**Novel Estimation of Penumbra Zone Based on Infarct Growth Using Machine Learning Techniques in Acute Ischemic Stroke**

**Yoon-Chul Kim 1, Hyung Jun Kim 2, Jong-Won Chung 2, In Gyeong Kim 2, Min Jung Seong 3, Keon Ha Kim 3, Pyoung Jeon 3, Hyo Suk Nam 4, Woo-Keun Seo 2, Gyeong-Moon Kim <sup>2</sup> and Oh Young Bang 2,\***


Received: 19 May 2020; Accepted: 22 June 2020; Published: 24 June 2020

**Abstract:** While the penumbra zone is traditionally assessed based on perfusion–diffusion mismatch, it can be assessed based on machine learning (ML) prediction of infarct growth. The purpose of this work was to develop and validate an ML method for the prediction of infarct growth distribution and volume, in cases of successful (SR) and unsuccessful recanalization (UR). Pre-treatment perfusion-weighted, diffusion-weighted imaging (DWI) data, and final infarct lesions annotated from day-7 DWI from patients with middle cerebral artery occlusion were utilized to develop and validate two ML models for prediction of tissue fate. SR and UR models were developed from data in patients with modified treatment in cerebral infarction (mTICI) scores of 2b–3 and 0–2a, respectively. When compared to manual infarct annotation, ML-based infarct volume predictions resulted in an intraclass correlation coefficient (ICC) of 0.73 (95% CI = 0.31–0.91, *p* < 0.01) for UR, and an ICC of 0.87 (95% CI = 0.73–0.94, *p* < 0.001) for SR. Favorable outcomes for mismatch presence and absence in SR were 50% and 36%, respectively, while they were 61%, 56%, and 25%, respectively, for the low, intermediate, and high infarct growth groups. The presented method can offer novel and alternative insights into selecting patients for recanalization therapy and predicting functional outcome.

**Keywords:** stroke; ischemia; machine learning; cerebral infarction

#### **1. Introduction**

Measurement of ischemic core and penumbra volumes in acute ischemic stroke provides clinicians with important clues for predicting clinical response after successful recanalization (SR) and selecting patients for treatments, such as intravenous thrombolysis and endovascular therapy (EVT) [1–3]. Diffusion-weighted imaging (DWI) and perfusion-weighted imaging (PWI) in magnetic resonance imaging (MRI) are neuroimaging modalities that can visualize the location and extent of ischemic stroke lesions with high sensitivity and specificity. The target mismatch criteria help determine patient selection for EVT and are evaluated using DWI and PWI lesion volumes [2].

Besides measurement of the PWI–DWI mismatch zone, estimation of infarct growth would provide a direct approach to the prediction of penumbra areas and the final infarct lesion volume [4].

There have been conflicting results on whether mismatch volume predicts infarct growth volume. It was reported that the infarct growth volume was greater than the mismatch volume, and half of the patients without target mismatch suffered from infarct growth regardless of the revascularization [5,6], while a recent randomized trial of EVT showed a good correlation between the predicted infarct volume and the 27-h infarct volume in target mismatch patients [7]. A clinical trial with a serial MRI study reported that ischemic lesion volume increased, by various degrees, from baseline to 12 weeks in untreated patients [8].

Machine learning (ML) has emerged as a promising methodology in acute stroke neuroimaging to predict ischemic stroke growth by interpolation of shape representations [9], and to predict the voxel-based tissue outcome [10,11] and the clinical outcome [12]. In particular, an ML method, referred to as fully automated stroke tissue estimation using random forest classifiers (FASTER) [13], showed potential to accurately predict final lesion volumes in cases of SR and unsuccessful recanalization (UR). However, the study did not investigate the performance of ML-based infarct growth estimation in comparison with target mismatch based on PWI–DWI mismatch volume and did not evaluate the prediction of clinical outcomes according to ML-based infarct growth estimation.

In the present study, we investigated the feasibility of an ML-based tissue outcome prediction technique using features derived from DWI and PWI data. Based on the ML method, we evaluated the performance of predicting final infarct volumes and infarct growth in the SR and UR cases. We also investigated how well the selection based on the ML-predicted infarct growth volume corresponded with the target mismatch classification in actual infarct growth and functional outcome distributions.

#### **2. Materials and Methods**

#### *2.1. Patients*

This study included patients who were admitted to a university medical center between 5 June 2005 and 31 December 2016. The study inclusion criteria were as follows: (1) symptomatic middle cerebral artery (MCA) occlusion, (2) MRI scan including DWI and PWI sequences prior to treatment and including DWI at 7 days after symptom onset, and (3) baseline MRI within 6 h of symptom onset.

Patients were divided into two groups: UR and SR. The UR group consisted of patients with modified treatment in cerebral infarction (mTICI) scores of 0–2a. The SR group consisted of patients with mTICI scores of 2b–3. For patients treated with EVT, the mTICI score was determined based on post procedural digital subtraction angiography (DSA), and for patients treated with intravenous (IV) tissue plasminogen activator (tPA), it was determined based on 24 h MRI and MR angiography (MRA).

The UR group was divided into the UR development and UR external validation groups, and the SR group was divided into the SR development and SR external validation groups. All patients in the SR group underwent EVT. The development/validation division criterion was based on the chronology of acute ischemic stroke occurrence: The UR and SR development groups corresponded to stroke onset dates prior to 31 December 2010 and 31 December 2011, respectively.

Demographic information including age and sex, baseline National Institutes of Health Stroke Scale (NIHSS) scores, and the 90-day modified Rankin Scale (mRS) scores were collected. The baseline NIHSS score is used to quantify the impairment due to a stroke and ranges from no symptoms at all (0) to the most severe stroke (42). The 90-day mRS score is used to assess clinical outcome and ranges from no symptoms at all (0) to death (6). Radiologic information, such as time interval from symptom onset to MRI, was obtained from electronic medical records.

