Citations

Clear

23 citation(s) found for "Lin, Wei".

A Shared Systemic Metabolic Signature Across the Neurological Disease Spectrum: A Summary-Level Analysis of 274,241 UK Biobank Participants
Yang, Likun, Lin, Wei , Mirsattari, Seyed M., Mirsattari, Seyed M.

Mayo Clinic

Biomedicines | August 6, 2026 | Volume: 14 | Issue: 8 | Pages: 1773

10.3390/biomedicines14081773 | PMID 42652156 Get It

Background: Neurological diseases share overlapping systemic and molecular features, but the specificity and interpretation of cross-disease metabolomic signatures remain uncertain. Methods: We analyzed summary-level Nightingale NMR metabolomic and genetic data from 274,241 UK Biobank participants across eight neurological endpoints. The disease rankings used here were derived from baseline plasma metabolites predicting future incident disease rather than from contemporaneous diagnostic case–control contrasts. Results: Among 57 priority metabolites with genome-wide significant instruments (2472 SNPs; mean F = 162.4), nine ranked in the top 30 for at least seven of the eight endpoints, defining a shared pre-diagnostic neurological signature confirmed as non-random cross-endpoint convergence by permutation testing (p < 0.0001). Because we did not perform a quantitative non-neurological disease control analysis, this signature should not be interpreted as neurologically specific and may partly reflect systemic morbidity, renal function, body composition, or frailty-related physiology. Forward Mendelian randomization across 45 metabolite–disease pairs found no Bonferroni-significant causal effects; three nominal protective associations are consistent with the number of false-positive findings expected under multiple testing and require replication. Conclusions: These results support a shared systemic, pre-diagnostic metabolic signature across the neurological disease spectrum, while the null MR findings and specificity limitations favor interpretation as a biomarker or prodromal downstream signal rather than a proven causal mechanism. External prospective validation is essential.
Learning Curves and Complexity Evolution in Robotic Liver Surgery
Kuemmerli, Christoph , Hess, Gabriel Fridolin , Billeter, Adrian T. , Benedetto, Fabrizio Di , Odorizzi, Roberta, Sucandy, Iswanto , Aegerter, Noa L. E., Berchtold, Caroline , Bernardi, L. , Cristaudi, Alessandra , Balzano, E , Koh, Ye-Xin, Lin, Charles Chung-Wei, Fukumori, Daisuke , Tschuor, Christoph , White, Steve, Kroge, P Von, Heumann, Asmus , Meyere, Céline De, D’hondt, M, Haak, Fabian, Stättner, Stefan, Primavesi, Florian, Morgül, Haluk , Andreou, A, Ratti, F, Esquer, Ignacio Sánchez, Robles-Campos, Ricardo, Lopez-Lopez, Victor, Wakabayashi, Go , Nebbia, Martina, Rahbari, Nuh N., Ferretti, Stefano, Ettore, Giuseppe , Croner, RS , Dong, Y, Starlinger, P, Nickel, F, Müller, Beat P., Hess, Gabriel F , Guidetti, Cristiano , di Benedetto, Fabrizio , Sucandy, Iswanto , Aegerter, Noa L E , Berchtold, Caroline , Bernardi, Lorenzo , Cristaudi, Alessandra , Balzano, Emanuele , Ghinolfi, Davide , Goh, Brian K P , Fukumori, Daisuke , Tschuor, Christoph , Pandanaboyana, Sanjay , De Meyere, Celine , Seehofer, Daniel , Esquer, Ignacio Sánchez , Müller, Philip C , Koh, Ye-Xin , Lin, Charles Chung-Wei, Tschuor, Christoph , Kroge, Philipp H. von , D’Hondt, Mathieu , Haak, Fabian , Stättner, Stefan , Primavesi, Florian , Ratti, Francesca , Robles-Campos, Ricardo, López‐López, Víctor , Wakabayashi, Taiga , Nebbia, Martina , Gilg, Stefan , Birgin, Emrullah , Ferretti, Stefano , Ettore, Giuseppe Maria, Rahimli, Mirhasan , Nickel, Felix , Müller-Stich, Beat Peter

Mayo Clinic

Annals of Surgery | June 17, 2026 | Volume: No volume available | Issue: No issue available | Pages: No page available

