Accuracy of CT perfusion ischemic core volume and location estimation: A comparison between four ischemic core estimation approaches using syngo.via.

BACKGROUND AND
OBJECTIVE: Computed tomography perfusion (CTP) is widely used in the evaluation of acute ischemic stroke patients for endovascular thrombectomy (EVT). The stability of CTP core estimation is suboptimal and varies between software packages. We aimed to quantify the volumetric and spatial agreement between the CTP ischemic core and follow-up infarct for four ischemic core estimation approaches using syngo.via.
METHODS: We included successfully reperfused, EVT-treated patients with baseline CTP and 24h follow-up diffusion weighted magnetic resonance imaging (DWI) (November 2017-September 2020). Data were processed with syngo.via VB40 using four core estimation approaches based on: cerebral blood volume (CBV)<1.2mL/100mL with and without smoothing filter, relative cerebral blood flow (rCBF)<30%, and rCBF<20%. The follow-up infarct was segmented on DWI.
RESULTS: In 59 patients, median estimated CTP core volumes for four core estimation approaches ranged from 12-39 mL. Median 24h follow-up DWI infarct volume was 11 mL. The intraclass correlation coefficient (ICC) showed moderate-good volumetric agreement for all approaches (range 0.61-0.76). Median Dice was low for all approaches (range 0.16-0.21). CTP core overestimation >10mL occurred least frequent (14/59 [24%] patients) using the CBV-based core estimation approach with smoothing filter.
CONCLUSIONS: In successfully reperfused patients who underwent EVT, syngo.via CTP ischemic core estimation showed moderate volumetric and spatial agreement with the follow-up infarct on DWI. In patients with complete reperfusion after EVT, the volumetric agreement was excellent. A CTP core estimation approach based on CBV<1.2 mL/100mL with smoothing filter least often overestimated the follow-up infarct volume and is therefore preferred for clinical decision making using syngo.via.

Introduction

Computed tomography perfusion (CTP) is widely used in the evaluation of acute ischemic stroke patients for endovascular thrombectomy (EVT) [1-3]. CTP is also used to identify patients who are eligible for IV alteplase treatment when EVT is contraindicated or not planned between 4.5–9 hours after stroke onset [4, 5]. CTP could quantify the cerebral perfusion based on the following parameters: cerebral blood volume (CBV), cerebral blood flow (CBF), mean transit time (MTT), time-to-maximum (Tmax), or time-to-peak (TTP). With these parameters, the extent of the severely hypoperfused ‘ischemic core’ and the hypoperfused, but–if timely reperfused–viable ‘penumbra’ can be estimated. Although the use of CTP ischemic core volume for the selection for EVT of patients in the 6-24h time window is included in the current guidelines, it is not recommended for selection in the 0-6h time window [6, 7]. Yet, CTP is still commonly performed in the 0-6h time window in the Netherlands, albeit usually without clinical consequenses. The various commercially available CTP post-processing software packages use different approaches based on CBV or relative CBF (rCBF), and Tmax or rCBF parameters thresholds to estimate the respective ischemic core and penumbra, which complicates the generalizability of CTP results [8-11]. Siemens syngo.via (Siemens Healthcare, Erlangen, Germany) is a widely used CTP post-processing software package, with a recommended CBV-based core (threshold CBV <1.2 mL/100mL) and CBF-based penumbra (threshold CBF<27 mL/100mL/min) estimation approach. Another frequently used CTP software package is RAPID (iSchemaView, Menlo Park, CA, USA), which uses relative CBF (rCBF) <30% and Tmax >6 seconds as the thresholds for ischemic core and penumbra, respectively. Previous studies have focused on differences between different CTP post-processing software packages and the comparison of the CTP results between vendors based on modifying specific thresholds [8, 10–19]. However, the agreement of core estimation for different CTP ischemic core estimation approaches using syngo.via has sparsely been studied [8, 12, 13]. Moreover, it is unclear how different core estimation approaches affect the accuracy of CTP in estimating the 24h diffusion-weighted imaging (DWI) follow-up infarct lesion. This is clinically important, since the follow-up infarct volume is, inter alia, a strong predictor of functional outcome [14]. We aimed to determine the extent of differences among four commonly used CTP core estimation approaches by quantifying the volumetric and spatial agreement between the CTP ischemic core and the 24h follow-up DWI infarct using syngo.via.

Materials and methods

Study population

We performed a single center, single acquisition protocol retrospective analysis of a prospectively collected registry of EVT-treated patients with baseline CTP. We included patients who were presented to our center for EVT between November 2017–September 2020. Other inclusion criteria were: admission within 24 hours after symptom onset, present occlusion of the anterior circulation, successful reperfusion (defined as expanded treatment in cerebral ischemia (eTICI) score 2b-3), and available 24h follow-up diffusion weighted imaging (DWI). Patients were not included for analysis if the CTP data could not be made available due to acquisition and storage at the primary stroke center according to the European General Data Protection Regulation (GDPR).