All patients or patient guardians provided informed consent for inclusion before they participated in the study. The study was conducted in accordance with the Declaration of Helsinki, and the protocol was approved by the institutional review board of the Samsung Medical Center (IRB no. 2016-08-064).

#### *2.2. Data Acquisition*

MRI data were acquired on a 3T scanner system (Philips Achieva, Best, the Netherlands). DWI sequence parameters were as follows: repetition time = 3 s, echo time = 81–88 ms, number of slices = 22, image matrix size = 256 × 256, pixel spacing = 0.9375 mm × 0.9375 mm, and spacing between slices = 6.5 mm. The apparent diffusion coefficient (ADC) was calculated in a pixel-by-pixel manner from DWI acquired with b = 0 and b = 1000 s/mm2. The PWI sequence was based on the commonly used T2\*-weighted dynamic susceptibility contrast with the injection of a bolus of extracellular gadolinium contrast agent, and the parameters were as follows: repetition time = 1.5–1.7 s, echo time = 35 ms, number of slices = 22, spacing between slices = 6.5 mm, echo train length = 67, number of frames = 50, image matrix = 256 × 256, pixel spacing = 0.9375 mm × 0.9375 mm, and spacing between slices = 6.5 mm.

#### *2.3. ADC*/*rTTP and Data Preprocessing*

Typically, ADC and time-to-maximum (Tmax) are used to determine diffusion and perfusion lesion volumes, respectively. A recent study suggested that relative time to peak (rTTP) > 4.5 s was empirically identical to Tmax (>6 s) in determining perfusion lesion volume and resulted in > 90% accuracy, when compared with the conventional Tmax (>6 s) [14]. Notably, in the present study we used the statistics of rTTP as features for ML training, as opposed to a previous study, which used the statistics of Tmax, cerebral blood flow (CBF), cerebral blood volume (CBV), and time to peak (TTP) values as features for ML training [13]. Tmax, CBF, and CBV calculations involve numerical deconvolution with arterial input function (AIF), which is known to be sensitive to image artifacts and the choice of region of interest in major cerebral arteries [15]. The present study only considered the calculation of rTTP (i.e., the TTP delay relative to the contralateral region) for the extraction of PWI-related imaging features, and it did not consider estimating the AIF and performing numerical deconvolution. We sought to evaluate the correspondence between the rTTP-derived PWI lesion volume and the Tmax-derived PWI lesion volume, and the rTTP > 4.5 s volume showed Pearson correlation coefficient of 0.84 with the Tmax > 6 s volume in the subjects considered in the study (Figure A1).

Baseline DWI/ADC, PWI, and day-7 DWI images were co-registered using Statistical Parametric Mapping (SPM) 12 [16]. The rTTP maps from baseline PWI data were automatically calculated as follows. At each voxel, the MR signal was converted to contrast agent concentration, and TTP was measured from the concentration curve. Median TTP values were computed from the right and left hemispheric regions, respectively. The hemisphere with the lower median TTP value was defined as the contralateral region, and the median TTP value in this region served as the baseline TTP. The rTTP maps were calculated by voxel-wise subtraction of the baseline TTP from TTP values.

The midsagittal plane in the axial brain slice was automatically estimated after determining the optimal values of the translation and rotation parameters. The parameter values were used to identify the midline of an image. The midline helped to identify the correct location of a symmetric contralateral voxel, given a stroke lesion voxel, when feature extraction was performed for ML model development.

#### *2.4. ML Model Development*

A schematic of the presented ML method is illustrated in Figure 1B, in comparison with the traditional approach based on the PWI/DWI target mismatch (Figure 1A).

Feature extraction was conducted as follows. Seven consecutive axial slices covering the brain tissue in the MCA territory were considered for feature extraction. All the voxels in the tissue of the lesion hemisphere were considered as candidates, where the infarct voxels were labeled as "1", and the non-infarct voxels were labeled as "0". For training, to avoid class imbalance issues, we set the number of infarct voxels to be equal to the number of non-infarct voxels. For each candidate voxel, we considered its neighborhood as 5 × 5 × 3 voxels surrounding the candidate voxel. From the ADC values of the neighborhood, a set of 12 features was computed, consisting of range, mean, median, min, max, standard deviation (SD), skew, kurtosis, 10th percentile, 25th percentile, 75th percentile, and 90th percentile. Another set of 12 features was computed from the rTTP values in the neighborhood. These 24 features were also computed from a neighborhood of the voxel in the contralateral hemisphere. Hence, the number of features was 24 × 2 = 48. For the UR model, the dimensions of (# of samples) × (# of features) in the training dataset were 297,192 × 48, which provided the input for the UR model. For the SR model, the dimensions in the training dataset were 193,510 × 48, which provided the input for the SR model. Binary infarct masks, manually delineated while referring to the day-7 DWI images, were obtained from all the datasets using the ITK-SNAP software [17]. The output class labels for supervised ML were obtained from the binary infarct masks.

**Figure 1.** Schematics of two approaches for the estimation of final infarct volume and functional outcome. (**A**) Traditional approach of penumbra zone estimation. (**B**) Machine learning (ML)-based infarct prediction.

With the training dataset, a five-fold cross-validation was performed to evaluate the performance of the UR and SR models. The splitting in the cross-validation folds was constructed on the basis of patients rather than voxels. The assignment of patients to each fold was performed randomly. The ML model of the random forest classifier, provided by the Python scikit-learn module, was used for cross-validation [18]. After hyperparameter selection of the model using a random search, the mean and SD of cross-validation accuracy was computed for each model.

#### *2.5. External Validation*

For external validation, we used baseline DWI and PWI images from the 12 UR and the 27 SR patients. Candidate voxels were chosen, and for each candidate voxel, the feature extraction procedures were performed in the same manner as those for the model development. For each voxel, the extracted features were input to the trained UR and SR models to result in a tissue outcome probability score *p*, where *p* ≥ 0.5 indicated an infarct voxel, and *p* < 0.5 indicated a non-infarct voxel. After processing all the candidate voxels, we obtained binary infarct prediction maps for both the UR and SR cases. The final infarct volume was calculated as the number of predicted infarct voxels multiplied by the voxel volume. Two final infarct volumes were calculated: one for the UR case and the other for the SR case.

