Machine learning-based pipeline for automated intracerebral hemorrhage and drain detection, quantification, and classification in non-enhanced CT images (NeuroDrAIn)

Abstract

Background and purpose: External drainage represents a well-established treatment option for acute intracerebral hemorrhage. The current standard of practice includes post-operative computer tomography imaging, which is subjectively evaluated. The implementation of an objective, automated evaluation of postoperative studies may enhance diagnostic accuracy and facilitate the scaling of research projects. The objective is to develop and validate a fully automated pipeline for intracerebral hemorrhage and drain detection, quantification of intracerebral hemorrhage coverage, and detection of malpositioned drains.

Materials and methods: In this retrospective study, we selected patients (n = 68) suffering from supratentorial intracerebral hemorrhage treated by minimally invasive surgery, from years 2010-2018. These were divided into training (n = 21), validation (n = 3) and testing (n = 44) datasets. Mean age (SD) was 70 (±13.56) years, 32 female. Intracerebral hemorrhage and drains were automatically segmented using a previously published artificial intelligence-based approach. From this, we calculated coverage profiles of the correctly detected drains to quantify the drains' coverage by the intracerebral hemorrhage and classify malpositioning. We used accuracy measures to assess detection and classification results and intraclass correlation coefficient to assess the quantification of the drain coverage by the intracerebral hemorrhage.

Results: In the test dataset, the pipeline showed a drain detection accuracy of 0.97 (95% CI: 0.92 to 0.99), an agreement between predicted and ground truth coverage profiles of 0.86 (95% CI: 0.85 to 0.87) and a drain position classification accuracy of 0.88 (95% CI: 0.77 to 0.95) resulting in area under the receiver operating characteristic curve of 0.92 (95% CI: 0.85 to 0.99).

Conclusion: We developed and statistically validated an automated pipeline for evaluating computed tomography scans after minimally invasive surgery for intracerebral hemorrhage. The algorithm reliably detects drains, quantifies drain coverage by the hemorrhage, and uses machine learning to detect malpositioned drains. This pipeline has the potential to impact the daily clinical workload, as well as to facilitate the scaling of data collection for future research into intracerebral hemorrhage and other diseases.

Fig 1. Chart delineating the inputs and outputs of each step of the pipeline.

1: Detection of true positive drain objects in binary mask, 2: quantification of the coverage profile and 3: classification of the drain position.

Fig 2

Concordance (A) and Bland-Altman plots (B) of the coverage profiles of predicted and GT drains in the testing dataset. Regression line (blue) and 95% confidence interval of predicted values (shaded area).

Fig 3

Non-contrast axial CT scans (A, C) of a patient from testing dataset following minimally invasive surgery showing a left frontal hemorrhage (red) and drain (green). Corresponding Plots of the coverage profiles (B, D). Vertical dashed line separates the distal 15 mm (green) from the rest of the drain (blue). (A) There is only marginal contact between the drain and the bleeding and (B) a coverage reaching approximately 25% between 15 and 25 mm from the tip. (C) Following surgical correction there was an optimal positioning of the drain and (D) an almost complete coverage of the drain with the hemorrhage.

Fig 4. Figure depicting predicted (blue) and ground truth (green) coverage profiles of drains that were misclassified by the classification model.