A novel systematic pipeline for increased predictability and explainability of growth patterns in children using trajectory features

Imagem de Miniatura
Arquivos
art_MASSARA_A_novel_systematic_pipeline_for_increased_predictability_and_2023.PDF (2.76 MB)
Citações na Scopus
0
Tipo de produção
article
Data de publicação
2023
DOI
Scopus
Web of Science
PubMed
Título da Revista
ISSN da Revista
Título do Volume
Editora
ELSEVIER IRELAND LTD
Autores
MASSARA, Paraskevi
LOPEZ-DOMINGUEZ, Lorena
BOURDON, Celine
BASSANI, Diego G.
KEOWN-STONEMAN, Charles D. G.
BIRKEN, Catherine S.
MAGUIRE, Jonathon L.
SANTOS, Ina S.
BANDSMA, Robert H. J.
Citação
INTERNATIONAL JOURNAL OF MEDICAL INFORMATICS, v.177, article ID 105143, 9p, 2023
Projetos de Pesquisa
Unidades Organizacionais
Fascículo
Resumo
Objective: Longitudinal patterns of growth in early childhood are associated with health conditions throughout life. Knowledge of such patterns and the ability to predict them can lead to better prevention and improved health promotion in adulthood. However, growth analyses are characterized by significant variability, and pattern detection is affected by the method applied. Moreover, pattern labelling is typically performed based on ad hoc methods, such as visualizations or clinical experience. Here, we propose a novel pipeline using features extracted from growth trajectories using mathematical, statistical and machine-learning approaches to predict growth patterns and label them in a systematic and unequivocal manner. Methods: We extracted mathematical and clinical features from 9577 children growth trajectories embedded with machine-learning predictions of the growth patterns. We experimented with two sets of features (CAnonical Time-series Characteristics and trajectory features specific to growth), developmental periods and six machine-learning classifiers. Clinical experts provided labels for the detected patterns and decision rules were created to associate the features with the labelled patterns. The predictive capacity of the extracted features was validated on two heterogenous populations (The Applied Research Group for Kids and the 2004 Pelotas Birth Cohort, based in Canada and Brazil, respectively). Results: Features predictive ability measured by accuracy and F1 score was & GE; 80% and & GE; 0.76 respectively in both cohorts. A small number of features (n = 74) was sufficient to distinguish between growth patterns in both cohorts. Slope, intercept of the trajectory, age at peak value, start value and change of the growth measure were among the top identified features. Conclusion: Growth features can be reliably used as predictors of growth patterns and provide an unbiased understanding of growth patterns. They can be used as tool to reduce the effort to repeat analysis and variability concerning anthropometric measures, time points and analytical methods, in the context of the same or similar populations.
Palavras-chave
Growth patterns, Children, Explainability, Prediction, Trajectory, Features
URI
https://observatorio.fm.usp.br/handle/OPI/56168
Referências
  1. Antonisamy B, 2017, J PEDIATR-US, V180, P53, DOI 10.1016/j.jpeds.2016.09.059
  2. Aris IM, 2019, INT J EPIDEMIOL, V48, P157, DOI 10.1093/ije/dyy286