For the UR (or SR) patient group, the final infarct volumes from the UR (or SR) model were compared with the final infarct volumes from the manually annotated day-7 DWI infarct masks. The Dice similarity coefficient (i.e., Dice score) was computed for each patient to evaluate the overlap between the ML predicted infarct mask and the manually segmented day-7 DWI infarct mask.

#### *2.6. Mismatch and Infarct Growth Estimation*

Target mismatch classification requires information on baseline diffusion and perfusion lesion volumes. We computed the Tmax > 6 s volume (i.e., perfusion lesion volume), the Tmax > 10 s volume (i.e., severe ischemic lesion volume), and the ischemic core volume based on the ADC <sup>&</sup>lt; <sup>600</sup> <sup>×</sup> <sup>10</sup>−<sup>6</sup> mm2/<sup>s</sup> threshold. Target mismatch criteria were defined as (1) a perfusion lesion volume to ischemic core volume ratio of 1.8 or more, (2) a difference between perfusion lesion volume and ischemic core volume of 15 mL or more, (3) the ischemic core volume less than 70 mL, and (4) the Tmax > 10 s volume less than 100 mL.

Actual infarct growth volume was calculated as the difference between the day-7 final infarct volume and the baseline diffusion lesion volume (i.e., ischemic core volume). The day-7 final infarct volume was measured after manual segmentation of the infarct lesions. Predicted infarct growth volume was calculated as the difference between the SR (or UR) predicted infarct volume and the baseline diffusion lesion volume. The SR (or UR) patients were categorized into tertiles according to the predicted infarct growth volume.

The percentage of favorable functional outcome was evaluated for the patient groups with and without target mismatch and for the patient groups categorized into tertiles based on SR (or UR) predicted infarct growth. Favorable functional outcome was defined as a 90-day mRS score of ≤ 1.

#### *2.7. Statistical Analysis*

We conducted statistical analysis using the R software package (R Foundation for Statistical Computing, Vienna, Austria). Descriptive demographics, and clinical and radiological data are shown as mean ± SDs, numbers, or median and interquartile ranges. An unpaired two-sample Student's t-test was performed to determine if the development and external validation data were significantly different in terms of age, NIHSS score, lesion volume, or onset to MRI time. A *p*-value < 0.05 was considered statistically significant. The intraclass correlation coefficient (ICC) and its 95% confidence interval (CI) were computed between the two infarct volume measurements. A *p*-value < 0.05 was considered to identify a statistically significant correlation, given the null hypothesis of no relationship between the two measurements. Bland–Altman analysis was performed by computing the mean difference and 95% limits of agreement (LOA) between the two volume measurements.

#### *2.8. Data Statement*

The datasets in the present study are not publicly available since they have private patient information. The de-identified data are available from the corresponding author upon request after approval of the institutional review board of Samsung Medical Center.

#### **3. Results**

A total of 102 patients satisfied the criteria of < 6 h stroke onset to MRI time, MCA occlusion, and baseline NIHSS score of ≥ 4. Of 102 patients, 40 patients had an mTICI score of 0–2a (comprising the UR group), and 62 patients had an mTICI score of 2b–3 (comprising the SR group). A total of 10 patients were excluded, since their image data were problematic in terms of (1) severe motion artifacts in PWI data (*n* = 8), (2) susceptibility artifact near frontal brain regions (*n* = 1), and (3) infarcts detected in unexpected locations on day-7 DWI, which were related to stroke recurrence or procedural complications (*n* = 1). Hence, out of 92 patients, the final numbers of SR and UR patients were 56 and 36, respectively (Table 1). In both the SR and UR groups, there were no statistically significant differences in age, male sex, baseline NIHSS score, onset to MRI time, baseline DWI and PWI lesion volumes, and day-7 DWI lesion volumes between the development and external validation groups, except for the baseline NIHSS score between the UR development and external validation groups (Table 1).

**Table 1.** Patient characteristics for machine learning (ML) model development and external validation.


\* Median (IQR), Abbreviations: SR, successful recanalization; UR, unsuccessful recanalization; mTICI, modified treatment in cerebral infarction; SD, standard deviation; NIHSS, National Institutes of Health Stroke Scale; IQR, interquartile range; MRI, magnetic resonance imaging; DWI, diffusion-weighted imaging; PWI, perfusion-weighted imaging; IV, intravenous; tPA, tissue plasminogen activator; EVT, endovascular treatment.

For the UR model, the five-fold cross-validation resulted in a mean accuracy (SD) of 74.6% (2.5%) and a 95% CI of 69.8%–79.4%. For the SR model, the five-fold cross-validation resulted in a mean accuracy (standard deviation) of 76.4% (6.5%) and a 95% CI of 63.7%–89.2%.

Final infarct predictions using the UR and SR models are shown for four different cases (Figure 2; patient A: UR with large infarct growth; patient B: SR with small infarct growth; patient C: UR with small infarct growth; and patient D: SR with large infarct growth). The large differences between the UR-predicted and SR-predicted volumes were observed in both patients A and B, but the failure of revascularization led to the large final infarct volume (243 mL) and unfavorable outcome (90-day mRS = 5) in patient A, whereas the success of revascularization led to the small final infarct volume (4 mL) and favorable outcome (90-day mRS = 0) in patient B. Regarding the actual infarct growth, patient A's (239 mL) was significantly larger than patient C's (28 mL), while patient D's (69 mL) was significantly larger than patient B's (3 mL). All four cases had target mismatch and showed substantial differences in infarct growth volume, regardless of recanalization status.