Image acquisition

CTP images were acquired on a dual source 192-slice scanner (70 kVp, 12 cm coverage; SOMATOM Force, Siemens Healthcare, Forchheim, Germany) with a slice thickness of 1mm and 0.7mm increment. All acquisitions were reconstructed to 5mm slices. All CTP scans were acquired after intravenous injection of 35 mL iodinated non-ionic contrast agent (Iomeron 300, iomeprol, 300mg iodine/mL; Bracco Imaging Deutschland GmbH, Konstanz, Germany) with an injection rate of 6 mL/s. CTP data were acquired using the following protocol: 15 scans 1.5 seconds apart, followed by 15 scans 3 seconds apart, resulting in a total of 30 scans over a period of 60 seconds. Baseline non-contrast CT and CT angiography images were acquired on the same dual source CT scanner, with the CTP acquired after the non-contrast and before the CTA. Depending on scanner availability, follow-up MRI DWI (b = 0 s/mm2 and b = 1000 s/mm2) and apparent diffusion coefficient (ADC) images were acquired on a 1.5 T scanner (n = 43; MAGNETOM Avanto fit, Siemens Healthcare, Erlangen, Germany) or a 3.0 T scanner (n = 14; Ingenia 3.0T, Philips Healthcare, Best, the Netherlands), both with a slice thickness of 5 mm.

Posttreatment imaging assessment

Posttreatment recanalization rate was scored as the extended Thrombolysis in Cerebral Infarction (eTICI) score by an independent core lab from the Collaboration for New Treatments of Acute Stroke (CONTRAST) consortium (n = 40). For patients not included in one of the CONTRAST trials or for whom there was no core lab observation available (n = 19), posttreatment DSAs were evaluated by an independent, blinded observer (>5 years of experience) who is part of the CONTRAST consortium core lab.

CTP data post-processing

CTP data were processed using Siemens syngo.via CT Neuro Perfusion (version VB40; Siemens Healthcare, Erlangen, Germany). The software was used in a research environment in order to automatically produce the results and to be able to modify the parameters and thresholds. CTP data were checked for severe patient motion, arterial input function curve, and presence of metal artifacts. Four core estimation approaches were investigated: Approach 1 represents the CBV-based core estimation approach (threshold CBV<1.2mL/100mL) with additional smoothing filter [8]. Approach 2 represents the conventional, CBV-based core estimation approach (threshold CBV<1.2mL/100mL) without smoothing filter, approach 3 represents a rCBF-based approach with thresholds derived from another commercial package (RAPID; iSchemaView) (threshold rCBF<30% + smoothing filter), and approach 4 represents the rCBF-based core estimation approach from syngo.via used for research purposes (threshold rCBF<20% + smoothing filter). We chose these specific four approaches since these are commonly used in daily clinical practice or recommended in the current guidelines [6, 7].

Data co-registration and follow-up imaging assessment

The follow-up DWI images (median 23h) were registered (rigid registration) to the baseline CTP using Elastix [15]. Results were visually inspected to assure correct alignment (JWH). The follow-up infarct volumes were segmented on DWI using a semi-automated segmentation method [16] with subsequent visual assessment by an expert neuroradiologist with >15 years of experience (CBLMM) who was blinded to all clinical information but occlusion side. If necessary, the automated segmentation results were manually adjusted after visual expert assessment using ITK-SNAP [17]. DWI images were screened for scattered lesions. We defined a scattered lesion as at least two separate hyperintense DWI lesions within the territory of one of the major cerebral arteries [18]. DWI scattered lesion patterns were scored as absent (no hyperintense lesions), mild (>0 but <5 hyperintense lesions), moderate (≥5 but <10 separate hyperintense lesions), or severe (≥10 separate hyperintense lesions). All hyperintense lesions on DWI were included for both volumetric and spatial analysis. DWI was chosen for follow-up infarct segmentation as it shows good agreement with the follow-up infarct volume and it is more sensitive than FLAIR for the detection of acute ischemic stroke lesions [19, 20]. The ADC maps were consulted to prevent including T2 shine-through lesions in the DWI lesion. For spatial agreement analysis, the CTP-estimated ischemic core segmentation was co-registered to the baseline CTP.

Assessment of volumetric and spatial agreement and other statistical analyses

We calculated the volume difference between the estimated CTP ischemic core and DWI follow-up infarct volume as

Negative values indicate overestimation of the ischemic core by the CTP software. The intraclass correlation coefficient (ICC) estimates and their 95% confidence intervals were calculated to assess the agreement between the estimated CTP ischemic core volume for each approach and the follow-up DWI infarct volume. We based the ICC estimates on a mean-rating (k = 2), absolute agreement, 2-way mixed-effects model and classified the degree of agreement as previously suggested (ICC <0.5 = poor agreement, ICC 0.5–0.75 = moderate agreement, ICC 0.75–0.9 = good agreement, and ICC >0.9 = excellent agreement) [21]. Bland-Altman analyses were performed to compare the estimated CTP ischemic core and follow-up DWI infarct volumes. Proportional bias was assessed using linear regression. The spatial overlap between the CTP and DWI segmentation was calculated using FSLMaths [22]. We calculated the Dice similarity coefficient using the ‘fslr’ package in R to quantify the spatial agreement between the CTP-estimated ischemic core and follow-up DWI lesion for the four core estimation approaches [23].

We reported the frequency and summary statistics for ordinal and continuous baseline characteristics. We performed linear regression analyses to determine the association between the time from imaging to reperfusion and spatial and volume difference. To determine if there were statistically significant differences in volumetric or spatial agreement between the estimated CTP ischemic core and the follow-up DWI lesion for the four approaches, we performed Friedman tests. Finally, we performed a sensitivity analysis of the volumetric and spatial agreement between CTP ischemic core and 24h follow-up DWI infarct lesion in patients with incomplete (eTICI 2b) vs. complete reperfusion (eTICI 3) to determine the degree of differences between subgroups in our analysis. All statistical analyses were performed using R (R Core Team (2020). R: A language and environment for statistical computing, R Foundation for Statistical Computing, Vienna, Austria. https://www.R-project.org).

