Browsing by Author "Lin, Jeffrey"
Now showing 1 - 9 of 9
- Results Per Page
- Sort Options
Item Open Access Deep learning for the dynamic prediction of multivariate longitudinal and survival data.(Statistics in medicine, 2022-03-28) Lin, Jeffrey; Luo, ShengThe joint model for longitudinal and survival data improves time-to-event predictions by including longitudinal outcome variables in addition to baseline covariates. However, in practice, joint models may be limited by parametric assumptions in both the longitudinal and survival submodels. In addition, computational difficulties arise when considering multiple longitudinal outcomes due to the large number of random effects to be integrated out in the full likelihood. In this article, we discuss several recent machine learning methods for incorporating multivariate longitudinal data for time-to-event prediction. The presented methods use functional data analysis or convolutional neural networks to model the longitudinal data, both of which scale well to multiple longitudinal outcomes. In addition, we propose a novel architecture based on the transformer neural network, named TransformerJM, which jointly models longitudinal and time-to-event data. The prognostic abilities of each model are assessed and compared through both simulation and real data analysis on Alzheimer's disease datasets. Specifically, the models were evaluated based on their ability to dynamically update predictions as new longitudinal data becomes available. We showed that TransformerJM improves upon the predictive performance of existing methods across different scenarios.Item Open Access Expanded and Independent Spanish validation of the MDS ‐ Non Motor Rating Scale(Movement Disorders Clinical Practice) Cubo, Esther; Luo, Sheng; Martínez-Martín, Pablo; Stebbins, Glenn T; Lin, Jeffrey; Choi, Dongrak; García-Bustillo, Alvaro; Mir, Pablo; Santos-Garcia, Diego; Serrano-Dueñas, Marcos; Rodriguez-Violante, Mayela; Singer, Carlos; and the Spanish MDS‐NMS Validation Study GroupItem Open Access Functional survival forests for multivariate longitudinal outcomes: Dynamic prediction of Alzheimer's disease progression.(Statistical methods in medical research, 2020-07-29) Lin, Jeffrey; Li, Kan; Luo, ShengThe random survival forest (RSF) is a non-parametric alternative to the Cox proportional hazards model in modeling time-to-event data. In this article, we developed a modeling framework to incorporate multivariate longitudinal data in the model building process to enhance the predictive performance of RSF. To extract the essential features of the multivariate longitudinal outcomes, two methods were adopted and compared: multivariate functional principal component analysis and multivariate fast covariance estimation for sparse functional data. These resulting features, which capture the trajectories of the multiple longitudinal outcomes, are then included as time-independent predictors in the subsequent RSF model. This non-parametric modeling framework, denoted as functional survival forests, is better at capturing the various trends in both the longitudinal outcomes and the survival model which may be difficult to model using only parametric approaches. These advantages are demonstrated through simulations and applications to the Alzheimer's Disease Neuroimaging Initiative.Item Open Access Prognostic Modeling of Parkinson's Disease Progression Using Early Longitudinal Patterns of Change.(Movement disorders : official journal of the Movement Disorder Society, 2021-07-30) Ren, Xuehan; Lin, Jeffrey; Stebbins, Glenn T; Goetz, Christopher G; Luo, ShengBackground
Predicting Parkinson's disease (PD) progression may enable better adaptive and targeted treatment planning.Objective
Develop a prognostic model using multiple, easily acquired longitudinal measures to predict temporal clinical progression from Hoehn and Yahr (H&Y) stage 1 or 2 to stage 3 in early PD.Methods
Predictive longitudinal measures of PD progression were identified by the joint modeling method. Measures were extracted by multivariate functional principal component analysis methods and used as covariates in Cox proportional hazards models. The optimal model was developed from the Parkinson's Progression Marker Initiative (PPMI) data set and confirmed with external validation from the Longitudinal and Biomarker Study in PD (LABS-PD) study.Results
The proposed prognostic model with longitudinal information of selected clinical measures showed significant advantages in predicting PD temporal progression in comparison to a model with only baseline information (iAUC = 0.812 vs. 0.743). The modeling results allowed the development of a prognostic index for categorizing PD patients into low, mid, and high risk of progression to HY 3 that is offered to facilitate physician-patient discussion on prognosis.Conclusion
