Forskning
Udskriv Udskriv
Switch language
Rigshospitalet - en del af Københavns Universitetshospital
Udgivet

Machine learning can identify newly diagnosed patients with CLL at high risk of infection

Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

DOI

  1. An atlas of O-linked glycosylation on peptide hormones reveals diverse biological roles

    Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

  2. A phenome-wide association and Mendelian Randomisation study of polygenic risk for depression in UK Biobank

    Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

  3. Author Correction: Retrospective evaluation of whole exome and genome mutation calls in 746 cancer samples

    Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

  4. SUCLA2 mutations cause global protein succinylation contributing to the pathomechanism of a hereditary mitochondrial disease

    Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

  5. PCM1 is necessary for focal ciliary integrity and is a candidate for severe schizophrenia

    Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

  1. Effectiveness and safety of IFN-free DAA HCV therapy in HIV/HCV co-infected persons: Results from a pan-European study

    Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

  2. Pulmonary Arterial Enlargement in Well-Treated Persons With Human Immunodeficiency Virus

    Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

  3. Changes in primary and secondary hemostasis in patients with CLL treated with venetoclax and ibrutinib

    Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

Vis graf over relationer

Infections have become the major cause of morbidity and mortality among patients with chronic lymphocytic leukemia (CLL) due to immune dysfunction and cytotoxic CLL treatment. Yet, predictive models for infection are missing. In this work, we develop the CLL Treatment-Infection Model (CLL-TIM) that identifies patients at risk of infection or CLL treatment within 2 years of diagnosis as validated on both internal and external cohorts. CLL-TIM is an ensemble algorithm composed of 28 machine learning algorithms based on data from 4,149 patients with CLL. The model is capable of dealing with heterogeneous data, including the high rates of missing data to be expected in the real-world setting, with a precision of 72% and a recall of 75%. To address concerns regarding the use of complex machine learning algorithms in the clinic, for each patient with CLL, CLL-TIM provides explainable predictions through uncertainty estimates and personalized risk factors.

OriginalsprogEngelsk
TidsskriftNature Communications
Vol/bind11
Udgave nummer1
Sider (fra-til)363
ISSN2041-1723
DOI
StatusUdgivet - 17 jan. 2020

ID: 59339866