Ethics statement

This study was reviewed by the Medical Ethical Committee Board of the Amsterdam University Medical Centers (location AMC) and informed consent was waived (Reference W19_281#19.334) as retrospective, observational studies do not fall under the scope Medical Research Involving Human Subject Act (WMO).

Results

Three hundred one patients were presented to our comprehensive stroke center for EVT between November 2017 and October 2020 and received baseline CTP imaging. For 284/301 patients, baseline CTA imaging showed an anterior circulation large vessel occlusion. Eighty-four (30%) patients received follow-up DWI at median 23h (IQR 18–34) after CTP. Most patients (49/59; 83%) in our study cohort were included in one of the randomized controlled trials of the CONTRAST consortium (i.e., MR CLEAN-NO IV, MR CLEAN-MED or MR CLEAN-LATE) [24] and received 24h follow-up DWI as part of the pre-specified follow-up imaging of the concerning trial [25-27]. After application of the exclusion criteria, we included a total of 59 patients in our analysis (Fig 1).

Fig 1

Flowchart of patient selection.

Table 1 shows a detailed description of the baseline characteristics. Median age was 71 (IQR 58–76) years. Median National Institutes of Health Stroke Scale Score (NIHSS) was 15 (IQR 9–18) and most patients presented within 6 hours after symptom onset (55/59; 93%). Compared to the overall Dutch stroke population described in the MR CLEAN Registry [28], more patients were female in our cohort (59% vs. 47%) and less patients received IV alteplase (49% vs. 78%).

Table 1

Baseline characteristics.

	Study cohort (n = 59)
Clinical characteristics
Age (yr)–median (IQR)	71 (58–77)
Female–n (%)	36 (59)
NIHSS score–median (IQR) [known in]	15 (9–19) [n = 57]
IVT administered–n (%)	29 (49)
Onset-to-imaging time (min)–median (IQR) [known in]	83 (58–188) [n = 57]
Imaging-to-reperfusion time (min)–median (IQR) [known in]	83 (63–114) [n = 58]
Onset-to-groin time (min)–median (IQR) [known in]	140 (105–222) [n = 53]
Imaging characteristics
Occlusion location on baseline CTA–n (%)
Intracranial ICA	3 (5)
ICA-T	11 (19)
M1	42 (70)
M2	3 (5)
Collateral status–n (%) [known in]	[n = 58]
0	2 (3)
1	18 (31)
2	23 (39)
3	15 (25)
Median baseline ischemic core volume on CTP (mL); approach 1 –median (IQR)	15 (4–31)
Median baseline ischemic core volume on CTP (mL); approach 2 –median (IQR)	27 (11–53)
Median baseline ischemic core volume on CTP (mL); approach 3 –median (IQR)	39 (20–95)
Median baseline ischemic core volume on CTP (mL); approach 4 –median (IQR)	11 (3–31)
Posttreatment recanalization rate (eTICI)
0	1 (2)
2a	3 (5)
2b	20 (34)
2c	8 (14)
3	27 (46)
Follow-up DWI infarct volume (mL)–median (IQR)	11 (5–42)
Median time between baseline CTP and follow-up DWI (hrs)–median (IQR)	23 (18–34)

ASPECTS = Alberta Stroke Program Early CT Score; CTP = CT perfusion; ICA = intracranial carotid artery; ICA-T = intracranial carotid artery terminus; IVT = IV alteplase; IQR = interquartile range; NIHSS = National Institute of Health Stroke Scale. If the [known in] number is not shown, the variable was known in all patients.

All other clinical and imaging characteristics were comparable to the population from the MR CLEAN Registry [28].

Volumetric agreement analysis

The median volume difference between the follow-up DWI infarct and the CTP ischemic core per estimation approach was: approach 1: 0 (IQR -10 to 18) mL, approach 2: -8 (IQR -23 to 13) mL, approach 3: -16 (IQR -44 to 2) mL, and approach 4: 2 (IQR -6 to 20) mL. Most patients showed a moderate (14/59; 24%) or severe (28/59; 34%) scattered lesion pattern on the follow-up DWI. The CTP ischemic core volumes were significantly different (p<0.01) for all combinations of core estimation approaches, except for the combination approach 1 vs. approach 4 (p = 0.4). Of note, for individual cases, we still found volume differences up to 40 mL between approach 1 and approach 4 (Fig 2). ICC estimates showed moderate–good volumetric agreement for all core estimation approaches (approach 1: ICC 0.61[95% CI 0.39–0.75], approach 2: ICC 0.61[95% CI 0.40–0.75], approach 3: ICC 0.76[95% CI 0.64–0.85], approach 4: ICC 0.67[95% CI 0.48–0.78]).

Fig 2

Bland-Altman plots comparing the estimated CTP ischemic core volume and DWI follow-up infarct volume for (a) approach 1, (b) approach 2, (c) approach 3, and (d) approach 4. The mean bias (blue), lower (red) and upper (green) Limits of Agreement are shown with 95% confidence intervals. The bias with 95% confidence intervals is shown in blue. Negative values indicate overestimation by CTP. CTP = CT perfusion; DWI = diffusion weighted imaging.

The Bland-Altman plots are shown in Fig 2. Linear regression showed that proportional bias was present for 3 of the 4 core estimation approaches (approach 1: p<0.01, approach 2: p<0.01, and approach 4: p<0.01).

