Integrating landmark modeling framework and machine learning algorithms for dynamic prediction of tuberculosis treatment outcomes
Закрыть
Conţinutul numărului revistei
Articolul precedent
Articolul urmator
209 0
SM ISO690:2012
KHEIRANDISH, Maryam, CATANZARO, Donald G., KRUDU, V., ZHANG, Shengfan. Integrating landmark modeling framework and machine learning algorithms for dynamic prediction of tuberculosis treatment outcomes. In: Journal of the American Medical Informatics Association, 2022, nr. 5(29), pp. 900-908. ISSN 1067-5027. DOI: https://doi.org/10.1093/jamia/ocac003
EXPORT metadate:
Google Scholar
Crossref
CERIF

DataCite
Dublin Core
Journal of the American Medical Informatics Association
Numărul 5(29) / 2022 / ISSN 1067-5027 /ISSNe 1527-974X

Integrating landmark modeling framework and machine learning algorithms for dynamic prediction of tuberculosis treatment outcomes

DOI: https://doi.org/10.1093/jamia/ocac003

Pag. 900-908

Kheirandish Maryam1, Catanzaro Donald G.1, Krudu V.23, Zhang Shengfan1
 
1 University of Arkansas,
2 Institute of Phtysiopneumology „Chiril Draganiuc”,
3 ”Nicolae Testemițanu” State University of Medicine and Pharmacy
 
Disponibil în IBN: 20 mai 2022


Rezumat

 Objective: This study aims to establish an informative dynamic prediction model of treatment outcomes using follow-up records of tuberculosis (TB) patients, which can timely detect cases when the current treatment plan may not be effective. Materials and Methods: We used 122 267 follow-up records from 17 958 new cases of pulmonary TB in the Republic of Moldova. A dynamic prediction framework integrating landmark modeling and machine learning algorithms was designed to predict patient outcomes during the course of treatment. Sensitivity and positive predictive value (PPV) were calculated to evaluate performance of the model at critical time points. New measures were defined to determine when follow-up laboratory tests should be conducted to obtain most informative results. Results: The random-forest algorithm performed better than support vector machine and penalized multinomial logistic regression models for predicting TB treatment outcomes. For all 3 outcome classes (ie, cured, not cured, and died after 24 months following treatment initiation), sensitivity and PPV of prediction models improved as more follow-up information was collected. Specifically, sensitivity and PPV increased from 0.55 to 0.84 and from 0.32 to 0.88, respectively, for the not cured class. Conclusion: The dynamic prediction framework utilizes longitudinal laboratory test results to predict patient outcomes at various landmarks. Sputum culture and smear results are among the important variables for prediction; however, the most recent sputum result is not always the most informative one. This framework can potentially facilitate a more effective treatment monitoring program and provide insights for policymakers toward improved guidelines on follow-up tests.

Cuvinte-cheie
Follow-Up Studies, mortality, supervised machine learning, treatment outcome, tuberculosis