Dataset Information

Predicting Infarct Core From Computed Tomography Perfusion in Acute Ischemia With Machine Learning: Lessons From the ISLES Challenge.

ABSTRACT:

Background and purpose

The ISLES challenge (Ischemic Stroke Lesion Segmentation) enables globally diverse teams to compete to develop advanced tools for stroke lesion analysis with machine learning. Detection of irreversibly damaged tissue on computed tomography perfusion (CTP) is often necessary to determine eligibility for late-time-window thrombectomy. Therefore, the aim of ISLES-2018 was to segment infarcted tissue on CTP based on diffusion-weighted imaging as a reference standard.

Methods

The data, from 4 centers, consisted of 103 cases of acute anterior circulation large artery occlusion stroke who underwent diffusion-weighted imaging rapidly after CTP. Diffusion-weighted imaging lesion segmentation was performed manually and acted as a reference standard. The data were separated into 63 cases for training and 40 for testing, upon which quality metrics (dice score coefficient, Hausdorff distance, absolute lesion volume difference, etc) were computed to rank methods based on their overall performance.

Results

Twenty-four different teams participated in the challenge. Median time to CTP was 185 minutes (interquartile range, 180-238), the time between CTP and magnetic resonance imaging was 36 minutes (interquartile range, 25-79), and the median infarct lesion size was 15.2 mL (interquartile range, 5.7-45). The best performance for Dice score coefficient and absolute volume difference were 0.51 and 10.1 mL, respectively, from different teams. Based on the ranking criteria, the top team's algorithm demonstrated for average Dice score coefficient and average absolute volume difference 0.51 and 10.2 mL, respectively, outperforming the conventional threshold-based method (dice score coefficient, 0.3; volume difference, 15.3). Diverse algorithms were used, almost all based on deep learning, with top-ranked approaches making use of the raw perfusion data as well as methods to synthetically generate complementary information to boost prediction performance.

Conclusions

Machine learning methods may predict infarcted tissue from CTP with improved accuracy compared with threshold-based methods used in clinical routine. This dataset will remain public and can be used to test improvement in algorithms over time.

SUBMITTER: Hakim A

PROVIDER: S-EPMC8240494 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:IntroductionComputed tomography perfusion (CTP) imaging has become an important tool in evaluating acute recanalization treatment candidates. Large clinical trials have successfully used RAPID automated imaging analysis software for quantifying ischemic core and penumbra, yet other commercially available software vendors are also on the market. We evaluated the possible difference in ischemic core and perfusion lesion volumes and the agreement rate of target mismatch between OLEA, MIStar, and Syngo.Via versus RAPID software in acute recanalization treatment candidates.Patients and methodsAll consecutive stroke-code patients with baseline CTP RAPID imaging at Helsinki University Hospital during 8/2018-9/2021 were included. Ischemic core was defined as cerebral blood flow <30% than the contralateral hemisphere and within the area of delay time (DT) >3s with MIStar. Perfusion lesion volume was defined as DT > 3 s (MIStar) and Tmax > 6 s with all other software. A perfusion mismatch ratio of ⩾1.8, a perfusion lesion volume of ⩾15 mL, and ischemic core <70 mL was defined as target mismatch. The mean pairwise differences of the core and perfusion lesion volumes between software were calculated using the Bland-Altman method and the agreement of target mismatch between software using the Pearson correlation.ResultsA total of 1606 patients had RAPID perfusion maps, 1222 of which had MIStar, 596 patients had OLEA, and 349 patients had Syngo.Via perfusion maps available. Each software was compared with simultaneously analyzed RAPID software. MIStar showed the smallest core difference compared with RAPID (-2 mL, confidence interval (CI) from -26 to 22), followed by OLEA (2 mL, CI from -33 to 38). Perfusion lesion volume differed least with MIStar (4 mL, CI from -62 to 71) in comparison with RAPID, followed by Syngo.Via (6 mL, CI from -94 to 106). MIStar had the best agreement rate with target mismatch of RAPID followed by OLEA and Syngo.Via.Discussion and conclusionComparison of RAPID with three other automated imaging analysis software showed variance in ischemic core and perfusion lesion volumes and in target mismatch.