CTP ischemic core overestimation of >10 mL was not uncommon (approach 1:14/59 (24%), approach 2: 26/59 (48%), approach 3: 33/59 (56%), approach 4: 12/59 (20%). Severe volume overestimation >50 mL by CTP occurred significantly more often for approach 3 (24%, 14/59) compared to approach 1 (3% 2/59; p<0.01), approach 2 (7%, 4/59; p = 0.01), and approach 4 (7%, 4/59; p = 0.01). Please see S1 Fig for a schematic overview of the CTP ischemic core volumes per core estimation approach. Please see S2 Fig for the scatter plots of the volumetric agreement between the estimated CTP core and the 24h follow-up DWI infarct.

Fig 3 shows examples of CTP ischemic core estimates (red) for the four estimation approaches and follow-up DWI and ADC images.

Fig 3

Baseline CTP of a patient with a right-sided M1 occlusion with successful reperfusion (eTICI 3).

The ischemic core (red) and penumbra (green) for (a) approach 1, (b) approach 2, (c) approach 3, and (d) approach 4. (e) Follow-up DWI acquired at 17 hours after baseline imaging and (f) follow-up DWI image with follow-up infarct segmentation (red). A = anterior; CTP = CT perfusion; DWI = diffusion weighted imaging; MRI = magnetic resonance imaging; R = right.

Spatial agreement analysis

The median Dice was low for all core estimation approaches (approach 1: 0.16[IQR 0.02–0.31], approach 2: 0.15[IQR 0.02–0.30], approach 3: 0.21[IQR 0.06–0.35], approach 4: 0.15[IQR 0.01–0.32]). See S3 Fig for more details.

Effect of imaging-to-reperfusion time on the spatial and volumetric accuracy

The median time between CTP acquisition and reperfusion was 83 (IQR 63–114) minutes. Longer time between imaging and reperfusion was not associated with spatial accuracy, but was associated with increased volume difference for all approaches (range 0.9–1.0 mL per minute). Scatter plots are shown in S4 and S5 Figs.

Sensitivity analysis comparing patients with incomplete vs. complete reperfusion (eTICI 2b vs. eTICI 3)

For patients with incomplete (eTICI 2b; n = 20) vs. complete reperfusion (eTICI 3; n = 27), median CTP ischemic core volumes were as follows: approach 1: 12 (IQR 4–26) mL vs. 16 (IQR 11–34) mL, approach 2: 18 (IQR 11–41) mL vs. 31 (IQR 21–55) mL, approach 3: 34 (IQR 16–76) mL vs. 42 (IQR 24–112) mL, and approach 4: 7 (IQR 2–24) mL vs. 12 (IQR 5–33) mL.

The ICC estimates for both patients with eTICI 2b and eTICI 3 ranged from moderate-good (Table 2).

Table 2

Volumetric agreement for patients with eTICI 2b vs. eTICI 3.

	Approach 1	Approach 2	Approach 3	Approach 4
ICC eTICI 2b (n = 20)	0.62 (95% CI 0.18–0.83)	0.62 (95% CI 0.18–0.83)	0.78 (95% CI 0.53–0.90)	0.69 (95% CI 0.33–0.86)
ICC eTICI 3 (n = 27)	0.65 (95% CI 0.33–0.82)	0.68 (95% CI 0.38–0.83)	0.80 (95% CI 0.62–0.90)	0.69 (95% CI 0.41–0.84)

The median Dice indicated low spatial agreement for all estimation approaches for both patients with eTICI 2b and eTICI3 (Table 3).

Table 3

Spatial agreement for patients with eTICI 2b vs. eTICI 3.

	Approach 1	Approach 2	Approach 3	Approach 4
Dice TICI 2b (n = 20)–median (IQR)	0.16 (IQR 0.01–0.33)	0.13 (IQR 0.01–0.32)	0.16 (IQR 0.06–0.35)	0.05 (IQR 0.00–0.32)
Dice TICI 3 (n = 27)–median (IQR)	0.19 (IQR 0.07–0.29)	0.22 (IQR 0.19–0.28)	0.26 (IQR 0.18–0.36)	0.18 (IQR 0.07–0.33)

Median times between CTP and FU DWI for patients who achieved non-complete (eTICI 2b) reperfusion and complete (eTICI 3) reperfusion were 21 (IQR 19–33) and 23 (IQR 17–31) hours, respectively (p = 0.8). Median follow-up DWI volumes were 12 (IQR 4–22) mL and 16 (IQR 11–31) mL for the eTICI 2b and eTICI 3 subgroup, respectively (p = 0.3).”

Discussion

Our study showed good–excellent volumetric agreement between the CTP ischemic core and the follow-up DWI lesion for four core estimation approaches for EVT-treated patients with complete reperfusion. The core estimation approach based on CBV<1.2 mL/100 mL with smoothing filter showed the best volumetric agreement. Overall, the spatial agreement was low (Dice range: 0.16–0.21). We found the highest spatial accuracy for the core estimation approach based on rCBF <30% with smoothing filter. If we also included patients with successful–yet not complete–reperfusion, we found moderate spatial and volumetric agreement. Volumetric overestimation >10mL was not uncommon and occurred in 20–56% of the successfully reperfused patients depending on the core estimation approach used. For patients with volumetric overestimation of approximately 10mL it is not likely that this would have affected the treatment decision for these patients. For 14/59 (24%) patients, the rCBF-based core estimation approach resulted in severe volumetric overestimation >50 mL whereas severe volumetric overestimation occurred in 2/59 (3%) patients using the core estimation approach based on CBV <1.2 mL/100mL with smoothing filter. In contrast to volumetric overestimation of approximately 10mL, it is possible that this could have resulted in falsely withholding these patients from EVT from. Our results suggest that severe volumetric overestimation was associated with the occlusion location if the ischemic core approach was based on rCBF with a threshold of rCBF <30% or rCBF <20% (approach 3 and approach 4).