**Figure 2.** Sample external validation cases of SR- and UR-based infarct predictions. (**A**) A case of large infarct growth in a UR patient (intravenous thrombolysis, modified treatment in cerebral infarction (mTICI) = 0, and 90-day modified Rankin Scale (mRS) = 5). (**B**) A case of small infarct growth in an SR patient (endovascular therapy, mTICI = 2b, and 90-day mRS = 0). (**C**) A case of small infarct growth in a UR patient (endovascular therapy, mTICI = 1, and 90-day mRS = 2). (**D**) A case of large infarct growth in an SR patient (endovascular therapy, mTICI = 2b, and 90-day mRS = 5). Predicted infarct masks, shown in red, are overlaid on pre-treatment DWI images for UR and SR model predictions. All four patients had target mismatch presence but showed various degrees of infarct growth. The yellow numbers indicate the following volumes: (from left to right) baseline DWI lesion volume, baseline PWI lesion volume, UR-predicted infarct volume, SR-predicted infarct volume, and day-7 infarct volume.

In the UR model validation, the ICC between manual infarct volume and UR-predicted infarct volume was 0.73 (95% CI = 0.31–0.91, *p* < 0.01; Figure 3A). In the SR model validation, the ICC between manual infarct volume and SR-predicted infarct volume was 0.87 (95% CI = 0.73–0.94, *p* < 0.001; Figure 3B). The mean difference (95% LOA obtained from Bland–Altman analysis) was −32.5 mL (−126.9 mL, 61.9 mL) for the UR model validation and 3.5 mL (−48.2 mL, 55.2 mL) for the SR model validation.

In all the external validation subjects, the presented ML model had a median Dice similarity coefficient (DSC) of 0.49 (IQR, 0.37–0.59), which was comparable to the DSC of 0.53 (IQR, 0.31–0.68) in the study by Yu et al. [19]. The presented ML model had a median DSC of 0.43 (IQR, 0.20–0.52) in the SR external validation subjects and a median DSC of 0.58 (IQR, 0.55–0.67) in the UR external validation subjects.

In external validation SR patients (*n* = 27), there was a statistically significant difference in actual infarct growth volume between the mismatch and non-mismatch groups (*p* = 0.02; Figure 4A). It was also observed that there was a statistically significant difference in actual infarct growth volume between the low and high SR predicted infarct growth volume groups (*p* = 0.01; Figure 4B). *p*-values between the low and intermediate groups and between the intermediate and high groups were 0.08 and 0.15, respectively (Figure 4B).

**Figure 3.** Correlation plots of (**A**) the UR-predicted infarct volume and (**B**) the SR-predicted infarct volume on the external validation UR and SR cohorts, respectively. The manual infarct volume measurement on day-7 DWI served as the reference.

**Figure 4.** Infarct growth distribution in the external validation SR patients (*n* = 27). (**A**) Target mismatch classification. (**B**) SR-predicted infarct growth classification. In (**A**), the mismatch absence group shows a heterogeneous distribution of actual infarct growth. In (**B**), the high group (SR-predicted infarct growth ≥ 35 mL) shows a heterogeneous distribution of actual infarct growth, and the low group (SR-predicted infarct growth < 15 mL) shows the narrowest distribution of the actual infarct growth.

The percentage values of favorable clinical outcomes of the day-90 mRS score ≤ 1 for the mismatch presence and absence groups in overall SR patients were 50% and 36%, respectively (Figure 5A). The percentage values of favorable clinical outcomes were 61%, 56%, and 25%, respectively, for the low, intermediate, and high SR-predicted infarct growth groups (Figure 5A). In UR patients, the percentage values of favorable outcomes for the mismatch presence and absence groups were 14% and 0%, respectively, while they were 9%, 18%, and 7%, respectively, for the low, intermediate, and high UR-predicted infarct growth groups (Figure 5B).

**Figure 5.** Distribution of modified Rankin Scale (mRS) scores at day-90 by target mismatch (top) and ML-predicted infarct growth (bottom). In both (**A**) SR and (**B**) UR patients, a lower ML-predicted infarct growth was related to a lower mRS score at day 90.

#### **4. Discussion**

This study suggests that the ML-predicted infarct growth volumes can offer a novel insight into selecting patients and predicting clinical outcome. The method may be a useful alternative to target mismatch, as it provides a direct way to measure infarct distribution and growth. Although the number of the mismatch absence group is small in this study, it indicates that the patient selection based on the ML-predicted infarct growth volume may have the infarct growth prediction performance comparable to target mismatch-based selection. In particular, the low infarct growth group showed a narrow distribution of the actual infarct growth, and this may indicate that once a patient was classified into the low group, a small infarct growth as well as a favorable clinical outcome would be highly likely. It is important to note that traditional PWI–DWI mismatch is evaluated only based on baseline DWI and PWI images without consideration of the final infarct images, while, in the presented approach, the SR and UR infarct volumes are estimated using ML models trained on baseline DWI and PWI images and the final infarct annotated images.

Ischemic penumbra is defined as a hypoperfused area that can regain function with rapid reperfusion [20]. This area is at a risk of infarction without rapid reperfusion, and infarct growth occurs in the absence of reperfusion. In the present study, mean infarct growth in the UR group was higher than that in the SR group, but the degree of infarct growth was highly variable in both UR and SR groups. The large variability of infarct growth within the SR (or UR) group may be attributed to the variable degrees of perfusion defect severities and ischemic damage in the penumbra zone or ischemic core.

Traditional threshold-based lesion volume estimation approaches have limitations, since the appropriate threshold values often depend on MRI sequence parameters, scanner vendors, and software analysis tools. ML takes a different perspective by learning a non-linear function from input/output relationships and has the potential to overcome the limitations of traditional threshold-based volume estimation approaches. Although the evaluation regarded traditional threshold-based methods used for target mismatch assessment as a reference, it is expected that the ML-based method could provide additional information in cases where threshold-based volume estimation is unsatisfactory. Until now, the Diffusion and perfusion imaging Evaluation For Understanding Stroke Evolution (DEFUSE) trial, Echo-Planar Imaging Thrombolytic Evaluation Trial (EPITHET), and other studies have

attempted to find the optimal Tmax threshold and mismatch ratio [1,2,21,22]. Similarly, Olivot and colleagues showed that besides mismatch volume, lesion geography and structure also determine infarct growth [23]. Lastly, the rate of infarct growth during the few days after stroke onset varies substantially between patients, and certain patients have slow infarct growth while others have rapid infarct growth [24]. ML-based prediction has the potential to overcome the limitations of traditional approaches of estimating ischemic penumbra.

