Research ExplorerPublicationsImaging-Based Outcome Prediction of Acute Intracerebral Hemo...

Imaging-Based Outcome Prediction of Acute Intracerebral Hemorrhage

Standard

Imaging-Based Outcome Prediction of Acute Intracerebral Hemorrhage. / Nawabi, Jawed; Kniep, Helge; Elsayed, Sarah; Friedrich, Constanze; Sporns, Peter; Rusche, Thilo; Böhmer, Maik; Morotti, Andrea; Schlunk, Frieder; Dührsen, Lasse; Broocks, Gabriel; Schön, Gerhard; Quandt, Fanny; Thomalla, Götz; Fiehler, Jens; Hanning, Uta.

In: TRANSL STROKE RES, Vol. 12, No. 6, 12.2021, p. 958-967.

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

Harvard

Nawabi, J, Kniep, H, Elsayed, S, Friedrich, C, Sporns, P, Rusche, T, Böhmer, M, Morotti, A, Schlunk, F, Dührsen, L, Broocks, G, Schön, G, Quandt, F, Thomalla, G, Fiehler, J & Hanning, U 2021, 'Imaging-Based Outcome Prediction of Acute Intracerebral Hemorrhage', TRANSL STROKE RES, vol. 12, no. 6, pp. 958-967. https://doi.org/10.1007/s12975-021-00891-8

APA

Nawabi, J., Kniep, H., Elsayed, S., Friedrich, C., Sporns, P., Rusche, T., Böhmer, M., Morotti, A., Schlunk, F., Dührsen, L., Broocks, G., Schön, G., Quandt, F., Thomalla, G., Fiehler, J., & Hanning, U. (2021). Imaging-Based Outcome Prediction of Acute Intracerebral Hemorrhage. TRANSL STROKE RES, 12(6), 958-967. https://doi.org/10.1007/s12975-021-00891-8

Vancouver

Nawabi J, Kniep H, Elsayed S, Friedrich C, Sporns P, Rusche T et al. Imaging-Based Outcome Prediction of Acute Intracerebral Hemorrhage. TRANSL STROKE RES. 2021 Dec;12(6):958-967. https://doi.org/10.1007/s12975-021-00891-8

Bibtex

@article{53639b1697f349fa887b5e80be2ed0ff,
title = "Imaging-Based Outcome Prediction of Acute Intracerebral Hemorrhage",
abstract = "We hypothesized that imaging-only-based machine learning algorithms can analyze non-enhanced CT scans of patients with acute intracerebral hemorrhage (ICH). This retrospective multicenter cohort study analyzed 520 non-enhanced CT scans and clinical data of patients with acute spontaneous ICH. Clinical outcome at hospital discharge was dichotomized into good outcome and poor outcome using different modified Rankin Scale (mRS) cut-off values. Predictive performance of a random forest machine learning approach based on filter- and texture-derived high-end image features was evaluated for differentiation of functional outcome at mRS 2, 3, and 4. Prediction of survival (mRS ≤ 5) was compared to results of the ICH Score. All models were tuned, validated, and tested in a nested 5-fold cross-validation approach. Receiver-operating-characteristic area under the curve (ROC AUC) of the machine learning classifier using image features only was 0.80 (95% CI [0.77; 0.82]) for predicting mRS ≤ 2, 0.80 (95% CI [0.78; 0.81]) for mRS ≤ 3, and 0.79 (95% CI [0.77; 0.80]) for mRS ≤ 4. Trained on survival prediction (mRS ≤ 5), the classifier reached an AUC of 0.80 (95% CI [0.78; 0.82]) which was equivalent to results of the ICH Score. If combined, the integrated model showed a significantly higher AUC of 0.84 (95% CI [0.83; 0.86], P value <0.05). Accordingly, sensitivities were significantly higher at Youden Index maximum cut-offs (77% vs. 74% sensitivity at 76% specificity, P value <0.05). Machine learning-based evaluation of quantitative high-end image features provided the same discriminatory power in predicting functional outcome as multidimensional clinical scoring systems. The integration of conventional scores and image features had synergistic effects with a statistically significant increase in AUC.",
author = "Jawed Nawabi and Helge Kniep and Sarah Elsayed and Constanze Friedrich and Peter Sporns and Thilo Rusche and Maik B{\"o}hmer and Andrea Morotti and Frieder Schlunk and Lasse D{\"u}hrsen and Gabriel Broocks and Gerhard Sch{\"o}n and Fanny Quandt and G{\"o}tz Thomalla and Jens Fiehler and Uta Hanning",
year = "2021",
month = dec,
doi = "10.1007/s12975-021-00891-8",
language = "English",
volume = "12",
pages = "958--967",
journal = "TRANSL STROKE RES",
issn = "1868-4483",
publisher = "SPRINGER US",
number = "6",

