Automated image registration of cerebral digital subtraction angiography.

PURPOSE: Our aim is to automatically align digital subtraction angiography (DSA) series, recorded before and after endovascular thrombectomy. Such alignment may enable quantification of procedural success. METHODS: Firstly, we examine the inherent limitations for image registration, caused by the projective characteristics of DSA imaging, in a representative set of image pairs from thrombectomy procedures. Secondly, we develop and assess various image registration methods (SIFT, ORB). We assess these methods using manually annotated point correspondences for thrombectomy image pairs. RESULTS: Linear transformations that account for scale differences are effective in aligning DSA sequences. Two anatomical landmarks can be reliably identified for registration using a U-net. Point-based registration using SIFT and ORB proves to be most effective for DSA registration and are applicable to recordings for all patient sub-types. Image-based techniques are less effective and did not refine the results of the best point-based registration method. CONCLUSION: We developed and assessed an automated image registration approach for cerebral DSA sequences, recorded before and after endovascular thrombectomy. Accurate results were obtained for approximately 85% of our image pairs.

Spatio-temporal deep learning for automatic detection of intracranial vessel perforation in digital subtraction angiography during endovascular thrombectomy.

Intracranial vessel perforation is a peri-procedural complication during endovascular therapy (EVT). Prompt recognition is important as its occurrence is strongly associated with unfavorable treatment outcomes. However, perforations can be hard to detect because they are rare, can be subtle, and the interventionalist is working under time pressure and focused on treatment of vessel occlusions. Automatic detection holds potential to improve rapid identification of intracranial vessel perforation. In this work, we present the first study on automated perforation detection and localization on X-ray digital subtraction angiography (DSA) image series. We adapt several state-of-the-art single-frame detectors and further propose temporal modules to learn the progressive dynamics of contrast extravasation. Application-tailored loss function and post-processing techniques are designed. We train and validate various automated methods using two national multi-center datasets (i.e., MR CLEAN Registry and MR CLEAN-NoIV Trial), and one international multi-trial dataset (i.e., the HERMES collaboration). With ten-fold cross-validation, the proposed methods achieve an area under the curve (AUC) of the receiver operating characteristic of 0.93 in terms of series level perforation classification. Perforation localization precision and recall reach 0.83 and 0.70 respectively. Furthermore, we demonstrate that the proposed automatic solutions perform at similar level as an expert radiologist.

autoTICI: Automatic Brain Tissue Reperfusion Scoring on 2D DSA Images of Acute Ischemic Stroke Patients.

The Thrombolysis in Cerebral Infarction (TICI) score is an important metric for reperfusion therapy assessment in acute ischemic stroke. It is commonly used as a technical outcome measure after endovascular treatment (EVT). Existing TICI scores are defined in coarse ordinal grades based on visual inspection, leading to inter- and intra-observer variation. In this work, we present autoTICI, an automatic and quantitative TICI scoring method. First, each digital subtraction angiography (DSA) acquisition is separated into four phases (non-contrast, arterial, parenchymal and venous phase) using a multi-path convolutional neural network (CNN), which exploits spatio-temporal features. The network also incorporates sequence level label dependencies in the form of a state-transition matrix. Next, a minimum intensity map (MINIP) is computed using the motion corrected arterial and parenchymal frames. On the MINIP image, vessel, perfusion and background pixels are segmented. Finally, we quantify the autoTICI score as the ratio of reperfused pixels after EVT. On a routinely acquired multi-center dataset, the proposed autoTICI shows good correlation with the extended TICI (eTICI) reference with an average area under the curve (AUC) score of 0.81. The AUC score is 0.90 with respect to the dichotomized eTICI. In terms of clinical outcome prediction, we demonstrate that autoTICI is overall comparable to eTICI.