Current mismatch concepts provide information on the likelihood ratio and number needed to treat for a favorable outcome. In the Highly Effective Reperfusion evaluated in Multiple Endovascular Stroke trials (HERMES), a meta-analysis of individual patient data from five randomized trials of EVT, the odds ratio was 2.49 and the number needed to treat with EVT to reduce disability by at least one level of the mRS for one patient was 2.6 [25]. On the contrary, our ML-based prediction method can produce the visualization of the degree of DWI lesion growth, which can be shown to patients or their guardians, prior to revascularization therapy. ML predictions using our custom tool via post processing typically took approximately 8 min. Using the current strict recanalization criteria based on the mismatch concept, we can calculate treatment effect sizes, but at the cost of limiting the benefit to a small portion of our stroke patients [26]. With ML, treatment approaches may be more customized, rather than dichotomized into either withholding or offering treatment. In addition, Oppenheim and colleagues showed that in patients without SR, DWI volume > 145 mL within 14 h of onset reliably predicted a malignant MCA infarction [27]. ML-based prediction of infarct volume at the peak time of cytotoxic edema (at 3–4 days after the infarction) may guide early management, such as decompressive hemicraniectomy.

This study has several limitations. First, a larger mean difference of predicted infarct volume was observed for UR than for SR. The inaccuracy may be attributed to the large variability of infarct growth rate, which varies from person to person and is highly unpredictable [28]. It may be worth investigating the division of patient grouping based on the infarct growth rate and develop individual ML models. Second, a custom software tool was used to measure the pre-treatment PWI and DWI lesion volumes based on Tmax and ADC thresholds, for the target mismatch assessment. ML-based methods for lesion volume estimation may be alternative tools for mismatch evaluation [29,30]. Third, only ADC and rTTP were used for the ML model development in the present study. Consideration of other MRI sequences, such as fluid-attenuated inversion recovery (FLAIR), may improve the accuracy of infarct growth prediction [31]. Onset to imaging time can affect the volume of the ischemic penumbra zone, as well as the final infarct volume. Recently, the infarct growth rate estimated from the baseline DWI lesion volume and time of stroke onset was reported to be associated with penumbral salvage and clinical outcomes after EVT reperfusion [28]. In a similar fashion, we are currently investigating the feasibility of an improved prediction of infarct growth with the incorporation of relative FLAIR, which is indicative of a 'tissue clock', into the prediction model. Fourth, Lev and colleagues reported the importance of 'location-weighted' scoring over simple volumetric data of penumbra areas [32,33]. For instance, in patients with similar infarct volumes, different severities of neurologic deficits can be observed, depending on the clinical features and lesion site. Hence, location-weighted ML-based prediction of functional outcome is worthy of investigation. Fifth, clinical variables were not used in this study. Previous studies demonstrated the effectiveness of clinical variables in the prediction of infarct growth [34] and clinical outcome [35], while our study was solely based on image features. The inclusion of clinical variables may help improve the outcome prediction. Finally, this was a single-center study performed on a 3T scanner, with only a small number of patients available for analysis. A larger prospective study is necessary to evaluate the benefit of the ML-based method.

#### **5. Conclusions**

ML-based prediction of tissue fate was demonstrated using two ML models trained on data from patients with and without recanalization. The presented ML-based method provides the estimations of voxel-wise infarct distributions and final infarct volumes in cases of SR and UR. The classification of patients in terms of ML-predicted infarct growth offers novel and alternative ways to predict infarct growth and functional outcome, when compared with the traditional target mismatch classification. As ML is data-driven and allows sophisticated feature engineering, the ML prediction is expected to improve as more data are collected and careful ML modeling is made based on multi-modal MRI. The advancement of technology will implicate more accurate predictions of the final infarct distribution, potentially providing clinicians with precise information for the guidance of treatment selection and the prediction of clinical outcome.

**Author Contributions:** Conceptualization, O.Y.B. and Y.-C.K.; methodology, O.Y.B. and Y.-C.K.; software, Y.-C.K.; validation, O.Y.B., Y.-C.K., H.J.K., J.-W.C.; formal analysis, O.Y.B.; investigation, O.Y.B.; resources, O.Y.B., H.S.N., P.J., K.H.K., M.J.S.; data curation, O.Y.B., I.K.; writing—original draft preparation, O.Y.B., Y.-C.K.; writing—review and editing, O.Y.B., Y.-C.K., J.-W.C., H.K., W.-K.S., G.-M.K.; visualization, O.Y.B., Y.-C.K.; supervision, O.Y.B., G.-M.K.; project administration, O.Y.B., I.G.K.; funding acquisition, O.Y.B. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was supported by the National Research Foundation of Korea, grant numbers 2018R1A2B2003489 and 2018R1D1A1B07042692.

**Conflicts of Interest:** The authors declare no conflict of interest. The funding sources had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.

#### **Appendix A**

#### *Comparison of Tmax* > *6 s Volume and rTTP* > *4.5 s Volume*

A custom MATLAB (Mathworks, Natick, MA, USA) module was developed to calculate Tmax from PWI time series image data and compare it with rTTP. An axial slice with middle cerebral arteries was selected, and an arterial input function (AIF) was automatically calculated by the cluster analysis proposed by Mouridsen et al. [36]. Standard singular value decomposition (SVD) [37] was used to estimate Tmax, defined to be the delay of the peak of the residue function. Figure A1-A compares Tmax > 6 s volume with rTTP > 4.5 s volume. Pearson correlation coefficient between Tmax > 6 s and rTTP > 4.5 s volumes was 0.84 (95% CI = 0.75–0.89, *p*-value < 0.0001). The mean difference (95% limits of agreement from Bland-Altman analysis) was −2.4 mL (−66.3 mL, 61.5 mL). Figure A1-B compares Tmax > 10 s volume with rTTP > 9.5 s volume. Pearson correlation coefficient between Tmax > 10 s and rTTP > 9.5 s volumes was 0.81 (95% CI = 0.72–0.87, *p*-value < 0.0001). The mean difference (95% limits of agreement from Bland–Altman analysis) was 7.5 mL (−42.9 mL, 57.9 mL).