Incorporating longitudinal information of multiple clinical measures significantly enhances predictive performance of prognostic models. Furthermore, the proposed prognostic index enables clinicians to classify patients into different risk groups, which could be adaptively updated as new longitudinal information becomes available. Modeling of this type allows clinicians to utilize observational data sets that inform on disease natural history and specifically, for precision medicine, allows the insertion of a patient's clinical data to calculate prognostic estimates at the individual case level. © 2021 International Parkinson and Movement Disorder Society.Item Open Access Successful use of the Unified Dyskinesia Rating Scale regardless of PD- or dyskinesia-duration.(Parkinsonism & related disorders, 2019-10) Ren, Xuehan; Lin, Jeffrey; Luo, Sheng; Goetz, Christopher G; Stebbins, Glenn T; Cubo, EstherOBJECTIVE:We assessed differential item functioning (DIF) in the Unified Dyskinesia Rating Scale (UDysRS) to evaluate bias risk from the duration of Parkinson's Disease (PD) and duration of dyskinesia. BACKGROUND:Assessing DIF is a core validation step for rating scales. If DIF is present for an item, interpretation must consider influences from the tested covariates. DIF can be uniform or non-uniform, depending on the consistency of influence from the given covariate across all levels of dyskinesia. METHODS:Using a large UDysRS database (N = 2313), uniform and non-uniform DIF related to the duration of PD and/duration of dyskinesia were tested. Unidimensionality of UDysRS was first confirmed using confirmatory factor analysis. DIF analysis was conducted using two independent latent models. DIF in an item was confirmed if both methods independently identified DIF at a significance level using Bonferroni correction. McFadden pseudo R^2 measured clinical relevancy of DIF magnitude (negligible, moderate, and large) for items identified with DIF, and items with DIF were considered clinically relevant if they exceeded a negligible designation. RESULTS:Most items did not show uniform or non-uniform DIF based on PD and dyskinesia duration in isolation or in combination. For all items where DIF was identified, the magnitude statistic was in the negligible range (McFadden pseudo R^2 < 0.035) and the combined impact of multiple identified DIF items on UDysRS likewise did not exceed the negligible designation. CONCLUSION:The absence of clinically relevant DIF suggests that the UDysRS can be applied across all patients regardless of their PD- or dyskinesia-duration.Item Open Access Unmasking the sky: high-resolution PM2.5 prediction in Texas using machine learning techniques.(Journal of exposure science & environmental epidemiology, 2024-04) Zhang, Kai; Lin, Jeffrey; Li, Yuanfei; Sun, Yue; Tong, Weitian; Li, Fangyu; Chien, Lung-Chang; Yang, Yiping; Su, Wei-Chung; Tian, Hezhong; Fu, Peng; Qiao, Fengxiang; Romeiko, Xiaobo Xue; Lin, Shao; Luo, Sheng; Craft, ElenaBackground
Although PM2.5 (fine particulate matter with an aerodynamic diameter less than 2.5 µm) is an air pollutant of great concern in Texas, limited regulatory monitors pose a significant challenge for decision-making and environmental studies.Objective
This study aimed to predict PM2.5 concentrations at a fine spatial scale on a daily basis by using novel machine learning approaches and incorporating satellite-derived Aerosol Optical Depth (AOD) and a variety of weather and land use variables.Methods
We compiled a comprehensive dataset in Texas from 2013 to 2017, including ground-level PM2.5 concentrations from regulatory monitors; AOD values at 1-km resolution based on images retrieved from the MODIS satellite; and weather, land-use, population density, among others. We built predictive models for each year separately to estimate PM2.5 concentrations using two machine learning approaches called gradient boosted trees and random forest. We evaluated the model prediction performance using in-sample and out-of-sample validations.Results
Our predictive models demonstrate excellent in-sample model performance, as indicated by high R2 values generated from the gradient boosting models (0.94-0.97) and random forest models (0.81-0.90). However, the out-of-sample R2 values fall within a range of 0.52-0.75 for gradient boosting models and 0.44-0.69 for random forest models. Model performance varies slightly across years. A generally decreasing trend in predicted PM2.5 concentrations over time is observed in Eastern Texas.Impact statement
We utilized machine learning approaches to predict PM2.5 levels in Texas. Both gradient boosting and random forest models perform well. Gradient boosting models perform slightly better than random forest models. Our models showed excellent in-sample prediction performance (R2 > 0.9).Item Open Access Validation of the Arabic Version of the Movement Disorder Society-Unified Parkinson's Disease Rating Scale.(Movement disorders : official journal of the Movement Disorder Society, 2022-04) Khalil, Hanan; Aldaajani, Zakiyah F; Aldughmi, Mayis; Al-Sharman, Alham; Mohammad, Tareq; Mehanna, Raja; El-Jaafary, Shaimaa I; Dahshan, Ahmed; Ben Djebara, Mouna; Kamel, Walaa A; Amer, Hanan A; Farghal, Mohammed; Abdulla, Fatema; Al-Talai, Nouf; Snineh, Muneer Abu; Farhat, Nouha; Jamali, Fatima A; Matar, Rawan K; Abdelraheem, Heba S; Ghonimi, Nesma AM; Al-Melh, Mishal Abu; Elbhrawy, Sonia; Alotaibi, Majid S; Elaidy, Shaimaa A; Almuammar, Shahad A; Al-Hashel, Jasem Y; Gouider, Riadh; Samir, Hatem; Mhiri, Chokri; Skorvanek, Matej; Lin, Jeffrey; Martinez-Martin, Pablo; Stebbins, Glenn T; Luo, Sheng; Goetz, Christopher G; Bajwa, Jawad ABackground
The Movement Disorder Society-Unified Parkinson's Disease Rating Scale (MDS-UPDRS) has become the gold standard for evaluating different domains in Parkinson's disease (PD), and it is commonly used in clinical practice, research, and clinical trials.Objectives
The objectives are to validate the Arabic-translated version of the MDS-UPDRS and to assess its factor structure compared with the English version.Methods
