Machine learning for the prediction of toxicities from head and neck cancer treatment: A systematic review with meta-analysis

Imagem de Miniatura
Arquivos
art_ARAUJO_Machine_learning_for_the_prediction_of_toxicities_from_2023.PDF (4.75 MB)
Citações na Scopus
6
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
Autores
MORAES, Matheus Cardoso
PEREZ-DI-OLIVEIRA, Maria Eduarda
SILVA, Viviane Mariano da
SALDIVIA-SIRACUSA, Cristina
PEDROSO, Caique Mariano
LOPES, Marcio Ajudarte
VARGAS, Pablo Agustin
KOCHANNY, Sara
PEARSON, Alexander
Citação
ORAL ONCOLOGY, v.140, article ID 106386, 15p, 2023
Projetos de Pesquisa
Unidades Organizacionais
Fascículo
Resumo
Introduction: The aim of the present systematic review (SR) is to summarize Machine Learning (ML) models currently used to predict head and neck cancer (HNC) treatment-related toxicities, and to understand the impact of image biomarkers (IBMs) in prediction models (PMs). The present SR was conducted following the guidelines of the PRISMA 2022 and registered in PROSPERO database (CRD42020219304). Methods: The acronym PICOS was used to develop the focused review question (Can PMs accurately predict HNC treatment toxicities?) and the eligibility criteria. The inclusion criteria enrolled Prediction Model Studies (PMSs) with patient cohorts that were treated for HNC and developed toxicities. Electronic database search encompassed PubMed, EMBASE, Scopus, Cochrane Library, Web of Science, LILACS, and Gray Literature (Google Scholar and ProQuest). Risk of Bias (RoB) was assessed through PROBAST and the results were synthesized based on the data format (with and without IBMs) to allow comparison. Results: A total of 28 studies and 4,713 patients were included. Xerostomia was the most frequently investigated toxicity (17; 60.71 %). Sixteen (57.14 %) studies reported using radiomics features in combination with clinical or dosimetrics/dosiomics for modelling. High RoB was identified in 23 studies. Meta-analysis (MA) showed an area under the receiver operating characteristics curve (AUROC) of 0.82 for models with IBMs and 0.81 for models without IBMs (p value < 0.001), demonstrating no difference among IBM-and non-IBM-based models. Discussion: The development of a PM based on sample-specific features represents patient selection bias and may affect a model's performance. Heterogeneity of the studies as well as non-standardized metrics prevent proper comparison of studies, and the absence of an independent/external test does not allow the evaluation of the model's generalization ability. Conclusion: IBM-featured PMs are not superior to PMs based on non-IBM predictors. The evidence was appraised as of low certainty.
Palavras-chave
Machine learning, Prediction model studies, Convolutional neural network, Xerostomia, Mucosistis, Dysphagia, Osteoradionecrosis, Hypothyroidism, Hearing loss
URI
https://observatorio.fm.usp.br/handle/OPI/54529
Referências
  1. Abdollahi H, 2018, PHYS MEDICA, V45, P192, DOI 10.1016/j.ejmp.2017.10.008
  2. Beetz I, 2012, RADIOTHER ONCOL, V105, P101, DOI 10.1016/j.radonc.2012.03.004
  3. Beetz I, 2012, RADIOTHER ONCOL, V105, P86, DOI 10.1016/j.radonc.2011.05.010
  4. Borenstein, 2009, INTRO META ANAL
  5. Bradley AP, 1997, PATTERN RECOGN, V30, P1145, DOI 10.1016/S0031-3203(96)00142-2
  6. Buettner F, 2012, RADIOTHER ONCOL, V103, P82, DOI 10.1016/j.radonc.2012.02.006
  7. Carballo-Rubio R, 2021, J HIGH ENERGY PHYS, DOI 10.1007/JHEP05(2021)132
  8. Cheng Z, 2019, JCO CLIN CANCER INFO, V3, P1, DOI 10.1200/CCI.18.00058