10.1097/sla.0000000000007123 | PMID 42307103 Get It

Objective: To evaluate the institutional learning curve of robotic liver resection (RLR), its association with surgical complexity and outcomes in an international multicenter cohort. Summary of Background Data: RLR is increasingly adopted worldwide, but the dynamics of the learning process and its impact on surgical outcomes remain poorly defined. Methods: This international, multicenter, retrospective cohort study included consecutive patients who underwent RLR between 06/2013 and 12/2024 across 22 expert centers. Based on the first 150 cases at each center, three phases were used to define the learning curve: operative time for competency, major complications for proficiency, and textbook outcome for mastery. Risk-adjusted generalized additive models were used to assess the learning process. Benchmark performance was compared to international reference values. Results: Among 2,331 patients, the most common indications were hepatocellular carcinoma (36%), benign lesions (23%) and colorectal liver metastases (21%). Operative time decreased after achieving competency (case #62), from 240 to 220 minutes ( P <0.001). Major morbidity reached a plateau during the proficiency phase (cases #63–138; 9% vs. 11%, P =0.144), while progression to mastery (cases #139–150) was associated with a higher rate of textbook outcome (76% vs. 87%, P =0.021). In contrast, conversion (5.1% vs. 4.9% vs. 2%), bile leak (4.0% vs. 5.3% vs. 3.8%), and 90-day mortality (0.7% vs. 0.9% vs. 0.9%) remained low and stable across all learning phases. Importantly, benchmark criteria for operative time, hospital stay, and textbook outcome were met across all complexity groups, including during the early learning phase. Conclusions: Robotic liver surgery shows progressive outcome improvement with early achievement of benchmark-level results and stable performance across increasing complexity. These findings indicate that, in experienced centers, RLR can be safely implemented and integrated into clinical practice, supporting its key role in modern hepatobiliary surgery.
Machine-Learning-Based Targeted Plasma Proteomic Analysis for Predicting Motor Progression in Parkinson’s Disease: An Interpretable Approach to Personalized Disease Management
Lin, Wei, Grewal, Sanjeet S.

Mayo Clinic

Bioengineering | March 26, 2026 | Volume: 13 | Issue: 4 | Pages: 380

10.3390/bioengineering13040380 | PMID 42072174 Get It

The accurate prediction of motor progression in Parkinson’s disease (PD) remains a major clinical challenge that limits personalized treatment planning and efficient clinical trial design. In this study, we developed and validated a machine-learning framework integrating a targeted panel of plasma proteins measured by Olink proximity extension assays with clinical variables to stratify patients according to their progression risk. We analyzed baseline plasma samples from 211 early-stage PD patients enrolled in the Parkinson’s Progression Markers Initiative (PPMI) cohort using four targeted Olink panels, from which 28 circulating proteins were retained after quality-control filtering. Patients were classified as rapid or slow progressors based on their annualized change in MDS-UPDRS Part III scores. Among the algorithms tested, Random Forest achieved the highest discriminative performance with an area under the receiver operating characteristic curve (AUC) of 0.751 (95% CI: 0.684–0.811), which exceeded that of clinical predictors alone (AUC 0.666). The integration of targeted proteomic and clinical features further improved model performance (AUC 0.773; p = 0.009). Nested cross-validation confirmed minimal optimistic bias (AUC 0.743). To enhance clinical interpretability, we applied SHapley Additive exPlanations (SHAP) analysis, which identified interleukin-6 (IL-6), brain-derived neurotrophic factor (BDNF), and vascular endothelial growth factor A (VEGF-A) as the most influential predictors. SHAP feature rankings were highly stable across cross-validation folds (mean Spearman ρ = 0.91). The robustness of these findings was confirmed through sensitivity analyses using extreme quartile comparisons (AUC 0.823), treatment-naïve subgroup analysis (AUC 0.738), and a clinically anchored outcome definition based on the minimal clinically important difference (AUC 0.739). A decision curve analysis demonstrated a net clinical benefit across threshold probabilities of 0.25–0.70. Our results establish targeted plasma protein profiling combined with interpretable machine learning as a promising tool for PD motor progression risk stratification, with potential applications in individualized patient counseling regarding motor prognosis and the selection of candidates for disease-modifying trials.
Deep Learning for Freezing of Gait Detection: Cross-Dataset Validation Reveals Critical Deployment Gaps Between Laboratory and Daily Living Wearable Monitoring
Lin, Wei, Grewal, Sanjeet S., Grewal, Sanjeet S.