A recent study compared baseline estimated ischemic core volumes of syngo.via to estimated ischemic core volumes from RAPID [12]. They found similar results between syngo.via (version VB30) and RAPID if rCBF <20% was used for ischemic core estimation by syngo.via. However, since the current recommended core estimation approach of syngo.via is not based on rCBF, these results are not generalizable to a clinical setting where a recommended CBV-based approach is used. Also, this study did not compare the CTP results to follow-up DWI, in contrast to the present study. Another study compared the core estimates from four software packages (i.e., RAPID, VEOcore, syngo.via, and Olea) and found volume differences up to 33 mL between the different software packages [11].

Other studies on CTP accuracy found median volume differences between the CTP ischemic core volume and follow-up imaging of 30 mL and 13 mL [29, 30]. These studies, however, focused on different CTP software packages (i.e., IntelliSpace Portal and RAPID) with different ischemic core parameters and thresholds (relative mean transit time (rMTT) ≥145% + CBV <2.0 mL/100mL and rCBF <30%), which hampers the comparison of these results to our findings.

The spatial agreement between the estimated CTP ischemic core and follow-up DWI infarct seems poor. However, these results are in line with a previous comparison between RAPID CTP estimated ischemic core and 24h follow-up DWI infarct volume [31]. Comparing Dice scores with the aim to determine the best clinical performance should be performed with caution as the Dice score is very easily negatively affected by ‘false negative voxels’–especially in relatively small segmentations, such as infarct segmentations [32]. This would result in a higher Dice for a core estimation approach with more frequent CTP ischemic core overestimation (e.g., for approach 3 in our analysis). Yet, this approach would not be optimal for clinical decision making as overestimation could lead to falsely withholding patients from EVT.

The moderate volumetric agreement (ICC) in our study could be a result of infarct growth in case of incomplete (micro- or macrovascular) reperfusion or delay between imaging and reperfusion. The significant association between longer imaging-to-reperfusion time and larger volume difference for all core estimation approaches supports this hypothesis. Moreover, if only patients with complete reperfusion were taken into account, we found good–excellent volumetric accuracy for all core estimation approaches.

The volumetric agreement (ICC) outperformed the spatial agreement for all ischemic core estimation approaches. It is plausible that scattered subcortical white matter lesions–which are likely to be a result of distal embolization during the EVT procedure–contribute to this finding. These lesions do not have a major impact on the volumetric agreement, but do affect the degree of ‘false negative voxels’ and thus the spatial agreement between the CTP ischemic core and follow-up infarct as mentioned earlier.

Several limitations to our study should be noted. First, all included patients who underwent EVT had relatively small ischemic core volumes (median range 13–40 mL). More specifically, 43/59 (73%) patients received baseline imaging in the hyperacute time window (i.e., within 3 hours after symptom onset) where CTP is not recommended for selection for EVT. The median onset-to-imaging and onset-to-groin times were 83 and 140 minutes, respectively. Therefore, we are not able to draw conclusions about the volumetric or spatial accuracy of syngo.via for large core volumes >70 mL or for patients who presented outside hyperacute time window, for example due to transfer from a mothership hospital. Of note, for patients who present within this time window, CTP is not recommended for selection for EVT. Second, we only included patients who received follow-up DWI which could have introduced selection bias. In the Netherlands, follow-up DWI is generally only acquired for research purposes and therefore sparsely performed in clinical practice. However, from a local cohort of 291 patients who received EVT outside one of the Collaboration for New Treatments of Acute Stroke (CONTRAST) trials, the median CTP-estimated core volume was 11 (IQR 5–32) mL, so it is unlikely that only patients with smaller core volumes were selected in our current study population. Third, we compared the estimated CTP ischemic core volume at baseline with the 24h follow-up DWI infarct volume in patients with successful reperfusion as previously suggested [33]. Although DWI accurately differentiates between cytotoxic and vasogenic edema [34] and is commonly used to determine the follow-up infarct volume, there is no ideal reference standard to determine the infarct volume [33]. Fourth, the infarct is likely to expand in the time between CTP acquisition and reperfusion–especially in patients with incomplete micro- or macrovascular reperfusion (i.e., eTICI 2b or 2c)–, which make the degree of reperfusion and timing of the follow-up imaging important factors to consider when performing accuracy assessments [35]. Also, co-registration between CTP and DWI is not optimal and could have negatively influenced our findings. To minimize this effect, we visually inspected all registration results. Previous studies have assessed this by determining the ventricle overlap between CTP-DWI and found suboptimal spatial agreement (Dice 0.8) [25, 27]. Fifth, we included patients with both 1.5 and 3.0 T follow-up DWI scans. As it has been shown that there might be great volumetric and spatial differences between the two field strengths, this could have influenced our results [36]. Finally, we were not able to assess the accuracy of the penumbra estimations since patients who did not undergo EVT were not included in our cohort.