**Figure A1.** Comparison of time-to-maximum (Tmax) and relative time to peak (rTTP) lesion volumes. (**A**) Perfusion lesion volumes measured by Tmax and rTTP. (**B**) Severe ischemic lesion volumes measured by Tmax and rTTP.

#### **References**


#### *J. Clin. Med.* **2020**, *9*, 1977


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

### *Case Report* **Acute Middle Cerebral Artery Occlusion Detection Using Mobile Non-Imaging Brain Perfusion Ultrasound—First Case**

**Mustafa Kilic 1, Christina Wendl 2, Sibylle Wilfling 1, David Olmes 1, Ralf Andreas Linker <sup>1</sup> and Felix Schlachetzki 1,\***


**Abstract:** Mobile brain perfusion ultrasound (BPU) is a novel non-imaging technique creating only hemispheric perfusion curves following ultrasound contrast injection and has been specifically designed for early prehospital large vessel occlusion (LVO) stroke identification. We report on the first patient investigated with the SONAS® system, a portable point-of-care ultrasound system for BPU. This patient was admitted into our stroke unit about 12 h following onset of a fluctuating motor aphasia, dysarthria and facial weakness resulting in an NIHSS of 3 to 8. Occlusion of the left middle cerebral artery occlusion was diagnosed by computed tomography angiography. BPU was performed in conjunction with injection of echo-contrast agent to generate hemispheric perfusion curves and in parallel, conventional color-coded sonography (TCCS) assessing MCAO. Both assessments confirmed the results of angiography. Emergency mechanical thrombectomy (MT) achieved complete recanalization (TICI 3) and post-interventional NIHSS of 2 the next day. Telephone follow-up after 2 years found the patient fully active in professional life. Point-of-care BPU is a non-invasive technique especially suitable for prehospital stroke diagnosis for LVO. BPU in conjunction with prehospital stroke scales may enable goal-directed stroke patient placement, i.e., directly to comprehensive stroke centers aiming for MT. Further results of the ongoing phase II study are needed to confirm this finding.

**Keywords:** prehospital stroke diagnosis; large vessel occlusion; ultrasound; thrombectomy; brain perfusion; SONAS®; prehospital stroke scales; point-of-care diagnostics

#### **1. Introduction**

Mechanical thrombectomy (MT) has evolved as the most efficient therapy in cerebral ischemic stroke due to large vessel occlusion (LVO) achieving a 20–27% absolute increase in patients regaining functional independence compared to patients without receiving MT mboxciteB1-jcm-1747802,B2-jcm-1747802. These impressive results, however, are being achieved by rigorous patient selection by cerebral perfusion imaging, achievement of successful recanalization and short symptom to recanalization times including transport to comprehensive stroke centers (CSC) [3,4]. In rural areas, stroke units (SU) and telestroke networks have been established only with intravenous thrombolysis (IVT) capabilities more than 20 years ago, and further transfer to CSC for MT is time-consuming, causing delay in the initiation of recanalization therapy.

Still, prehospital stroke detection algorithms employ clinical scales deemed imperfect for detecting LVO. Thus, transfer of severe stroke patient to either CSCs or the next regional SU remains a complex decision [5]. Ultrasound is currently the only mobile imaging modality widely available for brain vascular diagnostics. Our group demonstrated that transcranial color-coded sonography (TCCS) using mobile color-Duplex point-of care

**Citation:** Kilic, M.; Wendl, C.; Wilfling, S.; Olmes, D.; Linker, R.A.; Schlachetzki, F. Acute Middle Cerebral Artery Occlusion Detection Using Mobile Non-Imaging Brain Perfusion Ultrasound—First Case. *J. Clin. Med.* **2022**, *11*, 3384. https:// doi.org/10.3390/jcm11123384

Academic Editors: Hyo Suk Nam and Byung Moon Kim

Received: 13 May 2022 Accepted: 2 June 2022 Published: 13 June 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

ultrasound (POCUS) can be performed in the prehospital phase with very high sensitivity and specificity for the detection of middle cerebral artery (MCA) occlusion [6,7]. Another highly efficient alternative, the mobile stroke units (mSU) with computed tomography (CT), including CT-angiography installed in an ambulance to perform at the patient's site, is of limited availability, costly and has a rather small range. The air-mobile SU version of the mSU is still in the concept phase [8,9]. However, the drawback of TCCS is the high neurological expertise requiring extensive practical and theoretical training programs for paramedics to employ the pathophysiological driven diagnostic concept for TCCS, i.e., focusing on the right middle cerebral artery in left-sided hemiparesis and neglect [10].

We recently published the results of a phase I study using a mobile battery driven brain perfusion ultrasound (BPU) device SONAS® (BURL Concepts, Inc., San Diego, CA, USA) in healthy volunteers with good correlation of measured hemispheric brain perfusion to perfusion weighted MRI [11]. This case report describes the BPU finding in a patient with acute MCA occlusion confirmed by CT-angiography and TCCS prior to MT.

#### **2. Materials and Methods**

During initiation of a phase I–II study at the University of Regensburg, Department of Neurology at the medbo Bezirksklinikum (approval by the local ethics committee, IRB protocol number 2018-001279-19, in accordance with the World Medical Association Declaration of Helsinki), an acute stroke patient was admitted and individual informed consent in the presence of his parents as well as post-hoc consent for publication 1 year later given. The phase I–II study SONAS enrolled acute stroke patients with perfusion CT and was registered at ClinicalTrials.gov (NCT03897153). In brief, patients with acute ischemic stroke within 24 h will be investigated using the SONAS® device after confirmation of an either proximal, middle or distal MCA main stem occlusion or distal internal carotid artery occlusion (including carotid-T occlusion). Confirmation of the LVO will be performed by cerebral magnetic resonance imaging (cMRI) or cerebral imaging computed tomography (cCT), including perfusion weighted (pw) imaging sequences. The study has now been completed and data analysis is in preparation.

