A machine-learning framework for robust and reliable prediction of short- and long-term treatment response in initially antipsychotic-naïve schizophrenia patients based on multimodal neuropsychiatric data

Publikation: Bidrag til tidsskriftTidsskriftartikelfagfællebedømt

Standard

A machine-learning framework for robust and reliable prediction of short- and long-term treatment response in initially antipsychotic-naïve schizophrenia patients based on multimodal neuropsychiatric data. / Ambrosen, Karen S.; Skjerbæk, Martin W.; Foldager, Jonathan; Axelsen, Martin C.; Bak, Nikolaj; Arvastson, Lars; Christensen, Søren R.; Johansen, Louise B.; Raghava, Jayachandra M.; Oranje, Bob; Rostrup, Egill; Nielsen, Mette; Osler, Merete; Fagerlund, Birgitte; Pantelis, Christos; Kinon, Bruce J.; Glenthøj, Birte Y.; Hansen, Lars K.; Ebdrup, Bjørn H.

I: Translational Psychiatry, Bind 10, Nr. 1, 276, 2020.

Publikation: Bidrag til tidsskriftTidsskriftartikelfagfællebedømt

Harvard

Ambrosen, KS, Skjerbæk, MW, Foldager, J, Axelsen, MC, Bak, N, Arvastson, L, Christensen, SR, Johansen, LB, Raghava, JM, Oranje, B, Rostrup, E, Nielsen, M, Osler, M, Fagerlund, B, Pantelis, C, Kinon, BJ, Glenthøj, BY, Hansen, LK & Ebdrup, BH 2020, 'A machine-learning framework for robust and reliable prediction of short- and long-term treatment response in initially antipsychotic-naïve schizophrenia patients based on multimodal neuropsychiatric data', Translational Psychiatry, bind 10, nr. 1, 276. https://doi.org/10.1038/s41398-020-00962-8

APA

Ambrosen, K. S., Skjerbæk, M. W., Foldager, J., Axelsen, M. C., Bak, N., Arvastson, L., Christensen, S. R., Johansen, L. B., Raghava, J. M., Oranje, B., Rostrup, E., Nielsen, M., Osler, M., Fagerlund, B., Pantelis, C., Kinon, B. J., Glenthøj, B. Y., Hansen, L. K., & Ebdrup, B. H. (2020). A machine-learning framework for robust and reliable prediction of short- and long-term treatment response in initially antipsychotic-naïve schizophrenia patients based on multimodal neuropsychiatric data. Translational Psychiatry, 10(1), [276]. https://doi.org/10.1038/s41398-020-00962-8

Vancouver

Ambrosen KS, Skjerbæk MW, Foldager J, Axelsen MC, Bak N, Arvastson L o.a. A machine-learning framework for robust and reliable prediction of short- and long-term treatment response in initially antipsychotic-naïve schizophrenia patients based on multimodal neuropsychiatric data. Translational Psychiatry. 2020;10(1). 276. https://doi.org/10.1038/s41398-020-00962-8

Author

Ambrosen, Karen S. ; Skjerbæk, Martin W. ; Foldager, Jonathan ; Axelsen, Martin C. ; Bak, Nikolaj ; Arvastson, Lars ; Christensen, Søren R. ; Johansen, Louise B. ; Raghava, Jayachandra M. ; Oranje, Bob ; Rostrup, Egill ; Nielsen, Mette ; Osler, Merete ; Fagerlund, Birgitte ; Pantelis, Christos ; Kinon, Bruce J. ; Glenthøj, Birte Y. ; Hansen, Lars K. ; Ebdrup, Bjørn H. / A machine-learning framework for robust and reliable prediction of short- and long-term treatment response in initially antipsychotic-naïve schizophrenia patients based on multimodal neuropsychiatric data. I: Translational Psychiatry. 2020 ; Bind 10, Nr. 1.

Bibtex

@article{db570fb1e4fd4e768c0bd960d7c82b10,
title = "A machine-learning framework for robust and reliable prediction of short- and long-term treatment response in initially antipsychotic-na{\"i}ve schizophrenia patients based on multimodal neuropsychiatric data",
abstract = "The reproducibility of machine-learning analyses in computational psychiatry is a growing concern. In a multimodal neuropsychiatric dataset of antipsychotic-na{\"i}ve, first-episode schizophrenia patients, we discuss a workflow aimed at reducing bias and overfitting by invoking simulated data in the design process and analysis in two independent machine-learning approaches, one based on a single algorithm and the other incorporating an ensemble of algorithms. We aimed to (1) classify patients from controls to establish the framework, (2) predict short- and long-term treatment response, and (3) validate the methodological framework. We included 138 antipsychotic-na{\"i}ve, first-episode schizophrenia patients with data on psychopathology, cognition, electrophysiology, and structural magnetic resonance imaging (MRI). Perinatal data and long-term outcome measures were obtained from Danish registers. Short-term treatment response was defined as change in Positive And Negative Syndrome Score (PANSS) after the initial antipsychotic treatment period. Baseline diagnostic classification algorithms also included data from 151 matched controls. Both approaches significantly classified patients from healthy controls with a balanced accuracy of 63.8% and 64.2%, respectively. Post-hoc analyses showed that the classification primarily was driven by the cognitive data. Neither approach predicted short- nor long-term treatment response. Validation of the framework showed that choice of algorithm and parameter settings in the real data was successfully guided by results from the simulated data. In conclusion, this novel approach holds promise as an important step to minimize bias and obtain reliable results with modest sample sizes when independent replication samples are not available.",
author = "Ambrosen, {Karen S.} and Skjerb{\ae}k, {Martin W.} and Jonathan Foldager and Axelsen, {Martin C.} and Nikolaj Bak and Lars Arvastson and Christensen, {S{\o}ren R.} and Johansen, {Louise B.} and Raghava, {Jayachandra M.} and Bob Oranje and Egill Rostrup and Mette Nielsen and Merete Osler and Birgitte Fagerlund and Christos Pantelis and Kinon, {Bruce J.} and Glenth{\o}j, {Birte Y.} and Hansen, {Lars K.} and Ebdrup, {Bj{\o}rn H.}",
year = "2020",
doi = "10.1038/s41398-020-00962-8",
language = "English",
volume = "10",
journal = "Translational Psychiatry",
issn = "2158-3188",
publisher = "nature publishing group",
number = "1",

