying cohort sizes (5 to 67 subjects). An iterative non-linear normalization pipeline was employed to optimize template generation. Template variability was assessed using Dice overlap of anatomical structures. The templates created with 44 subjects achieved an optimal balance between variability and precision. Tukey’s HSD test confirmed significant differences between iterations and cohort sizes. This study underscores the importance of cohort size and iterative registration methods in creating high-quality anatomical templates.Clinical relevance— The findings provide insights into the optimal cohort size for creating group-specific anatomical brain templates.

Deep Brain Stimulation (DBS) is an effective treatment for movement disorders. Optimizing stimulation parameters remains, however, a trial-and-error process. Datadriven models leveraging Probabilistic Mapping have shown promise in predicting DBS outcomes, yet current studies rely on chronic stimulation data. This study explores the feasibility of using intra-operative stimulation test data for DBS effect prediction. Probabilistic volumes of beneficial and adverse effects were computed from intra-operative stimulation test data of 65 patients (23 with Essential Tremor + 42 with Parkinson’s Disease). A prediction dataset was generated including clinical, morphological, stimulation features along with features derived from probabilistic maps and simulated Volumes of Tissue Activated. Three machine learning models (Adaboost, Support Vector Classifier and Naïve Bayes) were implemented to predict stimulation effects in a classification task. The models were validated in a leave-one-out crossvalidation and their performances were compared. All the developed models were able to predict DBS outcome classes. The best predictive performance was achieved by the Adaboost model with a maximum balanced accuracy of 0.71 on 3 classes. These results show that intra-operative stimulation test data can predict DBS effects with a similar approach and comparable accuracy to post-operative monopolar review data.

Real-time guidance for the implantation of deep-brain-stimulation (DBS) electrodes in the context of stereotactic neurosurgery is essential but currently unavailable. Electromagnetic tracking (EMT) systems offer high-accuracy localization of tools in restricted volumes but face compatibility issues with stereotactic procedures due to electromagnetic distortions. This paper aims to evaluate and compare the localization performance (position and orientation) of a novel EMT system, the ManaDBS, specifically designed for stereotactic surgical environments, against the NDI Aurora, a commercially available EMT system. Two studies were conducted to assess the suitability of each EMT system for stereotactic DBS surgery. The first study evaluated performance accuracy within the measurement volume in the presence of two different stereotactic systems (Frame G and Vantage system, Elekta). The second study simulated a DBS surgical theater, performing implantation procedures with each EMT system and evaluating the position accuracy of the EMT sensor. The localization errors of Aurora (0.66 mm and 0.89°) were lower to those of ManaDBS (1.57 mm and 1.01°). However, in the presence of a stereotactic system, Aurora exhibited notable degradation (2.34 mm and 1.03°), whereas ManaDBS remained unaffected. This pattern persisted during simulated implantation in a DBS surgical environment, where nonlinear trajectories with significant error fluctuations along the implantation path were observed with Aurora system. The significant electromagnetic-field distortions render the Aurora system incompatible for stereotactic DBS surgery. However, the ManaDBS system exhibited no impact from these distortions, suggesting its potential suitability for DBS surgery and other potential applications in stereotactic neurosurgery.

Digital biomarkers offer novel approaches to non-invasive health monitoring, particularly in Palliative Care, for example in cancer patients, where the identification and relief of symptom burden and distress are the leading goals of care. This study investigates the correlation between acoustic speech features and distress severity in female cancer patients, using the Edmonton Symptom Assessment System (ESAS) as a reference. Speech recordings were collected from 28 cancer patients at up to four different time points, with acoustic features extracted using the openSMILE toolkit (ComParE 2016 feature set). The analysis focused on the 23rd and 24th Mel filter-bank bands—MFB 23 and MFB 24—which are two individual channels of the 26-channel Mel filter-bank computed by openSMILE. Pearson correlation analysis identified 13 spectral features significantly associated with the ESAS score of the female cohort. A multivariate Ordinary Least Squares (OLS) regression model demonstrated that selected acoustic parameters explained 29% of the variance in distress levels, with spectral flatness and mean energy in MFB 23 emerging as key predictors. These findings suggest that speech-based biomarkers may facilitate automated, objective distress screening in oncology patients. By integrating acoustic analysis into clinical workflows, this study highlights the potential of digital voice biomarkers for continuous symptom monitoring. Future research should refine predictive models and expand patient cohorts to enhance clinical applicability.Clinical relevance—This study highlights the potential of speech-derived digital biomarkers for distress screening in female palliative oncology patients. By correlating acoustic speech features with ESAS scores, it demonstrates a noninvasive, objective method for symptom monitoring. Integrating voice analysis into clinical workflows could enhance early intervention, reduce patient burden, and improve precision in symptom management, for example by integration into telemedicine interventions and follow-ups.