}

RIS

TY - JOUR

T1 - Imaging-Based Outcome Prediction of Acute Intracerebral Hemorrhage

AU - Nawabi, Jawed

AU - Kniep, Helge

AU - Elsayed, Sarah

AU - Friedrich, Constanze

AU - Sporns, Peter

AU - Rusche, Thilo

AU - Böhmer, Maik

AU - Morotti, Andrea

AU - Schlunk, Frieder

AU - Dührsen, Lasse

AU - Broocks, Gabriel

AU - Schön, Gerhard

AU - Quandt, Fanny

AU - Thomalla, Götz

AU - Fiehler, Jens

AU - Hanning, Uta

PY - 2021/12

Y1 - 2021/12

N2 - We hypothesized that imaging-only-based machine learning algorithms can analyze non-enhanced CT scans of patients with acute intracerebral hemorrhage (ICH). This retrospective multicenter cohort study analyzed 520 non-enhanced CT scans and clinical data of patients with acute spontaneous ICH. Clinical outcome at hospital discharge was dichotomized into good outcome and poor outcome using different modified Rankin Scale (mRS) cut-off values. Predictive performance of a random forest machine learning approach based on filter- and texture-derived high-end image features was evaluated for differentiation of functional outcome at mRS 2, 3, and 4. Prediction of survival (mRS ≤ 5) was compared to results of the ICH Score. All models were tuned, validated, and tested in a nested 5-fold cross-validation approach. Receiver-operating-characteristic area under the curve (ROC AUC) of the machine learning classifier using image features only was 0.80 (95% CI [0.77; 0.82]) for predicting mRS ≤ 2, 0.80 (95% CI [0.78; 0.81]) for mRS ≤ 3, and 0.79 (95% CI [0.77; 0.80]) for mRS ≤ 4. Trained on survival prediction (mRS ≤ 5), the classifier reached an AUC of 0.80 (95% CI [0.78; 0.82]) which was equivalent to results of the ICH Score. If combined, the integrated model showed a significantly higher AUC of 0.84 (95% CI [0.83; 0.86], P value <0.05). Accordingly, sensitivities were significantly higher at Youden Index maximum cut-offs (77% vs. 74% sensitivity at 76% specificity, P value <0.05). Machine learning-based evaluation of quantitative high-end image features provided the same discriminatory power in predicting functional outcome as multidimensional clinical scoring systems. The integration of conventional scores and image features had synergistic effects with a statistically significant increase in AUC.

AB - We hypothesized that imaging-only-based machine learning algorithms can analyze non-enhanced CT scans of patients with acute intracerebral hemorrhage (ICH). This retrospective multicenter cohort study analyzed 520 non-enhanced CT scans and clinical data of patients with acute spontaneous ICH. Clinical outcome at hospital discharge was dichotomized into good outcome and poor outcome using different modified Rankin Scale (mRS) cut-off values. Predictive performance of a random forest machine learning approach based on filter- and texture-derived high-end image features was evaluated for differentiation of functional outcome at mRS 2, 3, and 4. Prediction of survival (mRS ≤ 5) was compared to results of the ICH Score. All models were tuned, validated, and tested in a nested 5-fold cross-validation approach. Receiver-operating-characteristic area under the curve (ROC AUC) of the machine learning classifier using image features only was 0.80 (95% CI [0.77; 0.82]) for predicting mRS ≤ 2, 0.80 (95% CI [0.78; 0.81]) for mRS ≤ 3, and 0.79 (95% CI [0.77; 0.80]) for mRS ≤ 4. Trained on survival prediction (mRS ≤ 5), the classifier reached an AUC of 0.80 (95% CI [0.78; 0.82]) which was equivalent to results of the ICH Score. If combined, the integrated model showed a significantly higher AUC of 0.84 (95% CI [0.83; 0.86], P value <0.05). Accordingly, sensitivities were significantly higher at Youden Index maximum cut-offs (77% vs. 74% sensitivity at 76% specificity, P value <0.05). Machine learning-based evaluation of quantitative high-end image features provided the same discriminatory power in predicting functional outcome as multidimensional clinical scoring systems. The integration of conventional scores and image features had synergistic effects with a statistically significant increase in AUC.

U2 - 10.1007/s12975-021-00891-8

DO - 10.1007/s12975-021-00891-8

M3 - SCORING: Journal article

C2 - 33547592

VL - 12

SP - 958

EP - 967

JO - TRANSL STROKE RES

JF - TRANSL STROKE RES

SN - 1868-4483

IS - 6

ER -