Conclusions

In successfully reperfused patients who underwent EVT, syngo.via CTP ischemic core estimation showed moderate volumetric and spatial agreement with the follow-up infarct on DWI. For patients with complete reperfusion after EVT, the volumetric agreement was excellent. A CTP core estimation approach based on CBV<1.2 mL/100mL with smoothing filter least often overestimated the follow-up infarct volume and is therefore preferred for clinical decision making using syngo.via.

Boxplots show the distribution of CTP ischemic core volume per estimation approach.

The estimated CTP ischemic core volume was statistically significantly different for the four core estimation approaches using Friedman test, χ2 = 102.31, p<0.001. Pairwise Wilcoxon signed rank test between groups revealed statistically significant differences in CTP ischemic core volume between approach 1-approach 2 (p<0.001), approach 1-approach 3 (p<0.001), approach 2-approach 3 (p<0.001), approach 2-approach 4 (p<0.001), and approach 3-approach 4 (p<0.001). CTP = computed tomography perfusion.

(TIF)

Click here for additional data file.

Scatter plots show the agreement between the estimated CTP ischemic core volume and the follow-up DWI infarct lesion for (A) approach 1, (B) approach 2, (C) approach 3, and (D) approach 4. The solid grey line represents the identity line. Points below the identity line (grey) indicate a larger CTP ischemic core volume compared to the follow-up DWI lesion, i.e., overestimation by CTP. Points above the identity indicate underestimation by CTP or infarct growth. CTA = CT angiography; CTP = CT perfusion; DWI = diffusion weighted imaging; ICA = intracranial carotid artery; ICA-T = intracranial carotid artery terminus; M1 = M1 (horizontal) segment of the middle cerebral artery; M2 = M2 (insular) segment of the middle cerebral artery.

(TIF)

Click here for additional data file.

Boxplots show the distribution of the Dice similarity coefficient (Dice) per estimation approach.

Dice was statistically significantly different for the different approaches using Friedman test, χ2 = 27.45, p<0.001. Pairwise Wilcoxon signed rank test between groups revealed statistically significant differences in Dice between approach 1-approach 3 (p = 0.005), approach 2-approach 3 (p = 0.001), and approach 3-approach 4 (p = 0.001). CTP = computed tomography perfusion.

(TIF)

Click here for additional data file.

Scatter plots show the association between time from CTP imaging to reperfusion and Dice similarity coefficient for (a) approach 1, (b) approach 2, (c) approach 3, and (d) approach 4. R = Pearson correlation coefficient.

(TIF)

Click here for additional data file.

Scatter plots show the association between time from CTP imaging to reperfusion and volume difference for (a) approach 1, (b) approach 2, (c) approach 3, and (d) approach 4. R = Pearson correlation coefficient.

(TIF)

Click here for additional data file.

23 Feb 2022

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Marios-Nikos Psychogios

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Thank you for stating the following in the Competing Interests section:

"HAM is co-founder and shareholder of Nicolab. All other authors report no conflicts of interest. BJE reports grants from Stryker Neurovascular and personal fees from Dekra and from Novartis, outside the submitted work. CBLMM reports grants from TWIN, during the conduct of the study and grants from CVON/Dutch Heart Foundation and from Stryker outside the submitted work (paid to institution) and is shareholder of Nicolab. All other authors did not receive support from any organization for the submitted work, had no financial relationships with any organizations that might have an interest in the submitted work in the previous three years, and had no other relationships or activities that could appear to have influenced the submitted work."

We note that you received funding from a commercial source: Stryker Neurovascular, Dekra, Novartis, TWIN, CVON/Dutch Heart Foundation and Stryker

Please provide an amended Competing Interests Statement that explicitly states this commercial funder, along with any other relevant declarations relating to employment, consultancy, patents, products in development, marketed products, etc.

Within this Competing Interests Statement, please confirm that this does not alter your adherence to all PLOS ONE policies on sharing data and materials by including the following statement: "This does not alter our adherence to PLOS ONE policies on sharing data and materials.” (as detailed online in our guide for authors http://journals.plos.org/plosone/s/competing-interests).  If there are restrictions on sharing of data and/or materials, please state these. Please note that we cannot proceed with consideration of your article until this information has been declared.

Please include your amended Competing Interests Statement within your cover letter. We will change the online submission form on your behalf.

3. We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For more information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions.

In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially sensitive information, data are owned by a third-party organization, etc.) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories.

We will update your Data Availability statement on your behalf to reflect the information you provide.

4. Please include your full ethics statement in the ‘Methods’ section of your manuscript file. In your statement, please include the full name of the IRB or ethics committee who approved or waived your study, as well as whether or not you obtained informed written or verbal consent. If consent was waived for your study, please include this information in your statement as well.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: No

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This is a well written study, which addresses an important clinical question with merging importance especially, as LVOs are more and more prone to treatment in expanded time windows and different approaches for estimation of penumbra and core assessment are used in the field.

This already leads to one of my major concerns regarding this work:

- the authors chose very early treatment time points (Mr Clean onset -door: 240min, here 140min; Hermes Onset to random 196min vs 83 onset to image here). Firstly these are timelines which in my eyes do not apply to most clinical settings, as the o-d times are much longer, especially for shipped patients. Secondly the perfusion approach is not applicable to the timelines studied here, as current guideline recommendation is beyond the 4,5 rep. 6hrs window. Thus it would have made more sense to focus on later time points.

-Another point is the delay of the MRI , which dependent on TICI success will influence the final DWI lesion size. can the authors comment on that and provide an analysis of 2b and final DWI size compared to 3?