#### *2.1. Case Report*

The 46-year-old male patient was transferred for suspected left hemispheric ischemia as on admission, he presented with moderate dysarthria, motor aphasia and mild facial paresis (NIHSS 3) starting early in the morning. He reported that about 4–5 weeks prior, one side of the face had felt strange for 10–15 min, suggestive of a transient ischemic attack. Admission was approximately 12 h after stroke onset. At this time, cCT was almost unremarkable (ASPECTS Score 9), while CT-angiography revealed proximal MCA occlusion (M1 segment) on the left side with good collateralization (collateral score 3–4, Figure 1). With fluctuating symptoms between progressive global aphasia and brachiocephalic hemiparesis (NIHSS 3 to 8), and an overall deficit relevant to daily living, the decision was made for MT, especially since no intravenous thrombolysis was initiated. Informed consent for additional ultrasonography was obtained from the patient and his parents and retrospectively. Both TCCS and BPU were performed in parallel to MT preparation in general anesthesia without causing a time delay.

Digital subtraction angiography demonstrated proximal MCA occlusion with good cortical collaterals via leptomeningeal anastomosis predominantly from the anterior cerebral artery (Figure 2). After successful MT (TICI 3), the patient was extubated and presented only mild word finding difficulties and minimal facial paresis (NIHSS 2). On telephone follow up two years later, he was back to work and neurologically unremarkable (modified Rankin Scale 0).

**Figure 1. Left**: unenhanced CT, putaminal hypodensity (arrow), ASPECTS score 9; **middle**: axial maximum intensity projection (MIP) reconstruction of CT-angiography showing left MCA occlusion (arrow); **right**: coronal MIP reconstruction of left MCA occlusion (arrow), good leptomeningeal collaterals can be seen.

**Figure 2.** Digital subtraction angiography. **Left** to **right**: phase after contrast injection showing proximal MCA occlusion (arrow); demonstration of extensive leptomeningeal collaterals arising from the ACA (arrows); immediately after mechanical thrombectomy by first use of the stent retriever showing suspicious residual thrombus with mild lumen narrowing (arrow) and, right, complete recanalization with mild stenosis/vasospasm (TICI 3).

#### *2.2. Transcranial Color-Coded Sonography (TCCS)*

TCCS was performed using a standard color Duplex ultrasound system equipped with a low-frequency phased array transducer (Philips, Affiniti 70, S5-1 cardiac sector probe, Amsterdam, The Netherlands). Transtemporal insonation for detection of the basal cerebral arteries was performed as previously described [12]. In brief, after detection of a sufficient temporal bone window by identification of the mesencephalic peduncle and the contralateral skull in B-mode color Duplex mode helped to identify patent cerebral arteries with flow towards the transducer depicted in red, and flow away from the transducer in blue. Pulsed Doppler measurements were performed to quantify flow velocities. An occluded MCA can be assumed when the anterior cerebral artery can be depicted confirming a diagnostic useful temporal bone window.

#### *2.3. Brain Perfusion Ultrasound (BPU) Using SONAS®*

BPU was performed as previously described employing the CE-certified (Class IIa), non-imaging ultrasound device SONAS®, a portable, battery-powered device to generate hemispheric time intensity curves after injection of an ultrasound enhancing agent, which in this case was SonoVue® [11,13]. In brief, bilateral low-frequency transducer with low transmit frequencies and power (220 kHz, 2% alternating duty cycle from right to left, TIC and MI < 1.0) measure the 4th to 6th harmonic frequencies generated by intravenous 2.4 mL microbubble injection. Time to peak (TTP) curves are generated based on measurements

contralateral, that is transmission and reception are on opposite sides of the head, and ipsilateral measurement, where transmission and receiving are ipsilateral (Figure 3). Using an automatic peak detection algorithm for both brain hemispheres separately, hemispheric TTP values are compared. The differential of peaks is computed and shown as a delta-TTP (dTTP) value.

**Figure 3.** Upper figure: signal transmission and recording from both ultrasound transducers after injection of echo-enhancing agents demonstrating the mode of operation. R—right, L—left. Lower figure: time intensity curves after contrast agent injection with X-axis time after injection and Y-axis maximum signal backscatter analyzed automatically by BPU machine (SONAS®). The first peak is due to the first appearance of contrast agent while the second and third peaks are to due systemic reflow phenomenon.

#### **3. Results**

Prior to embolectomy and 30 min after neurological deterioration, both, BPU und TCCS were performed. BPU demonstrated significant delay in left hemispheric perfusion with a delay of 6.6 s (Figure 4).

**Figure 4.** Analysis of time intensity curves representing brain perfusion ultrasound using SONAS®. Upper row: 1—first time-to-peak (TTP) in the "healthy" right hemisphere, 2—delayed left hemispheric peak, 3—peak of the whole brain, and 4—delayed second peak probably representing right anterior cerebral artery crossflow as the primary collateral pathway. Middle row: absolute time values of the peaks. Lower row: expression of the difference in time-to-peak values (delta-TTP) expressing the absolute perfusion deficit in the left hemisphere.

Conventional TCCS demonstrated only only slow proximal pseudo-venous flow in the proximal MCA (peak systolic flow less than 10 cm/s) consistent with MCA occlusion and relatively normal flow in the left ACA (Figure 5).