  9. Chinnery T, 2021, CAN ASSOC RADIOL J, V72, P73, DOI 10.1177/0846537120942134
  10. Collins GS, 2021, BMJ OPEN, V11, DOI 10.1136/bmjopen-2020-048008
  11. Collins GS, 2015, ANN INTERN MED, V162, P55, DOI [10.7326/M14-0697, 10.1186/s12916-014-0241-z, 10.1002/bjs.9736, 10.1136/bmj.g7594, 10.1016/j.eururo.2014.11.025, 10.1016/j.jclinepi.2014.11.010, 10.7326/M14-0698, 10.1038/bjc.2014.639]
  12. Dean JA, 2017, CLIN ONCOL-UK, V29, P263, DOI 10.1016/j.clon.2016.12.001
  13. Dean J, 2018, CLIN TRANSL RAD ONCO, V8, P27, DOI 10.1016/j.ctro.2017.11.009
  14. Debray TPA, 2017, BMJ-BRIT MED J, V356, DOI 10.1136/bmj.i6460
  15. Dohopolski M, 2022, RADIOTHER ONCOL, V171, P129, DOI 10.1016/j.radonc.2022.04.016
  16. Faria VD, 2021, J DIGIT IMAGING, V34, P1237, DOI 10.1007/s10278-021-00487-6
  17. Gabrys HS, 2018, FRONT ONCOL, V8, DOI 10.3389/fonc.2018.00035
  18. Giraud P, 2019, FRONT ONCOL, V9, DOI 10.3389/fonc.2019.00174
  19. GRADEpro GDT, 2022, GRADEPRO GUID DEV TO
  20. Huang J, 2005, IEEE T KNOWL DATA EN, V17, P299, DOI 10.1109/TKDE.2005.50
  21. Humbert-Vidan L, 2021, BRIT J RADIOL, V94, DOI 10.1259/bjr.20200026
  22. Jiang W, 2019, ADV RADIAT ONCOL, V4, P401, DOI 10.1016/j.adro.2018.11.008
  23. KUTCHER GJ, 1989, INT J RADIAT ONCOL, V16, P1623, DOI 10.1016/0360-3016(89)90972-3
  24. Liu YX, 2019, QUANT IMAG MED SURG, V9, P1288, DOI 10.21037/qims.2019.07.08
  25. Men K, 2019, INT J RADIAT ONCOL, V105, P440, DOI 10.1016/j.ijrobp.2019.06.009
  26. Moons KGM, 2019, ANN INTERN MED, V170, pW1, DOI 10.7326/M18-1377
  27. Moons KGM, 2015, ANN INTERN MED, V162, pW1, DOI 10.7326/M14-0698
  28. Moons KGM, 2014, PLOS MED, V11, DOI 10.1371/journal.pmed.1001744
  29. Nakatsugawa M, 2019, INT J RADIAT ONCOL, V103, P460, DOI 10.1016/j.ijrobp.2018.09.038
  30. Nardone V, 2018, RADIOL MED, V123, P415, DOI 10.1007/s11547-017-0850-7
  31. Ouzzani M, 2016, SYST REV-LONDON, V5, DOI 10.1186/s13643-016-0384-4
  32. Page MJ, 2022, REV PANAM SALUD PUBL, V46, DOI [10.1136/bmj.n160, 10.26633/RPSP.2022.112]
  33. Pota M, 2017, ARTIF INTELL MED, V81, P41, DOI 10.1016/j.artmed.2017.03.004
  34. Rosen BS, 2018, INT J RADIAT ONCOL, V102, P1319, DOI 10.1016/j.ijrobp.2018.06.048
  35. Schunemann H, 2013, GRADE HDB GRADING QU
  36. Shamseer L, 2015, BMJ-BRIT MED J, V349, DOI [10.1136/bmj.i4086, 10.1186/2046-4053-4-1, 10.1136/bmj.g7647, 10.1136/bmj.b2535]
  37. Sheikh K, 2019, RADIAT ONCOL, V14, DOI 10.1186/s13014-019-1339-4
  38. Smyczynska U, 2021, CANCERS, V13, DOI 10.3390/cancers13215584
  39. Soares I, 2018, HEALTH TECHNOL-GER, V8, P159, DOI 10.1007/s12553-017-0204-4
  40. Ursino S, 2021, STRAHLENTHER ONKOL, V197, P209, DOI 10.1007/s00066-020-01697-7
  41. van Dijk LV, 2019, SCI REP-UK, V9, DOI 10.1038/s41598-019-48184-3
  42. van Dijk LV, 2018, RADIOTHER ONCOL, V128, P459, DOI 10.1016/j.radonc.2018.06.012
  43. van Dijk LV, 2018, RADIOTHER ONCOL, V126, P89, DOI 10.1016/j.radonc.2017.08.024
  44. van Dijk LV, 2017, INT J RADIAT ONCOL, V99, P1101, DOI 10.1016/j.ijrobp.2017.08.003
  45. van Dijk LV, 2017, RADIOTHER ONCOL, V122, P185, DOI 10.1016/j.radonc.2016.07.007
  46. Wentzel A, 2020, RADIOTHER ONCOL, V148, P245, DOI 10.1016/j.radonc.2020.05.023
  47. Wolff RF, 2019, ANN INTERN MED, V170, P51, DOI 10.7326/M18-1376

Coleções
Artigos e Materiais de Revistas Científicas - FM/MCG
Artigos e Materiais de Revistas Científicas - HC/ICHC
Artigos e Materiais de Revistas Científicas - LIM/28
Artigos e Materiais de Revistas Científicas - ODS/03