Mayo Clinic

Sensors | February 20, 2026 | Volume: 26 | Issue: 4 | Pages: 1352

10.3390/s26041352 | PMID 41755290 Get It

Freezing of gait (FoG) affects 38–65% of advanced Parkinson’s disease patients, yet automated detection algorithms are often validated solely on laboratory datasets. This study quantifies the critical performance gap between laboratory and real-world performance—a prerequisite for clinical deployment. Using temporal convolutional networks (TCNs), we trained models on two public datasets representing ecological extremes: a daily living dataset (Figshare; n = 35, single-sensor) and a laboratory dataset (DAPHNET; n = 10, multi-sensor). We compared five training configurations to address class imbalance. Results showed that F1-based early stopping outperformed Area Under the Curve (AUC)-based stopping by 47% (F1: 0.55 vs. 0.37, p = 0.0008). Combining multiple imbalance corrections (focal loss, weighting, sampling) paradoxically degraded precision to 33% due to a ~60-fold over-weighting of the minority class. Most importantly, cross-dataset validation revealed an 83% performance gap: laboratory F1 reached 0.9999 ± 0.0002, whereas daily living F1 dropped to 0.55 ± 0.26 (p < 0.0001), with a 1299-fold increase in variance. These findings demonstrate that laboratory success does not guarantee real-world utility. We propose that the observed gap represents a “deployment gap” reflecting the combined influence of environmental complexity, sensor constraints, and physiological variability. These results provide an empirical framework for evaluating deployment readiness of wearable FoG detection systems and offer concrete training strategy recommendations for clinical translation.
Cross-National Disparities in End-of-Life Outcomes and Care Strategies for Parkinson's Disease: A Systematic Review and Meta-Analysis of China and the United States
Lin, Wei, Yang, Likun , Wu, Yan, Wang, Dan, Lin, Wei, Yang, Likun, Wang, Dan

Mayo Clinic

No container title available | January 12, 2026 | Volume: No volume available | Issue: No issue available | Pages: No page available

