Context-Aware Time Series Imputation for Multi-Analyte Clinical Data

Yin, Kejing; Feng, Liaoliao; Cheung, William K.

doi:10.1007/s41666-020-00075-3

Cited by 5 publications

(4 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our study delineated the frequency and clinical characteristics of treatment-resistant MG in mainland China, despite patients receiving adequate doses of corticosteroids and at least two concurrent immunosuppressive drugs for a minimum of one year. A key strength of this study lies in addressing the ambiguity surrounding the assessment of clinical symptoms in the ICG criteria by incorporating ndings from various original researches [23][24][25][26]. Our approach focused on utilizing operational assessment of clinical symptoms, in contrast to previous de nitions of refractory MG.…”

Section: Discussionmentioning

confidence: 99%

Clinical characteristics and outcome of treatment-refractory myasthenia gravis -a retrospective study

Li,

Yang,

Dong

et al. 2024

Preprint

View full text Add to dashboard Cite

Objective This study aimed to investigate the frequency, clinical characteristics, and outcomes of treatment-refractory myasthenia gravis (MG) in a Chinese cohort. Methods A retrospective cohort of 277 MG patients was conducted between August 2016 and May 2023. Patients were classified as refractory if their Myasthenia Gravis Foundation of America post-intervention status (MGFA-PIS) was categorized as “unchanged” or “worse”. Additionally, patients with persistent symptoms and functional limitations despite at least 12 months of concurrent immunosuppressive therapy, including adequately dosed steroids and two other immunosuppressive drugs, were also classified as refractory. The clinical features and outcomes at the end of follow-up of drug-refractory patients were compared with those of drug-responsive patients. Results Of the 277 patients, 36 were unequivocally diagnosed with refractory MG. Treatment-refractory patients frequently presented at a younger age, with a generalized form of the disease, and with bulbar/or respiratory symptoms at onset. These patients also exhibited higher disease severity than non-drug-refractory patients. Furthermore, patients classified as drug-resistant experienced a longer interval between disease onset and the initiation of immunotherapy. At the end of follow-up, poor outcome was more frequently found in treatment-refractory MG patients. Conclusion This study found that 13% of MG patients were classified as drug-refractory. There is a need for new, more specific drugs to treat drug-refractory MG patients.

show abstract

Section: Discussionmentioning

confidence: 99%

Clinical characteristics and outcome of treatment-refractory myasthenia gravis -a retrospective study

Li,

Yang,

Dong

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…The design principles of some of the challenge participating teams may also inform future algorithms to potentially overcome these issues. For example, extending on the global context vector idea by Team HKBU [ 20 ], one can add a time-varying bias vector to RNN and similar models in order to capture and quantify the underlying bias factors contributing to the observed data quality issues.…”

Section: Discussionmentioning

confidence: 99%

“…Team Hong Kong Baptist University (HKBU) [ 20 ] proposed Context-Aware Time Series Imputation (CATSI) to explicitly capture the patient admission condition by a global context vector to augment a bi-directional recurrent neural network (RNN). They used multilayer perceptron (MLP) to learn the model for cross-sectional imputation and used a fusion layer to integrate the temporal imputation and the cross-sectional imputation.…”

Section: Challenge Participating Systemsmentioning

confidence: 99%

See 1 more Smart Citation

Evaluating the state of the art in missing data imputation for clinical data

Luo

2021

Briefings in Bioinformatics

View full text Add to dashboard Cite

Clinical data are increasingly being mined to derive new medical knowledge with a goal of enabling greater diagnostic precision, better-personalized therapeutic regimens, improved clinical outcomes and more efficient utilization of health-care resources. However, clinical data are often only available at irregular intervals that vary between patients and type of data, with entries often being unmeasured or unknown. As a result, missing data often represent one of the major impediments to optimal knowledge derivation from clinical data. The Data Analytics Challenge on Missing data Imputation (DACMI) presented a shared clinical dataset with ground truth for evaluating and advancing the state of the art in imputing missing data for clinical time series. We extracted 13 commonly measured blood laboratory tests. To evaluate the imputation performance, we randomly removed one recorded result per laboratory test per patient admission and used them as the ground truth. DACMI is the first shared-task challenge on clinical time series imputation to our best knowledge. The challenge attracted 12 international teams spanning three continents across multiple industries and academia. The evaluation outcome suggests that competitive machine learning and statistical models (e.g. LightGBM, MICE and XGBoost) coupled with carefully engineered temporal and cross-sectional features can achieve strong imputation performance. However, care needs to be taken to prevent overblown model complexity. The challenge participating systems collectively experimented with a wide range of machine learning and probabilistic algorithms to combine temporal imputation and cross-sectional imputation, and their design principles will inform future efforts to better model clinical missing data.

show abstract

Deep imputation of missing values in time series health data: A review with benchmarking

Kazijevs

Samad

2023

Journal of Biomedical Informatics

View full text Add to dashboard Cite

Context-Aware Time Series Imputation for Multi-Analyte Clinical Data

Cited by 5 publications

References 11 publications

Clinical characteristics and outcome of treatment-refractory myasthenia gravis -a retrospective study

Clinical characteristics and outcome of treatment-refractory myasthenia gravis -a retrospective study

Evaluating the state of the art in missing data imputation for clinical data

Deep imputation of missing values in time series health data: A review with benchmarking

Contact Info

Product

Resources

About