Research ExplorerPublicationsMachine Learning for Onset Prediction of Patients with Intra...

Machine Learning for Onset Prediction of Patients with Intracerebral Hemorrhage

Standard

Machine Learning for Onset Prediction of Patients with Intracerebral Hemorrhage. / Rusche, Thilo; Wasserthal, Jakob; Breit, Hanns-Christian; Fischer, Urs; Guzman, Raphael; Fiehler, Jens; Psychogios, Marios-Nikos; Sporns, Peter B.

In: J CLIN MED, Vol. 12, No. 7, 2631, 31.03.2023.

Research output: SCORING: Contribution to journal › SCORING: Journal article › Research › peer-review

Harvard

Rusche, T, Wasserthal, J, Breit, H-C, Fischer, U, Guzman, R, Fiehler, J, Psychogios, M-N & Sporns, PB 2023, 'Machine Learning for Onset Prediction of Patients with Intracerebral Hemorrhage', J CLIN MED, vol. 12, no. 7, 2631. https://doi.org/10.3390/jcm12072631

APA

Rusche, T., Wasserthal, J., Breit, H-C., Fischer, U., Guzman, R., Fiehler, J., Psychogios, M-N., & Sporns, P. B. (2023). Machine Learning for Onset Prediction of Patients with Intracerebral Hemorrhage. J CLIN MED, 12(7), [2631]. https://doi.org/10.3390/jcm12072631

Vancouver

Rusche T, Wasserthal J, Breit H-C, Fischer U, Guzman R, Fiehler J et al. Machine Learning for Onset Prediction of Patients with Intracerebral Hemorrhage. J CLIN MED. 2023 Mar 31;12(7). 2631. https://doi.org/10.3390/jcm12072631

Bibtex

@article{6afcbf2f5371484ab338b7437c97bfa2,
title = "Machine Learning for Onset Prediction of Patients with Intracerebral Hemorrhage",
abstract = "OBJECTIVE: Intracerebral hemorrhage (ICH) has a high mortality and long-term morbidity and thus has a significant overall health-economic impact. Outcomes are especially poor if the exact onset is unknown, but reliable imaging-based methods for onset estimation have not been established. We hypothesized that onset prediction of patients with ICH using artificial intelligence (AI) may be more accurate than human readers.MATERIAL AND METHODS: A total of 7421 computed tomography (CT) datasets between January 2007-July 2021 from the University Hospital Basel with confirmed ICH were extracted and an ICH-segmentation algorithm as well as two classifiers (one with radiomics, one with convolutional neural networks) for onset estimation were trained. The classifiers were trained based on the gold standard of 644 datasets with a known onset of >1 and <48 h. The results of the classifiers were compared to the ratings of two radiologists.RESULTS: Both the AI-based classifiers and the radiologists had poor discrimination of the known onsets, with a mean absolute error (MAE) of 9.77 h (95% CI (confidence interval) = 8.52-11.03) for the convolutional neural network (CNN), 9.96 h (8.68-11.32) for the radiomics model, 13.38 h (11.21-15.74) for rater 1 and 11.21 h (9.61-12.90) for rater 2, respectively. The results of the CNN and radiomics model were both not significantly different to the mean of the known onsets (p = 0.705 and p = 0.423).CONCLUSIONS: In our study, the discriminatory power of AI-based classifiers and human readers for onset estimation of patients with ICH was poor. This indicates that accurate AI-based onset estimation of patients with ICH based only on CT-data may be unlikely to change clinical decision making in the near future. Perhaps multimodal AI-based approaches could improve ICH onset prediction and should be considered in future studies.",
author = "Thilo Rusche and Jakob Wasserthal and Hanns-Christian Breit and Urs Fischer and Raphael Guzman and Jens Fiehler and Marios-Nikos Psychogios and Sporns, {Peter B}",
year = "2023",
month = mar,
day = "31",
doi = "10.3390/jcm12072631",
language = "English",
volume = "12",
journal = "J CLIN MED",
issn = "2077-0383",
publisher = "MDPI AG",
number = "7",