- another drawback is the use of different MRI field strengths within the cohort (3 vs 1,5T), there is some evidence form stereotactic studies, that there might be great differences between the to field strengths (doi: 10.3340/jkns.2014.55.3.136). Can the authors please comment on that?

- Another problem could be the non blinded TICI evaluation , which was done by the interventionalist and might lead to BIAS, It would be worthwhile to crosscheck those by a blinded rater.

- Table 1 shows, that for some metrics not all informations form every patient were available....how did the authors deals with that?

Minor: Table 1:

median follow up DWI time 23min? probably 23hrs...?

Line 307 duplication of "from"

Suppl. figure S1 in my eyes does not add visual information, as I can visually barely detect sign. differences between the approaches.

Reviewer #2: Overall interesting topic which needs further research as these thresholds are still to be validated and fu infarct volume is a good approximation.

One reference you might add to the discussion as it looks at a similar topic outlining the discrepancies between different software packages:

https://pubmed.ncbi.nlm.nih.gov/33216157/

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

24 Mar 2022

Reviewer #1:

This is a well written study, which addresses an important clinical question with merging importance especially, as LVOs are more and more prone to treatment in expanded time windows and different approaches for estimation of penumbra and core assessment are used in the field.

1.1 This already leads to one of my major concerns regarding this work:

- the authors chose very early treatment time points (Mr Clean onset -door: 240min, here 140min; Hermes Onset to random 196min vs 83 onset to image here). Firstly, these are timelines which in my eyes do not apply to most clinical settings, as the o-d times are much longer, especially for shipped patients.

We thank the reviewer for pointing us to this limitation. We agree that the patients in our study cohort received imaging in the hyperacute time window and relatively shortly after symptom onset (median 83 minutes). This could partially be explained by the fact that most patients included in our analysis were directly admitted to the EVT-capable center and did not have to be transferred from a primary stroke center. Another reason could be that most patients who were included in the MR CLEAN-NO IV trial received follow-up DWI imaging and were subsequently included in our analysis. The MR CLEAN-NO IV trial studied the added value of IV alteplase prior to EVT in the 0-4.5h time window.

We have added this as a limitation to our study. (Page 20, line 557-564):

“Several limitations to our study should be noted. First, all included patients who underwent EVT had relatively small ischemic core volumes (median range 13-40 mL). More specifically, 43/59 (73%) patients received baseline imaging in the hyperacute time window (i.e., within 3 hours after symptom onset) where CTP is not recommended for selection for EVT. The median onset-to-imaging and onset-to-groin times were 83 and 140 minutes, respectively. Therefore, we are not able to draw conclusions about the volumetric or spatial accuracy of syngo.via for large core volumes >70 mL or for patients who presented outside hyperacute time window, for example due to transfer from a mothership hospital.”

1.2 Secondly, the perfusion approach is not applicable to the timelines studied here, as current guideline recommendation is beyond the 4,5 rep. 6hrs window. Thus it would have made more sense to focus on later time points.

We thank the reviewer for this comment and we acknowledge that CTP is indeed not recommended for selection for IVT or EVT in the 0-4.5, resp. 0-6h time windows. We agree that CTP should currently not be used for patient selection for EVT in the earlier time windows. However, we feel that for diagnostic purposes, it is relevant to quantify the spatial and volumetric accuracy, both for the hyperacute as in the later time windows. We agree that it would have been interesting to include more patients in the later time window. However, we were only able to include patients with follow-up DWI. These follow-up DWI data were mostly acquired in the setting of either the MR CLEAN-NO IV or the MR CLEAN-MED trials which were focusing on patients in the 0-4.5 or 0-6h time window.

We have specified the origin of our study cohort in the Results section and specified that patients received 24h follow-up imaging as part of follow-up imaging from the concerning randomized trial. (Page 11, line 275-279):

“Most patients (49/59; 83%) in our study cohort were included in one of the randomized controlled trials of the CONTRAST consortium (i.e., MR CLEAN-NO IV, MR CLEAN-MED or MR CLEAN-LATE) [24] and received 24h follow-up DWI as part of the pre-specified follow-up imaging of the concerning trial [25-27].”

We have addressed the Discussion section (Page 20, line 558-565):

“More specifically, 43/59 (73%) patients received baseline imaging in the hyperacute time window (i.e., within 3 hours after symptom onset) where CTP is not recommended for selection for EVT. The median onset-to-imaging and onset-to-groin times were 83 and 140 minutes, respectively. Therefore, we are not able to draw conclusions about the volumetric or spatial accuracy of syngo.via for large core volumes >70 mL or for patients who presented outside hyperacute time window, for example due to transfer from a mothership hospital. Of note, for patients who present within this time window, CTP is not recommended for selection for EVT.”

1.3 Another point is the delay of the MRI, which dependent on TICI success will influence the final DWI lesion size. can the authors comment on that and provide an analysis of 2b and final DWI size compared to 3?

We fully agree with the reviewer and acknowledge that this is a major limitation of our analysis.

We have emphasized this in the limitations section of our Discussion (Page 21, lines 589-593):

“Fourth, the infarct is likely to expand in the time between CTP acquisition and reperfusion – especially in patients with incomplete micro- or macrovascular reperfusion (i.e., eTICI 2b or 2c) –, which make the degree of reperfusion and timing of the follow-up imaging important factors to consider when performing accuracy assessments [35].”