The study was carried out in three phases: first, the English version of the MDS-UPDRS was translated into Arabic and subsequently back-translated into English by independent translation team; second, cognitive pretesting of selected items was performed; third, the Arabic version was tested in over 400 native Arabic-speaking PD patients. The psychometric properties of the translated version were analyzed using confirmatory factor analysis (CFA) as well as exploratory factor analysis (EFA).Results
The factor structure of the Arabic version was consistent with that of the English version based on the high CFIs for all four parts of the MDS-UPDRS in the CFA (CFI ≥0.90), confirming its suitability for use in Arabic.Conclusions
The Arabic version of the MDS-UPDRS has good construct validity in Arabic-speaking patients with PD and has been thereby designated as an official MDS-UPDRS version. The data collection methodology among Arabic-speaking countries across two continents of Asia and Africa provides a roadmap for validating additional MDS rating scale initiatives and is strong evidence that underserved regions can be energically mobilized to promote efforts that apply to better clinical care, education, and research for PD. © 2022 International Parkinson and Movement Disorder Society.Item Open Access Validation of the Finnish Version of the Unified Dyskinesia Rating Scale.(European neurology, 2021-07-14) Kaasinen, Valtteri; Scheperjans, Filip; Kärppä, Mikko; Korpela, Jaana; Brück, Anna; Sipilä, Jussi OT; Joutsa, Juho; Järvelä, Juha; Eerola-Rautio, Johanna; Martikainen, Mika H; Airaksinen, Katja; Stebbins, Glenn T; Martinez-Martin, Pablo; Goetz, Christopher G; Lin, Jeffrey; Luo, Sheng; Pekkonen, EeroIntroduction
The Unified Dyskinesia Rating Scale (UDysRS) was developed to provide a comprehensive rating tool of dyskinesia in Parkinson's disease (PD). Because dyskinesia therapy trials involve multicenter studies, having a scale that is validated in multiple non-English languages is pivotal to international efforts to treat dyskinesia. The aim of the present study was to organize and perform an independent validation of the UDysRS Finnish version.Methods
The UDysRS was translated into Finnish and then back-translated into English using 2 independent teams. Cognitive pretesting was conducted on the Finnish version and required modifications to the structure or wording of the translation. The final Finnish version was administered to 250 PD patients whose native language is Finnish. The data were analyzed to assess the confirmatory factor structure to the Spanish UDysRS (the reference standard). Secondary analyses included an exploratory factor analysis (EFA), independent of the reference standard.Results
The comparative fit index (CFI), in comparison with the reference standard factor structure, was 0.963 for Finnish. In the EFA, where variability from sample to sample is expected, isolated item differences of factor structure were found between the Finnish and Reference Standard versions of the UDysRS. These subtle differences may relate to differences in sample composition or variations in disease status.Conclusion
The overall factor structure of the Finnish version was consistent with that of the reference standard, and it can be designated as the official version of the UDysRS for Finnish speaking populations.Item Open Access Validation of the Polish version of the Unified Dyskinesia Rating Scale (UDysRS).(Neurologia i neurochirurgia polska, 2021-02-02) Siuda, Joanna; Boczarska-Jedynak, Magdalena; Budrewicz, Sławomir; Dulski, Jarosław; Figura, Monika; Fiszer, Urszula; Gajos, Agata; Gorzkowska, Agnieszka; Koziorowska-Gawron, Ewa; Koziorowski, Dariusz; Krygowska-Wajs, Anna; Rudzińska-Bar, Monika; Sławek, Jarosław; Toś, Mateusz; Wójcik-Pędziwiatr, Magdalena; Lin, Jeffrey; Luo, Sheng; Martinez-Martin, Pablo; Stebbins, Glenn T; Goetz, Christopher G; Opala, Grzegorz; Polish UDysRS Validation Task ForceBackground
In 2008, the Movement Disorders Society published the Unified Dyskinesia Rating Scale (UDysRS). This has become the established tool for assessing the severity and disability associated with dyskinesia in patients with Parkinson's Disease (PD). We translated and validated the Polish version of the UDysRS, explored its dimensionality, and compared it to the Spanish version, which is the Reference Standard for UDysRS translations.Material and methods
The UDysRS was translated into Polish by a team led by JS and GO. The back-translation, completed by colleagues fluent in both Polish and English who were not involved in the original translation, was reviewed and approved by the Executive Committee of the MDS Rating Scales Programme. Then the translated version of the UDysRS underwent cognitive pretesting, and the translation was modified based on the results. The approved version was considered to be the Official Working Document of the Polish UDysRS and was tested on 250 Polish PD patients recruited at movement disorder centres. Data was compared to the Reference Standard used for validating UDysRS translations.Results
The overall factor structure of the Polish version was consistent with that of the Reference Standard version, as evidenced by the high Confirmatory Fit Index score (CFI = 0.98). The Polish UDysRS was thus confirmed to share a common factor structure with the Reference Standard.Conclusions
The Official Polish UDysRS translation is recommended for use in clinical and research settings. Worldwide use of uniform rating measures offers a common ground to study similarities and differences in disease manifestations and progression across cultures.