Project description:Background and purposeThe Alberta Stroke Program Early Computed Tomography (CT) Score (ASPECTS) and CT perfusion (CTP) are commonly used to predict the ischemic core in acute ischemic strokes. CT angiography source images (CTA-SI) can also provide additional information to identify the extent of ischemia. Our objective was to investigate the correlation of noncontrast CT (NCCT) ASPECTS and CTA-SI ASPECTS with CTP core volumes.MethodsWe utilized a single institutional, retrospective registry of consecutive patients with acute ischemic stroke with large vessel occlusion between May 2016 and May 2018. We graded ASPECTS both on baseline NCCT and CTA-SI and measured CTP core using automated RAPID software (cerebral blood flow <30%). We used Spearman's correlation coefficients to evaluate the correlation between continuous variables.ResultsA total of 52 patients fit the inclusion criteria of large vessel occlusion in 6 to 24 hours and baseline imaging work up of NCCT, CTA, and CTP. The median age was 63 (interquartile range=53.5-75) and 38.46% were female. The median NCCT ASPECTS was 7 (interquartile range=6-9), CTA-SI ASPECTS was 5 (interquartile range=4-7), and CTP core was 14.5 mL (interquartile range=0-46 mL). There was a moderate correlation between NCCT ASPECTS and CTP core (rs=-0.55, P<0.0001) and between CTA-SI ASPECTS and CTP core (rs=-0.50, P=0.0002). The optimal NCCT ASPECTS cutoff score to detect CTP core ≤70 mL was ≥6 (sensitivity, 0.84; specificity, 0.57; positive predictive value, 0.93; negative predictive value, 0.36) and the optimal CTA-SI ASPECTS was ≥5 (sensitivity, 0.76; specificity, 0.71; positive predictive value, 0.94; negative predictive value, 0.31).ConclusionsThere was a moderate correlation between NCCT and CTA-SI ASPECTS in predicting CTP defined ischemic core in delayed time windows. Further studies are needed to determine if NCCT and CTA imaging could be used for image-based patient selection when CTP imaging is not available.

Project description:PurposeComputed tomography (CT)-derived ventilation methods compute respiratory induced volume changes as a surrogate for pulmonary ventilation. Currently, there are no known methods to derive perfusion information from noncontrast CT. We introduce a novel CT-Perfusion (CT-P) method for computing the magnitude mass changes apparent on dynamic noncontrast CT as a surrogate for pulmonary perfusion.MethodsCT-Perfusion is based on a mass conservation model which describes the unknown mass change as a linear combination of spatially corresponding inhale and exhale HU estimated voxel densities. CT-P requires a deformable image registration (DIR) between the inhale/exhale lung CT pair, a preprocessing lung volume segmentation, and an estimate for the Jacobian of the DIR transformation. Given this information, the CT-P image, which provides the magnitude mass change for each voxel within the lung volume, is formulated as the solution to a constrained linear least squares problem defined by a series of subregional mean magnitude mass change measurements. Similar to previous robust CT-ventilation methods, the amount of uncertainty in a subregional sample mean measurement is related to measurement resolution and can be characterized with respect to a tolerance parameter τ . Spatial Spearman correlation between single photon emission CT perfusion (SPECT-P) and the proposed CT-P method was assessed in two patient cohorts via a parameter sweep of τ . The first cohort was comprised of 15 patients diagnosed with pulmonary embolism (PE) who had SPECT-P and 4DCT imaging acquired within 24 h of PE diagnosis. The second cohort was comprised of 15 nonsmall cell lung cancer patients who had SPECT-P and 4DCT images acquired prior to radiotherapy. For each test case, CT-P images were computed for 30 different uncertainty parameter values, uniformly sampled from the range [0.01, 0.125], and the Spearman correlation between the SPECT-P and the resulting CT-P images were computed.ResultsThe median correlations between CT-P and SPECT-P taken over all 30 test cases ranged between 0.49 and 0.57 across the parameter sweep. For the optimal tolerance τ = 0.0385, the CT-P and SPECT-P correlations across all 30 test cases ranged between 0.02 and 0.82. A one-sample sign test was applied separately to the PE and lung cancer cohorts. A low Spearmen correlation of 15% was set as the null median value and two-sided alternative was tested. The PE patients showed a median correlation of 0.57 (IQR = 0.305). One-sample sign test was statistically significant with 96.5 % confidence interval: 0.20-0.63, P < 0.00001. Lung cancer patients had a median correlation of 0.57(IQR = 0.230). Again, a one-sample sign test for median was statistically significant with 96.5 percent confidence interval: 0.45-0.71, P < 0.00001.ConclusionCT-Perfusion is the first mechanistic model designed to quantify magnitude blood mass changes on noncontrast dynamic CT as a surrogate for pulmonary perfusion. While the reported correlations with SPECT-P are promising, further investigation is required to determine the optimal CT acquisition protocol and numerical method implementation for CT-P imaging.