10.21203/rs.3.rs-8124753/v1 Get It

Abstract Background Parkinson's disease (PD) represents a growing global health challenge, with prevalence doubling from 2.5 million cases in 1990 to 6.1 million in 2016, and projections suggesting it will exceed 12 million by 2040. This exponential growth has been termed the "Parkinson's pandemic." China and the United States, which together account for nearly half of the global PD population, have developed markedly different approaches to end-of-life (EOL) care, shaped by their distinct healthcare systems and cultural contexts. Understanding these differences is crucial for developing effective global strategies. This systematic review and meta-analysis comprehensively compares advanced-stage PD care between these two nations to quantify disparities and inform evidence-based policy interventions. Methods We conducted a rigorous systematic review following PRISMA 2020 guidelines, searching six databases (PubMed, Embase, Scopus, Web of Science, CNKI, Wanfang) from January 2015 to October 2025. The timeframe was selected to capture contemporary care patterns following major healthcare reforms in both countries (Affordable Care Act implementation in the US; Healthy China 2030 initiative). Inclusion criteria encompassed peer-reviewed studies reporting on advanced PD (Hoehn-Yahr stage ≥ 3) with outcome measures including survival, quality of life (QoL), palliative care access, caregiver burden, rehabilitation utilization, or healthcare costs. The sample size threshold (n ≥ 100) balanced statistical power with inclusivity. Two independent reviewers screened 3,847 records with excellent agreement (Cohen's κ = 0.82–0.89). Quality assessment employed the Newcastle-Ottawa Scale (NOS ≥ 8 indicating high quality) and ROBINS-I for observational studies, with GRADE criteria determining evidence certainty. Meta-analyses utilized random-effects models with comprehensive heterogeneity assessment (I² statistics, meta-regression) and publication bias evaluation (Egger's test, trim-and-fill analysis). Sensitivity analyses included high-quality studies only and examined the influence of large database studies. Results Our analysis of 42 high-quality studies encompassing over 125,000 patients revealed profound and systematic cross-national disparities across multiple domains. In terms of care access, US patients demonstrated significantly higher hospice enrollment rates (28–35% vs. <1% in China, p < 0.001; HIGH certainty evidence) and greater access to specialized neuropalliative programs (15% of US centers vs. <1% in China). These differences translated into measurable quality of life disparities, with Chinese patients scoring 17.6 points worse on the PDQ-39 scale (95% CI 14.2–21.0; MODERATE certainty, I²=68%), a difference exceeding the minimal clinically important difference nearly four-fold. The burden on families showed equally striking patterns: Chinese families provided 94% of EOL care compared to 61% in the US, resulting in significantly higher caregiver depression rates (47% vs. 32%, OR = 1.89, 95% CI 1.45–2.47) and burden scores (Zarit difference: 10.2 points, 95% CI 7.8–12.6). Access to interventions revealed additional gaps, with only 9.2% of Chinese patients receiving physical therapy versus 14.4% in the US (OR = 0.61, 95% CI 0.49–0.76), and a 5.25-fold disparity in deep brain stimulation access (4.2% vs. 0.8%). Economic analysis demonstrated substantial out-of-pocket burden differences: Chinese families paid 58% of total costs ($4,200 annually) versus 12% for US families ($2,800). Meta-regression identified healthcare system characteristics as the primary driver of these disparities (R²=0.42, p < 0.001). Conclusions This comprehensive analysis reveals that systematic differences in healthcare infrastructure, insurance coverage, and cultural care models produce markedly divergent EOL experiences for PD patients in China versus the United States. These findings have immediate policy implications. For China, we recommend: (1) establishing pilot neuropalliative programs in tertiary centers with phased national expansion; (2) developing caregiver support infrastructure including respite care and training programs; (3) leveraging telemedicine platforms to address urban-rural disparities. For the United States, priorities include: (1) addressing racial disparities in rehabilitation access through culturally competent outreach; (2) expanding Medicare coverage for maintenance therapy; (3) earlier integration of neuropalliative care. Implementation science research is urgently needed to test culturally adapted interventions. As global PD prevalence accelerates, lessons from this bilateral comparison provide a framework for other nations confronting similar challenges in neurological care delivery.
A Biomechanics-Guided and Time–Frequency Collaborative Deep Learning Framework for Parkinsonian Gait Severity Assessment
Lin, Wei, Zhou, Tianqi , Yang, Qiwen

Mayo Clinic

Mathematics | December 29, 2025 | Volume: 14 | Issue: 1 | Pages: 89

10.3390/math14010089 Get It

Parkinson’s Disease (PD) is a neurodegenerative disorder in which gait abnormalities serve as key indicators of motor impairment and disease progression. Although wearable sensor-based gait analysis has advanced, existing methods still face challenges in modeling multi-sensor spatial relationships, extracting adaptive multi-scale temporal features, and effectively integrating time–frequency information. To address these issues, this paper proposes a multi-sensor gait neural network that integrates biomechanical priors with time–frequency collaborative learning for the automatic assessment of PD gait severity. The framework consists of three core modules: (1) BGS-GAT (Biomechanics-Guided Graph Attention Network), which constructs a sensor graph based on plantar anatomy and explicitly models inter-regional force dependencies via graph attention; (2) AMS-Inception1D (Adaptive Multi-Scale Inception-1D), which employs dilated convolutions and channel attention to extract multi-scale temporal features adaptively; and (3) TF-Branch (Time–Frequency Branch), which applies Real-valued Fast Fourier Transform (RFFT) and frequency-domain convolution to capture rhythmic and high-frequency components, enabling complementary time–frequency representation. Experiments on the PhysioNet multi-channel foot pressure dataset demonstrate that the proposed model achieves 0.930 in accuracy and 0.925 in F1-score for four-class severity classification, outperforming state-of-the-art deep learning models.
Modeling Tdp‐43 dependent misprocessing of Stmn2 in vivo
Pickles, Sarah , Gaiani, Luca, Lin, Lilian Tsai‐Wei , Jansen‐West, Karen , Tong, Jimei, Rosso, Giulia del, Daughrity, Lillian M. , Zhang, Yongjie, Petrucelli, Leonard, Petrucelli, Leonard

Mayo Clinic

Alzheimer's & Dementia | December 24, 2025 | Volume: 21 | Issue: S1 | Pages: No page available