Also, we have added an additional sensitivity analysis (“Sensitivity analysis comparing patients with incomplete vs. complete reperfusion (eTICI 2b vs. eTICI 3”) between the (newly scored) eTICI 2b vs. eTICI 3 subgroups to the Results section. (Page 16-17, lines 399-481):

We have performed additional analyses on the volumetric (ICC) and spatial (Dice) agreement for these two subgroups and found that for patients with complete reperfusion, all approaches showed moderate-good volumetric agreement and low spatial agreement between the estimated CTP core the 24h follow-up DWI infarct. (Page 16-17, Tables 2-3)

1.4 Another drawback is the use of different MRI field strengths within the cohort (3 vs 1,5T), there is some evidence form stereotactic studies, that there might be great differences between the two field strengths (doi:10.3340/jkns.2014.55.3.136). Can the authors please comment on that?

We thank the reviewer for pointing us to this important issue. We have added this as a limitation to our study in the Discussion section of our manuscript. (Page 21, lines 594-597):

“Fifth, we included patients with both 1.5 and 3.0 T follow-up DWI scans. As it has been shown that there might be great volumetric and spatial differences between the two field strengths, this could have influenced our results [36].”

1.5 Another problem could be the non-blinded TICI evaluation, which was done by the interventionalist and might lead to BIAS. It would be worthwhile to crosscheck those by a blinded rater.

We agree that the TICI evaluation by the interventionalist could have introduced bias. Unfortunately, there were no independent core lab observations available at the time of the writing of the manuscript. However, in the meantime, the data of the MR CLEAN-NO IV trials has become available. As most patients included in our analysis were also included in the MR CLEAN-NO IV trial, we could resort to the independent core lab TICI evaluations. These observations were all scored blindly by an independent observer who was part of the Collaborations for New Treatment of Acute Stroke (CONTRAST) consortium core lab. For patients who were not included in the MR CLEAN-NO IV trial or for whom there was no core lab observation available (n=19), the post-EVT DSA images were crosschecked by an independent, blinded rater who is part of the CONTRAST core lab.

We have adapted the specification of the posttreatment imaging assessment this in our manuscript. (Page 7, lines 151-156):

“Posttreatment recanalization rate was scored as the extended Thrombolysis in Cerebral Infarction (eTICI) score by an independent core lab from the Collaboration for New Treatments of Acute Stroke (CONTRAST) consortium (n=40). For patients not included in one of the CONTRAST trials or for whom there was no core lab observation available (n=19), posttreatment DSAs were evaluated by an independent, blinded observer (>5 years of experience) who is part of the CONTRAST consortium core lab.”

We have adapted S2 Fig with the updated eTICI scores. (S2 Fig)

1.6 Table 1 shows, that for some metrics not all informations form every patient were available....how did the authors deals with that?

We thank the reviewer for noticing. Indeed, unfortunately, not all clinical baseline variables were available for all patients at the time of writing of the manuscript. However, in the meantime, clinical data from the MR CLEAN-NO IV trial has become available. Since a considerable number of patients included in our analysis were also included in the MR CLEAN-NO IV trial, we were able to add missing clinical baseline variables for patients who were also enrolled in the MR CLEAN-NO IV trial. Although we regret that not all clinical variables are available for all patients included in our analysis, we feel that the data represented in Table 1 still provides an accurate overview of the included population as most patients had baseline data available.

We have updated Table 1 with the updated values. (Table 1)

1.7 Table 1: median follow up DWI time 23min? probably 23hrs...?

Indeed, this should be ‘hours’. We thank the reviewer for noticing and have adapted this accordingly. (Page 11-13, Table 1)

1.8 Line 307 duplication of "from"

Thank you. Corrected.

1.9 Suppl. figure S1 in my eyes does not add visual information, as I can visually barely detect sign. differences between the approaches.

We agree with the reviewer that it is hard to clearly see differences in S1 Fig and now see that this figure has limited added value. Hence, we have decided to remove S1 Fig from the Supporting Information and removed the references to S1 Fig from the manuscript. S2-S6 Figs have been renamed to S1-S5 Figs, accordingly.

Reviewer #2:

Overall interesting topic which needs further research as these thresholds are still to be validated and fu infarct volume is a good approximation.

2.1 One reference you might add to the discussion as it looks at a similar topic outlining the discrepancies between different software packages: https://pubmed.ncbi.nlm.nih.gov/33216157/

We thank the reviewer for this suggestion and have added the reference to both the Introduction and Discussion section.

(Page 4, line 92-95):

“The various commercially available CTP post-processing software packages use different approaches based on CBV or relative CBF (rCBF), and Tmax or rCBF parameters thresholds to estimate the respective ischemic core and penumbra, which complicates the generalizability of CTP results [8-11].”

(Page 19, line 517-519):

“Another study compared the core estimates from four software packages (i.e., RAPID, VEOcore, syngo.via, and Olea) and found volume differences up to 33 mL between the different software packages [11].”

Submitted filename: Response_letter.docx

Click here for additional data file.

18 Jul 2022

Accuracy of CT perfusion ischemic core volume and location estimation: a comparison between four ischemic core estimation approaches using syngo.via

PONE-D-22-00274R1

Dear Dr. Hoving,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Miquel Vall-llosera Camps

Senior Editor

PLOS ONE

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #2: The Authors have sufficiently answered to all my comments and improved the manuscript very well.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #2: No

**********

21 Jul 2022

PONE-D-22-00274R1

Accuracy of CT perfusion ischemic core volume and location estimation: a comparison between four ischemic core estimation approaches using syngo.via

Dear Dr. Hoving:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Miquel Vall-llosera Camps

Staff Editor

PLOS ONE