}

RIS

TY - JOUR

T1 - Machine Learning for Onset Prediction of Patients with Intracerebral Hemorrhage

AU - Rusche, Thilo

AU - Wasserthal, Jakob

AU - Breit, Hanns-Christian

AU - Fischer, Urs

AU - Guzman, Raphael

AU - Fiehler, Jens

AU - Psychogios, Marios-Nikos

AU - Sporns, Peter B

PY - 2023/3/31

Y1 - 2023/3/31

N2 - OBJECTIVE: Intracerebral hemorrhage (ICH) has a high mortality and long-term morbidity and thus has a significant overall health-economic impact. Outcomes are especially poor if the exact onset is unknown, but reliable imaging-based methods for onset estimation have not been established. We hypothesized that onset prediction of patients with ICH using artificial intelligence (AI) may be more accurate than human readers.MATERIAL AND METHODS: A total of 7421 computed tomography (CT) datasets between January 2007-July 2021 from the University Hospital Basel with confirmed ICH were extracted and an ICH-segmentation algorithm as well as two classifiers (one with radiomics, one with convolutional neural networks) for onset estimation were trained. The classifiers were trained based on the gold standard of 644 datasets with a known onset of >1 and <48 h. The results of the classifiers were compared to the ratings of two radiologists.RESULTS: Both the AI-based classifiers and the radiologists had poor discrimination of the known onsets, with a mean absolute error (MAE) of 9.77 h (95% CI (confidence interval) = 8.52-11.03) for the convolutional neural network (CNN), 9.96 h (8.68-11.32) for the radiomics model, 13.38 h (11.21-15.74) for rater 1 and 11.21 h (9.61-12.90) for rater 2, respectively. The results of the CNN and radiomics model were both not significantly different to the mean of the known onsets (p = 0.705 and p = 0.423).CONCLUSIONS: In our study, the discriminatory power of AI-based classifiers and human readers for onset estimation of patients with ICH was poor. This indicates that accurate AI-based onset estimation of patients with ICH based only on CT-data may be unlikely to change clinical decision making in the near future. Perhaps multimodal AI-based approaches could improve ICH onset prediction and should be considered in future studies.

AB - OBJECTIVE: Intracerebral hemorrhage (ICH) has a high mortality and long-term morbidity and thus has a significant overall health-economic impact. Outcomes are especially poor if the exact onset is unknown, but reliable imaging-based methods for onset estimation have not been established. We hypothesized that onset prediction of patients with ICH using artificial intelligence (AI) may be more accurate than human readers.MATERIAL AND METHODS: A total of 7421 computed tomography (CT) datasets between January 2007-July 2021 from the University Hospital Basel with confirmed ICH were extracted and an ICH-segmentation algorithm as well as two classifiers (one with radiomics, one with convolutional neural networks) for onset estimation were trained. The classifiers were trained based on the gold standard of 644 datasets with a known onset of >1 and <48 h. The results of the classifiers were compared to the ratings of two radiologists.RESULTS: Both the AI-based classifiers and the radiologists had poor discrimination of the known onsets, with a mean absolute error (MAE) of 9.77 h (95% CI (confidence interval) = 8.52-11.03) for the convolutional neural network (CNN), 9.96 h (8.68-11.32) for the radiomics model, 13.38 h (11.21-15.74) for rater 1 and 11.21 h (9.61-12.90) for rater 2, respectively. The results of the CNN and radiomics model were both not significantly different to the mean of the known onsets (p = 0.705 and p = 0.423).CONCLUSIONS: In our study, the discriminatory power of AI-based classifiers and human readers for onset estimation of patients with ICH was poor. This indicates that accurate AI-based onset estimation of patients with ICH based only on CT-data may be unlikely to change clinical decision making in the near future. Perhaps multimodal AI-based approaches could improve ICH onset prediction and should be considered in future studies.

U2 - 10.3390/jcm12072631

DO - 10.3390/jcm12072631

M3 - SCORING: Journal article

C2 - 37048712

VL - 12

JO - J CLIN MED

JF - J CLIN MED

SN - 2077-0383

IS - 7

M1 - 2631

ER -