10.1002/alz70855_101194 Get It

Abstract Background Some cases of Alzheimer's Disease (AD) and Frontotemporal Dementia (FTD) feature the mislocalization of the normally nuclear TAR DNA binding protein (TDP‐43) to the cytoplasm where it is incorporated into inclusions in affected brain regions. TDP‐43 plays a crucial role in the suppression of cryptic exons in the mature RNA of many targets in both AD and FTD. One such target, Stathmin‐2 (STMN2), is misspliced in the absence of TDP‐43 leading to a reduction of STMN2 RNA and protein while the levels of the cryptic splice variant is increased. Cryptic exons are not conserved across species, meaning previous Tdp‐43 models did not capture Stmn2 processing defects and Stmn2 loss of function (LOF). We hypothesize that STMN2 LOF contributes to cognitive phenotypes observed in AD and FTD. Method We have generated a novel mouse that more accurately models human disease mechanisms by expressing short‐hairpin (shRNA) via adeno associated virus in the central nervous system of a recently described humanized Stmn2 mouse model in which Stmn2 splicing is dependent on Tdp‐43. In this mouse we will simultaneously model Tdp‐43 loss of function and aberrant Stmn2 splicing. Result Our preliminary experiments demonstrate that humanized Stmn2 mice expressing shRNA against Tardbp have a reduction of Tardbp RNA and TDP‐43 protein compared to mice expressing a non‐targeting shRNA. Tdp‐43 knock‐down is accompanied by cryptic splicing of endogenous mouse targets, Sortilin 1 and Translin . As anticipated, we also detected cryptic Stmn2 splicing, suggesting that correct Stmn2 splicing is dependent on Tdp‐43 in this model. We will further characterize motor and cognitive functions in addition to neurodegeneration and gliosis in these animals. Conclusion We anticipate this novel model, and our subsequent studies will identify relevant disease mechanisms and provide an invaluable tool with which to investigate therapies for AD and FTD.
Improving Spinal Cord Stimulation Patient Triage: Random Forest Model with Custom Evaluation Functions
Lin, Chen-Ta , Chao, Wei-Lin , Huang, Yu-Li , Liang, Sheng‐Fu , Lin, Chen-Ta , Chao, Wei-Lin , Huang, Yu-Li

Mayo Clinic

Studies in Health Technology and Informatics | August 7, 2025 | Volume: No volume available | Issue: No issue available | Pages: No page available

10.3233/shti251024 | PMID 40776042 Get It

Effective healthcare referral triage poses a significant challenge, especially extracting referral notes on time for scheduling. To offset this issue, this study aims to develop an optimized Random Forest classifier to recreate the referral notes content and identify patients requiring Spinal Cord Stimulation procedures. A performance index was introduced to improve accuracy, while a consistency index was designed to prevent overfitting. The optimized classifier, with the optimal selection at the number of estimates being 5 trees and tree depth being 4, achieved an accuracy of 0.74 with a lower standard deviation, demonstrating its reliability and effectiveness.
Follistatin drives neuropathic pain in mice through IGF1R signaling in nociceptive neurons
Jiang, Bao‐Chun , Ling, Yuejuan , Xu, Meng-Lin , Gu, Jun , Sha, Wei‐Lin, Tian, Tian , Bai, Xue‐Hui , Li, Nan , Jiang, Changyu , Ma, Ling-Jie , Zhang, Zhijun , Qin, Yibin , Zhu, Meixuan, Yuan, Hong-Jie, Wu, Long‐Jun , Ling, Yue-Juan , Wu, Xiao-Bo , Bai, Xue-Hui , Jiang, Chang-Yu , Chen, Ouyang , Ma, Ling-Jie , Zhang, Zhi-Jun , Wu, Long-Jun , Ji, Ru-Rong , Gao, Yong-Jing , Xu, Meng-Lin , Li, Nan

Mayo Clinic

Science Translational Medicine | October 16, 2024 | Volume: 16 | Issue: 769 | Pages: No page available