  3. Balantekin KN, 2018, EAT BEHAV, V31, P12, DOI 10.1016/j.eatbeh.2018.07.007
  4. Bhargava SK, 2004, NEW ENGL J MED, V350, P865, DOI 10.1056/NEJMoa035698
  5. Busert LK, 2016, J NUTR, V146, P1387, DOI 10.3945/jn.115.220137
  6. Carsley S, 2015, INT J EPIDEMIOL, V44, P776, DOI 10.1093/ije/dyu123
  7. CLEVELAND WS, 1979, J AM STAT ASSOC, V74, P829, DOI 10.2307/2286407
  8. Cole SZ, 2011, AM FAM PHYSICIAN, V83, P829
  9. Cossio-Bolaños MA, 2020, AM J HUM BIOL, V32, DOI 10.1002/ajhb.23398
  10. de Onis Mercedes, 2003, Forum Nutr, V56, P238
  11. Dietterich T. G., 2004, P 21 INT C MACH LEAR, P28, DOI [10.1145/1015330.1015428, DOI 10.1145/1015330.101542832]
  12. Fletcher RR, 2021, FRONT ARTIF INTELL, V3, DOI 10.3389/frai.2020.561802
  13. Frisk V, 2002, DEV NEUROPSYCHOL, V22, P565, DOI 10.1207/S15326942DN2203_2
  14. Fulcher B, 2017, CELL SYST, V5, P527, DOI 10.1016/j.cels.2017.10.001
  15. Gomula A., 2021, AM J HUM BIOL, V33
  16. Guyon I., 2003, Journal of Machine Learning Research, V3, P1157, DOI 10.1162/153244303322753616
  17. Herle M, 2020, EUR J EPIDEMIOL, V35, P205, DOI 10.1007/s10654-020-00615-6
  18. Imai CM, 2014, NUTR METAB CARDIOVAS, V24, P730, DOI 10.1016/j.numecd.2014.01.001
  19. Lavin A, 2022, NAT COMMUN, V13, DOI 10.1038/s41467-022-33128-9
  20. Law CM, 2002, CIRCULATION, V105, P1088, DOI 10.1161/hc0902.104677
  21. Laxer RE, 2018, PLOS ONE, V13, DOI 10.1371/journal.pone.0190405
  22. Lennon H, 2018, BMJ OPEN, V8, DOI 10.1136/bmjopen-2017-020683
  23. Leung M, 2018, PLOS ONE, V13, DOI 10.1371/journal.pone.0194565
  24. López-Domínguez L, 2023, SCI REP-UK, V13, DOI 10.1038/s41598-023-28485-4
  25. Lubba CH, 2019, DATA MIN KNOWL DISC, V33, P1821, DOI 10.1007/s10618-019-00647-x
  26. Marceau K, 2011, DEV PSYCHOL, V47, P1389, DOI 10.1037/a0023838
  27. Massara P, 2021, P 31 ANN INT C COMPU, P220
  28. Massara P, 2021, INT J EPIDEMIOL, V50, P1000, DOI 10.1093/ije/dyab021
  29. Mlakar M., 2023, PLOS ONE, V18
  30. Oda Junichi, 2009, Proceedings of the 2009 Fifth International Conference on Intelligent Information Hiding and Multimedia Signal Processing. IIH-MSP 2009, P1134, DOI 10.1109/IIH-MSP.2009.288
  31. Parent C, 2013, ACM COMPUT SURV, V45, DOI 10.1145/2501654.2501656
  32. Péneau S, 2017, J PEDIATR-US, V186, P64, DOI 10.1016/j.jpeds.2017.02.010
  33. Proust-Lima C, 2011, R USER C USER 2011, P66
  34. Proust-Lima C, 2017, J STAT SOFTW, V78, P1, DOI 10.18637/jss.v078.i02
  35. R Development Core Team, 2021, R LANG ENV STAT COMP, DOI 10.1038/sj.hdy.6800737
  36. Robinson HA, 2019, PREV MED REP, V14, DOI 10.1016/j.pmedr.2019.100834
  37. Santos IS, 2014, INT J EPIDEMIOL, V43, P1437, DOI 10.1093/ije/dyu144
  38. Santos IS, 2011, INT J EPIDEMIOL, V40, P1461, DOI 10.1093/ije/dyq130
  39. Sokol RL, 2019, INT J OBESITY, V43, P1113, DOI 10.1038/s41366-018-0194-y
  40. Speiser JL, 2019, EXPERT SYST APPL, V134, P93, DOI 10.1016/j.eswa.2019.05.028
  41. Tirosh A, 2011, NEW ENGL J MED, V364, P1315, DOI 10.1056/NEJMoa1006992
  42. Wen XZ, 2012, BMC MED RES METHODOL, V12, DOI 10.1186/1471-2288-12-38
  43. World Health Organization, 2020, LEVELS TRENDS CHILD

Coleções
Artigos e Materiais de Revistas Científicas - FM/MPR
Artigos e Materiais de Revistas Científicas - LIM/39