Abstract 1122‐000066: Number‐Needed‐to‐Review: A Novel Metric to Assess Triage Efficiency of Large Vessel Occlusion Detection Systems

Introduction : Endovascular thrombectomy is the gold standard treatment for acute ischemic strokes with large vessel occlusions (LVO). Manual image analysis is often time consuming and requires clinicians to be skilled in reading perfusion scans, as well as vessel images. RapidAI software has an automated processor to detect LVO of the middle cerebral artery and is analyzed in this study. A novel metric, number‐needed‐to‐review (NNR), is introduced to assess the clinical efficiency of this software. Methods : This is a retrospective review of patients with a suspected ischemic stroke and an image processed by RapidAI software from 11/1/2020 to 4/30/2021 at a regional hospital system. Only M1 LVOs were included. Sensitivities, specificities, positive predictive value (PPV), and negative predictive value (NPV) were calculated for the following: Rapid LVO detection, gaze deviation (GD), hyperdense sign (HDS), Tmax >6 seconds, and NIHSS at presentation. The NNR was calculated for an M1 occlusion. Results : 559 patients were included in this study. M1 occlusion was detected in 42 (7.5%) cases. Rapid LVO detection software was found to have a sensitivity of 71%, specificity of 94%, PPV of 49%, and NPV of 92% for a M1 occlusion. When both GD and HDS were combined with Rapid LVO, the specificity and PPV increased to 100% for a M1 occlusion. A negative LVO software combined with either a low (<15 mL on Tmax >6s) or high (<50 mL on Tmax >6s) Tmax threshold were found to have a specificity and PPV of 100% for no M1 occlusion. The combination of GD, HDS, Rapid LVO+ (for M1 occlusion) and Rapid LVO‐ with a low Tmax threshold (for no M1 occlusion) yielded 24 images NNR per 100 cases. When the combination of GD, HDS, Rapid LVO+ was combined with Rapid LVO‐ and a high Tmax threshold, the NNR per 100 cases was 16. With the addition of NIHSS<7 for the remaining cases in the high Tmax group, the NNR per 100 cases decreased to 9. Conclusions : The addition of GD and HDS to the Rapid LVO increases the specificity and PPV for a M1 occlusion. When combined with a negative Rapid LVO detection and either a low or high Tmax >6s threshold, the NNR is significantly decreased. As few as 9 images per 100 would be needed to be manually reviewed by a clinician during stroke triage.

EFM-Net: Feature Extraction and Filtration with Mask Improvement Network for Object Detection in Remote Sensing Images

Object detection is an essential task in computer vision. Many methods have made significant progress in ordinary object detection. Due to the particularity of remote sensing images, the detection target is tiny, the background is messy, dense, and has mutual occlusion, which makes the general detection method challenging to apply to remote sensing images. For these problems, we propose a new detection framework feature extraction and filtration method with a mask improvement network (EFM-Net) to enhance object detection ability. In EFM-Net, we designed a multi-branched feature extraction (MBFE) module to better capture the information in the feature graph. In order to suppress the background interference, we designed a background filtering module based on attention mechanisms to enhance the attention of objects. Finally, we proposed a mask generate the boundary improvement method to make the network more robust to occlusion detection. We tested the DOTA v1.0, NWPU VHR-10, and UCAS-AOD datasets, and the experimental results show that our method has excellent effects.

Validation of an artificial intelligence-driven large vessel occlusion detection algorithm for acute ischemic stroke patients

Rapid and accurate diagnosis of large vessel occlusions (LVOs) in acute ischemic stroke (AIS) patients using automated software could improve clinical workflow in determining thrombectomy in eligible patients. Artificial intelligence-based methods could accomplish this; however, their performance in various clinical scenarios, relative to clinical experts, must be thoroughly investigated. We aimed to assess the ability of Canon’s AUTOStroke Solution LVO application in properly detecting and locating LVOs in AIS patients. Data from 202 LVO and 101 non-LVO AIS patients who presented with stroke-like symptoms between March 2019 and February 2020 were collected retrospectively. LVO patients had either an internal carotid artery (ICA) ( n = 59), M1 middle cerebral artery (MCA) ( n = 82) or M2 MCA ( n = 61) occlusion. Computed tomography angiography (CTA) scans from each patient were pushed to the automation platform and analyzed. The algorithm’s ability to detect LVOs was assessed using accuracy, sensitivity and Matthews correlation coefficients (MCCs) for each occlusion type. The following results were calculated for each occlusion type in the study (accuracy, sensitivity, MCC): ICA = (0.95, 0.90, 0.89), M1 MCA = (0.89, 0.77, 0.78) and M2 MCA = (0.80, 0.51, 0.59). For the non-LVO cohort, 98% (99/101) of cases were correctly predicted as LVO negative. Processing time for each case was 69.8 ± 1.1 seconds (95% confidence interval). Canon’s AUTOStroke Solution LVO application was able to accurately identify ICA and M1 MCA occlusions in addition to almost perfectly assessing when an LVO was not present. M2 MCA occlusion detection needs further improvement based on the sensitivity results displayed by the LVO detection algorithm.