Deep Brain Stimulation (DBS) for movement disorders can greatly benefit from the insights provided by probabilistic mapping. This consists in the application of statistical approaches to stimulation data of multiple patients. Such analysis is influenced by the input data and chosen statistical method, hindering the generalizability of the obtained results. This study aims at determining the minimum sample size yielding stable results, and the statistical approach providing higher results consistency. Intra-operative stimulation test data of 36 patients who underwent DBS surgery for Parkinson’s Disease (PD), were used to compute Probabilistic Sweet Spots (PSS). The PSS were calculated with sample sizes ranging from 4 to 36 (steps of 2) using Bayesian t-test, Wilcoxon test with False Discovery Rate correction and Wilcoxon test with nonparametric permutations correction. Calculations were repeated 10 times. Obtained PSS were compared in terms of size and position variability across sample sizes and between statistical methods. Only the PSS computed with the Bayesian t-test reached stability in all the three chosen metrics. Stability in size and centroid location was reached from a sample size of 14 patients, while the covered volume (Dice coefficient) stabilized from a sample size of 18 patients. The Bayesian t-test also provided higher results consistency with respect to the other approaches. The composition of the dataset in cohorts with <20 patients has a greater influence on the extent and location of computed PSS. Moreover, the Bayesian t-test demonstrated the highest suitability to extrapolate results from analyses involving small sample sizes.Clinical Relevance— Determining the minimum number of patients and most suitable method to calculate stable PSS is crucial for ensuring the reliability and clinical applicability of Probabilistic Mapping approaches. This ensures the generalizability of findings, allowing for a more accurate understanding of the underlying pathology. These insights can be used to refine diagnostic approaches and optimize clinical interventions, ultimately improving patient care.

Electromagnetic navigation systems (ENS) provide a promising solution for tracking electrode positioning during deep brain stimulation (DBS) surgery, enhancing precision and potentially improving clinical outcomes. Recent studies suggest that novel ENS based on quasi-static magnetic fields may be compatible with the DBS surgical environment, particularly with stereotactic systems — the gold standard for electrode implantation. However, a key challenge persists: ensuring seamless integration into the surgical workflow requires an efficient and reliable patient-to-image registration system. In this study, we present a novel registration system tailored for stereotactic procedures. The system employs flexible markers attached to the stereotactic frame, incorporating magnetic sensors and wireless communication to enable seamless interaction with the surgical environment. These markers are visible on preoperative computed tomography (CT) scans, eliminating the need for additional intraoperative imaging. The registration method demonstrated robust performance with automatic segmentation of the markers on CT images, achieving an average point matching error of 1.51 mm. During implantation tests, the system localized the ENS sensors with a target registration error of 2.54 ± 0.92 mm. This innovative navigation approach ensures precise localization, reducing reliance on repeated CT imaging for verification, thus streamlining the surgical workflow. Clinical Relevance— This proposes a registration-free approach for stereotactic neurosurgery, enabling the monitoring of electrode position and orientation during implantation.

This study validates a smartphone app for hand tremor assessment in Parkinson’s disease (PD). Twenty-eight PD patients performed a weekly tremor test using the app while wearing a wrist-worn actigraphy device (GeneActiv), of which twenty-one yielded usable actigraphy data for comparative analysis. Features were extracted from both devices to compare smartphone application derived data against actigraphy measurements. Further analysis examined the validity of the smartphone data in ON versus OFF medication states and against clinical scales (MDS-UPDRS). Statistical tests show high correlations between the app and GeneActiv features whereas app-derived features show statistical significant differences on data between medication states (p < 0.05), and correlations (maximum correlation 0.47) with clinical scores (MDS-UPDRS). The findings support the clinical validity of smartphone-based tremor assessments in PD, which shows potential for ongoing symptom monitoring in individuals with PD from at-home environments.

Deep Brain Stimulation (DBS) is an established therapy for movement and neuropsychiatric disorders. Identifying brain regions (Probabilistic Sweet Spots, PSS) linked with the greatest symptom improvement is crucial for refining pre-operative targeting and post-operative programming. Probabilistic stimulation mapping is a powerful data-driven tool to delineate these regions. However, the chosen statistical methods can influence the identified PSS. A comprehensive evaluation of their impact is lacking in DBS research. The present study compares the PSS generated with four voxel-wise statistical approaches - t-test, Wilcoxon test, Linear Mixed Model, and Bayesian t-test - with the aim of assessing their influence on computed results on the same dataset. Intra-operative stimulation test data of 23 Essential Tremor (ET) patients was used to run patient-specific electric field simulations and to generate PSS in a group-specific anatomical template space. The PSS for the different statistical tests were first compared in terms of size and topography. Then, their correlation with clinical improvement was calculated in a leave-one-out cross-validation scheme and PSS consistency across datasets with different compositions was assessed. Our findings emphasize the impact of statistical test selection on both the anatomical location and volume of the extracted PSS, highlighting the importance of careful methodological choices in future DBS mapping studies. The Bayesian t-test and a voxel-wise application of nonparametric permutation testing, introduced for the first time in DBS research, showed promising results in identifying PSS representative of improvement and exhibited robustness to variations in the dataset.

Evaluating experimentally the human exposure to time varying magnetic fields is often complex due to the strongly non-homogeneous magnetic field observed near electromagnetic devices. This study investigates the use of an in-house ac magnetic field camera (MFC) to map the exposure coefficients defined by standards and guidelines (ICNIRP and ICES). As a case study this paper reports on the spatial distribution of the magnetic field exposure coefficients of a neurosurgical tracking system. Although initial field strengths were low, the frequency analysis revealed that exposure coefficients exceeded recommended limits in worst-case scenarios. To mitigate this, a bandwidth limitation of the magnetic field generated by the tracking system was implemented effectively reducing exposure levels. These results highlight the importance of optimized system design and signal processing to ensure compliance with safety standards in surgical applications.