10.1126/scitranslmed.adi1564 | PMID 39413164 Get It

Neuropathic pain is a debilitating chronic condition that lacks effective treatment. The role of cytokine- and chemokine-mediated neuroinflammation in its pathogenesis has been well documented. Follistatin (FST) is a secreted protein known to antagonize the biological activity of cytokines in the transforming growth factor–β (TGF-β) superfamily. The involvement of FST in neuropathic pain and the underlying mechanism remain largely unknown. Here, we report that FST was up-regulated in A-fiber sensory neurons after spinal nerve ligation (SNL) in mice. Inhibition or deletion of FST alleviated neuropathic pain and reduced the nociceptive neuron hyperexcitability induced by SNL. Conversely, intrathecal or intraplantar injection of recombinant FST, or overexpression of FST in the dorsal root ganglion (DRG) neurons, induced pain hypersensitivity. Furthermore, exogenous FST increased neuronal excitability in nociceptive neurons. The biolayer interferometry (BLI) assay and coimmunoprecipitation (co-IP) demonstrated direct binding of FST to the insulin-like growth factor–1 receptor (IGF1R), and IGF1R inhibition reduced FST-induced activation of extracellular signal–regulated kinase (ERK) and protein kinase B (AKT), as well as neuronal hyperexcitability. Further co-IP analysis revealed that the N-terminal domain of FST exhibits the highest affinity for IGF1R, and blocking this interaction with a peptide derived from FST attenuated Nav1.7-mediated neuronal hyperexcitability and neuropathic pain after SNL. In addition, FST enhanced neuronal excitability in human DRG neurons through IGF1R. Collectively, our findings suggest that FST, released from A-fiber neurons, enhances Nav1.7-mediated hyperexcitability of nociceptive neurons by binding to IGF1R, making it a potential target for neuropathic pain treatment.
A phase 2 study of thalidomide for the treatment of radiation-induced blood-brain barrier injury
Cheng, Jinping , Jiang, Jingru, He, Baixuan, Lin, Wei‐Jye , Li, Yi , Li, Honghong , Huang, Xiaolong , Xie, Jiatian, Zhang, Zhan , Yang, Yuhua , Xu, Yongteng , Hu, Xia , Wu, Minyi, Zhuo, Xiaohuang , Pei, Yu , Rong, Xiaoming , Ye, Xiaojing , Saw, Phei Er , Wu, Long‐Jun , Mai, Hai‐Qiang , Tang, Yamei , Li, Honghong , Zhang, Zhan , Xu, Yongteng , Shi, Zhongshan , Ye, Xiaojing , Saw, Phei Er , Wu, Long-Jun , Simone, Charles B , Chua, Melvin L K , Tang, Yamei , Duan, Jingjing, Cai, Jinhua , Hu, Xia, Liu, Qiang , Yu, Pei, Simone, Charles B. , Chua, Melvin L.K. , Mai, Hai‐Qiang

Mayo Clinic

Science Translational Medicine | February 22, 2023 | Volume: 15 | Issue: 684 | Pages: No page available

10.1126/scitranslmed.abm6543 | PMID 36812346 Get It

Radiation-induced brain injury (RIBI) is a debilitating sequela after radiotherapy to treat head and neck cancer, and 20 to 30% of patients with RIBI fail to respond to or have contraindications to the first-line treatments of bevacizumab and corticosteroids. Here, we reported a Simon’s minmax two-stage, single-arm, phase 2 clinical trial (NCT03208413) to assess the efficacy of thalidomide in patients with RIBI who were unresponsive to or had contraindications to bevacizumab and corticosteroid therapies. The trial met its primary endpoint, with 27 of 58 patients enrolled showing ≥25% reduction in the volume of cerebral edema on fluid-attenuated inversion recovery–magnetic resonance imaging (FLAIR-MRI) after treatment (overall response rate, 46.6%; 95% CI, 33.3 to 60.1%). Twenty-five (43.1%) patients demonstrated a clinical improvement based on the Late Effects Normal Tissues–Subjective, Objective, Management, Analytic (LENT/SOMA) scale, and 36 (62.1%) experienced cognitive improvement based on the Montreal Cognitive Assessment (MoCA) scores. In a mouse model of RIBI, thalidomide restored the blood-brain barrier and cerebral perfusion, which were attributed to the functional rescue of pericytes secondary to elevation of platelet-derived growth factor receptor β (PDGFRβ) expression by thalidomide. Our data thus demonstrate the therapeutic potential of thalidomide for the treatment of radiation-induced cerebral vasculature impairment.