Skip to main content
Advertisement
  • Loading metrics

PLoS Computational Biology Issue Image | Vol. 19(8) September 2023

Topological data analysis predicts recurrence in paediatric acute lymphoblastic leukaemia

Acute lymphoblastic leukaemia (ALL) is a blood cancer which mainly affects children and adolescents. Therapy fails for approximately 20% of patients who suffer relapse. We use methods from topological data analysis, which quantifies shapes in data, to analyse pre-treatment ALL datasets with known outcomes. We combine these analyses with machine learning to identify significant shape characteristics in the data (notably, isolated data islands and empty spaces between them) and show that they predict risk of relapse with high accuracy. We also confirm the predictive power of CD10, CD20, CD38 and CD45 as biomarkers for ALL diagnosis. Chulián et al 2023

Image Credit: Salvador Chulián, salvador.chulian@uca.es

thumbnail
Topological data analysis predicts recurrence in paediatric acute lymphoblastic leukaemia

Acute lymphoblastic leukaemia (ALL) is a blood cancer which mainly affects children and adolescents. Therapy fails for approximately 20% of patients who suffer relapse. We use methods from topological data analysis, which quantifies shapes in data, to analyse pre-treatment ALL datasets with known outcomes. We combine these analyses with machine learning to identify significant shape characteristics in the data (notably, isolated data islands and empty spaces between them) and show that they predict risk of relapse with high accuracy. We also confirm the predictive power of CD10, CD20, CD38 and CD45 as biomarkers for ALL diagnosis. Chulián et al 2023

Image Credit: Salvador Chulián, salvador.chulian@uca.es

https://doi.org/10.1371/image.pcbi.v19.i08.g001