**Figure 5.** TCCS scan through the left temporal bone window with transcranial color-coded sonography in the upper and Doppler spectrum from the Doppler gate. **Left**: Doppler gate placed at the origin of the middle cerebral artery with absent color filling (arrow) in the M1 segment and pseudo-venous flow at the beginning of the M1-Segment (Doppler gate depth 55 mm). **Right**: anterior cerebral artery color-coded in red due to aliasing due to increased flow velocities with the Doppler gate placed at a depth of 66 mm in the A1-segment showing increased flow of 63 cm/s mean maximum flow and decreased resistance index of 0.39 all suggestive of collateral flow.

#### *Case Report Continued*

Decision for embolectomy under general anesthesia was made based on the persistent MCA occlusion and fluctuating course of neurological deficits (NIHSS 3 to 8). The first angiographic image was made at 20:45 h and complete TICI 3 recanalization was noted at 21:17 h. After extubation, NIHSS score was 3 and 2 in the following days. Neuropsychological testing showed mild cognitive impairment and indication for neurorehabilitation. At telephone follow-up 2 years later, the patient was again able to work and independent in daily life (mRS score of 0–1).

#### **4. Discussion**

This is the first report describing the application of BPU in a patient with acute MCA occlusion shown by CT-angiography and later digital subtraction angiography using a portable POCUS device for BPU (SONAS®) with comparison to standard TCCS as another mobile method for acute stroke diagnosis, while TCCS allowed MCA occlusion imaging with similar information as CT-angiography and DSA BPU depicts the subsequent brain perfusion changes. If replicated in the ongoing phase II study, BPU may be feasible in prehospital stroke diagnostics.

The progressive developments in the field of MT with high recanalization rates along with decreasing peri-interventional risk challenge the identification and allocation of acute stroke patients in the prehospital phase. In stroke patients in need for MT, direct transfer to an endovascular MT-capable center may be preferred as opposed to transport to the closest primary stroke center and secondary transport for MT [14]. To date, efforts have been made to identify LVO employing prehospital stroke scales as shown in the PRESTO study comparing eight scales employed by paramedics in the prehospital phase and identifying three scales with acceptable-to-good accuracy [15]. In another prospective study on patients with ischemic stroke entered in the Dyon stroke registry, Duloquin et al. applied 16 different prehospital stroke scales; however, a priori excluding patients with ICH severely reducing its value as a prehospital study [16]. Even with this drawback sensitivities of scales were still low (identification of LVO ranging between 0.64 to 0.79, sensitivity 59% to 93%, and specificity ranging from 34% to 89%, c-statistic) with 174 of 971 patients (17.9%) had LVO (defined as occlusion of M1 and M2-segment of the MCA and basilar artery occlusion). In a topical review from 2020, van Gaal and Demchuck listed common flaws of studies on the validity of prehospital stroke scales, among them, exclusion of ICH patients, lack of prospective studies in the field as opposed to application of stroke scales to pure ischemic stroke databases (exclusion, i.e., TIA, ICH, stroke mimics, amongst others), and whether or not these scales have indeed been tested by paramedics [17]. However, in a recent pilot study in the Baltimore metro area patients were re-routed CSCs for MT upon application of the Los Angeles Motor Scale. Consequently, significantly shorter procedural times from symptom to MT onset of 119 min were achieved [18]. While the faster initiation of MT showed a strong non-significant trend for better outcome, more than 50% of patients were allocated to the wrong clinic.

However, fast and goal-directed prehospital point-of-care diagnostics such as blood serum biomarkers and TCCS have been advocated to further accelerate the diagnostic and therapeutic stroke pathways in stroke patients [19–22]. Apart from mobile stroke units, only portable TCCS is a neuroimaging tool capable of visualizing LVO in the field, preferentially employing echo-enhancing agents [23,24]. In the patient described here, conventional TCCS identified left MCA occlusion in line with CT-angiography and digital subtraction angiography. BPU, which involves single-slice transcranial ultrasound using phased-array transducers and echo-enhancing agents to produce perfusion maps comparable to perfusion CT, is not currently mature enough to be used for routine practice, let alone prehospital use [25,26]. Yet, for conventional TCCS, detailed pathophysiological hypothesis of the stroke location and hands-on expertise in identifying an adequate temporal bone window is needed. However, many of these obstacles may be solved by employing telemetric and artificial intelligence support [27,28]. BPU using the SONAS® device differs significantly from TCCS and other transcranial ultrasound techniques as it lacks anatomical information and presents only perfusion curves [11]. However, the demonstrated delay in time-to-peak times of the left hemisphere together with the clinical information can be easily gathered with little expertise, ideal for application in the prehospital phase. Table 1 compares the basic findings, techniques, advantages and disadvantages of TCCS and non-imaging BPU using SONAS®.

**Table 1.** Comparison of TCCS and BPU.


#### **5. Conclusions**

This is the first patient with MCA occlusion in whom both the BPU technique with the SONAS® device and standard TCCS were used to visualize MCA occlusion and the resulting perfusion deficit. The results of BPU using the SONAS® device were compared with conventional imaging (CT angiography, TCCS and DSA) and revealed a similar result. The user-friendly mode of operation of BPU requiring little additional skills indicate the potential of SONAS® to fasten stroke treatment when positively identifying LVO at the earliest time point possible after symptom onset. Goal-directed hospital admission for MT and fewer secondary transportations may enable faster symptom to recanalization times, and eventually better outcome. The results of the ongoing phase II study with the SONAS® device in acute ischemic stroke patients as well as further studies on lacunar infarction, intracerebral hemorrhage, minor strokes and stroke mimics are the prerequisite to define the full potential of BPU.

**Author Contributions:** M.K. and F.S. contributed equally to the patient investigation and data collection, analysis of the results, and writing of the manuscript. C.W., S.W., D.O. and R.A.L. contributed to the analytical methods and data analysis. The main draft was written by M.K. and F.S. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Consent to publish the results was given by all included participants especially the patient.

**Data Availability Statement:** The raw data underlying this article are intended for publication on a suitable platform and can be made available by the corresponding author on reasonable request.

**Acknowledgments:** The authors would like to thank the patient for approving the investigation and publication, and Burl Concepts, Inc. for technical support and data analysis of this extra patient.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**