}

RIS

TY - JOUR

T1 - A machine-learning framework for robust and reliable prediction of short- and long-term treatment response in initially antipsychotic-naïve schizophrenia patients based on multimodal neuropsychiatric data

AU - Ambrosen, Karen S.

AU - Skjerbæk, Martin W.

AU - Foldager, Jonathan

AU - Axelsen, Martin C.

AU - Bak, Nikolaj

AU - Arvastson, Lars

AU - Christensen, Søren R.

AU - Johansen, Louise B.

AU - Raghava, Jayachandra M.

AU - Oranje, Bob

AU - Rostrup, Egill

AU - Nielsen, Mette

AU - Osler, Merete

AU - Fagerlund, Birgitte

AU - Pantelis, Christos

AU - Kinon, Bruce J.

AU - Glenthøj, Birte Y.

AU - Hansen, Lars K.

AU - Ebdrup, Bjørn H.

PY - 2020

Y1 - 2020

N2 - The reproducibility of machine-learning analyses in computational psychiatry is a growing concern. In a multimodal neuropsychiatric dataset of antipsychotic-naïve, first-episode schizophrenia patients, we discuss a workflow aimed at reducing bias and overfitting by invoking simulated data in the design process and analysis in two independent machine-learning approaches, one based on a single algorithm and the other incorporating an ensemble of algorithms. We aimed to (1) classify patients from controls to establish the framework, (2) predict short- and long-term treatment response, and (3) validate the methodological framework. We included 138 antipsychotic-naïve, first-episode schizophrenia patients with data on psychopathology, cognition, electrophysiology, and structural magnetic resonance imaging (MRI). Perinatal data and long-term outcome measures were obtained from Danish registers. Short-term treatment response was defined as change in Positive And Negative Syndrome Score (PANSS) after the initial antipsychotic treatment period. Baseline diagnostic classification algorithms also included data from 151 matched controls. Both approaches significantly classified patients from healthy controls with a balanced accuracy of 63.8% and 64.2%, respectively. Post-hoc analyses showed that the classification primarily was driven by the cognitive data. Neither approach predicted short- nor long-term treatment response. Validation of the framework showed that choice of algorithm and parameter settings in the real data was successfully guided by results from the simulated data. In conclusion, this novel approach holds promise as an important step to minimize bias and obtain reliable results with modest sample sizes when independent replication samples are not available.

AB - The reproducibility of machine-learning analyses in computational psychiatry is a growing concern. In a multimodal neuropsychiatric dataset of antipsychotic-naïve, first-episode schizophrenia patients, we discuss a workflow aimed at reducing bias and overfitting by invoking simulated data in the design process and analysis in two independent machine-learning approaches, one based on a single algorithm and the other incorporating an ensemble of algorithms. We aimed to (1) classify patients from controls to establish the framework, (2) predict short- and long-term treatment response, and (3) validate the methodological framework. We included 138 antipsychotic-naïve, first-episode schizophrenia patients with data on psychopathology, cognition, electrophysiology, and structural magnetic resonance imaging (MRI). Perinatal data and long-term outcome measures were obtained from Danish registers. Short-term treatment response was defined as change in Positive And Negative Syndrome Score (PANSS) after the initial antipsychotic treatment period. Baseline diagnostic classification algorithms also included data from 151 matched controls. Both approaches significantly classified patients from healthy controls with a balanced accuracy of 63.8% and 64.2%, respectively. Post-hoc analyses showed that the classification primarily was driven by the cognitive data. Neither approach predicted short- nor long-term treatment response. Validation of the framework showed that choice of algorithm and parameter settings in the real data was successfully guided by results from the simulated data. In conclusion, this novel approach holds promise as an important step to minimize bias and obtain reliable results with modest sample sizes when independent replication samples are not available.

U2 - 10.1038/s41398-020-00962-8

DO - 10.1038/s41398-020-00962-8

M3 - Journal article

C2 - 32778656

AN - SCOPUS:85089271341

VL - 10

JO - Translational Psychiatry

JF - Translational Psychiatry

SN - 2158-3188

IS - 1

M1 - 276

ER -

ID: 248143711