Machine Learning-Enabled Determination of Diffuseness of Brain Arteriovenous Malformations from Magnetic Resonance Angiography.

The diffuseness of brain arteriovenous malformations (bAVMs) is a significant factor in surgical outcome evaluation and hemorrhagic risk prediction. However, there are still predicaments in identifying diffuseness, such as the judging variety resulting from different experience and difficulties in quantification. The purpose of this study was to develop a machine learning (ML) model to automatically identify the diffuseness of bAVM niduses using three-dimensional (3D) time-of-flight magnetic resonance angiography (TOF-MRA) images. A total of 635 patients with bAVMs who underwent TOF-MRA imaging were enrolled. Three experienced neuroradiologists delineated the bAVM lesions and identified the diffuseness on TOF-MRA images, which were considered the ground-truth reference. The U-Net-based segmentation model was trained to segment lesion areas. Eight mainstream ML models were trained through the radiomic features of segmented lesions to identify diffuseness, based on which an integrated model was built and yielded the best performance. In the test set, the Dice score, F2 score, precision, and recall for the segmentation model were 0.80 [0.72-0.84], 0.80 [0.71-0.86], 0.84 [0.77-0.93], and 0.82 [0.69-0.89], respectively. For the diffuseness identification model, the ensemble-based model was applied with an area under the Receiver-operating characteristic curves (AUC) of 0.93 (95% CI 0.87-0.99) in the training set. The AUC, accuracy, precision, recall, and F1 score for the diffuseness identification model were 0.95, 0.90, 0.81, 0.84, and 0.83, respectively, in the test set. The ML models showed good performance in automatically detecting bAVM lesions and identifying diffuseness. The method may help to judge the diffuseness of bAVMs objectively, quantificationally, and efficiently.

Evaluating the association of preoperative parecoxib with acute pain trajectories after video-assisted thoracoscopic surgery: a single-centre cohort study in Taiwan.

OBJECTIVE: The efficacy of parecoxib as pre-emptive analgesia still remains controversial. This study aimed to investigate how pre-emptive analgesia with parecoxib affected postoperative pain trajectories over time in patients undergoing thoracic surgery. DESIGN: Retrospective cohort study. SETTING: A single medical centre in Taiwan. PARTICIPANTS: We collected 515 patients undergoing video-assisted thoracoscopic surgery at a tertiary medical centre between September 2016 and August 2017. INTERVENTIONS: Pre-emptive parecoxib before surgery. PRIMARY AND SECONDARY OUTCOME MEASURES: Daily numeric rating pain scores in the first postoperative week. RESULTS: A total of 196 (38.1%) of the recruited patients received parecoxib preoperatively. The latent curve analysis revealed that woman, higher body weight and postoperative use of parecoxib were associated with increased baseline level of pain scores over time (p=0.035, 0.005 and 0.048, respectively) but epidural analgesia and preoperative use of parecoxib were inclined to decrease it (both p<0.001). Regarding the decreasing trends of changes in daily pain scores, older age and epidural analgesia tended to steepen the slope (p=0.014 and <0.001, respectively). Preoperative use of parecoxib were also related to decreased frequency of rescue morphine medication (HR=0.4; 95% CI 0.25 to 0.65). CONCLUSIONS: Pre-emptive analgesia with parecoxib was associated with decreased baseline pain scores but had no connection with pain decreasing trends over time. Latent curve analysis provided insights into the dynamic relationships among the analgesic modalities, patient characteristics and postoperative pain trajectories.

Safety of Aspirin Use in Patients With Stroke and Small Unruptured Aneurysms.

OBJECTIVE: We initiated a multicenter, prospective cohort study to test the hypothesis that aspirin is safe for patients with ischemic cerebrovascular disease (ICVD) harboring unruptured intracranial aneurysms (UIAs) <7 mm. METHODS: This prospective, multicenter cohort study consecutively enrolled 1,866 eligible patients with ICVD harboring UIAs <7 mm in diameter from 4 hospitals between January 2016 and August 2019. Baseline and follow-up patient information, including the use of aspirin, was recorded. The primary endpoint was aneurysm rupture. RESULTS: After a total of 4,411.4 person-years, 643 (37.2%) patients continuously received aspirin treatment. Of all included patients, rupture occurred in 12 (0.7%). The incidence rate for rupture (IRR) was 0.27 (95% confidence interval [CI] 0.15-0.48) per 100 person-years. The IRRs were 0.39 (95% CI 0.21-0.72) and 0.06 (95% CI 0.010-0.45) per 100 person-years for the nonaspirin and aspirin groups, respectively. In the multivariate analysis, uncontrolled hypertension and UIAs 5 to <7 mm were associated with a high rate of aneurysm rupture, whereas aspirin use was associated with a low rate of aneurysm rupture. Compared with other groups, the high-risk group (UIAs 5 to <7 mm with concurrent uncontrolled hypertension) without aspirin had higher IRRs. CONCLUSION: Aspirin is a safe treatment for patients with concurrent small UIAs and ICVD. Patients who are not taking aspirin in the high-risk group warrant intensive surveillance. CLINICALTRIALSGOV IDENTIFIER: NCT02846259. CLASSIFICATION OF EVIDENCE: This study provides Class III evidence that for patients harboring UIAs <7 mm with ICVD, aspirin does not increase the risk of aneurysm rupture.

Aspirin and Growth of Small Unruptured Intracranial Aneurysm: Results of a Prospective Cohort Study.

BACKGROUND AND PURPOSE: The role of aspirin in unruptured intracranial aneurysm (UIA) growth remains largely unknown. We aim to identify whether aspirin is associated with a lower rate of UIA growth in patients with UIA <7 mm. METHODS: This prospective cohort study consecutively enrolled patients with UIAs <7 mm with ischemic cerebrovascular disease between January 2016 and December 2019. Baseline and follow-up patient information, including the use of aspirin and blood pressure level, were recorded. Patients were considered aspirin users if they took aspirin, including standard- and low-dose aspirin, ≥3× per week. The primary end point was aneurysm growth in any direction or an indisputable change in aneurysm shape. RESULTS: Among the 315 enrolled patients, 272 patients (86.3%) underwent imaging examinations during follow-up (mean follow-up time, 19.6±12.7 months). A total of 113 patients were continuously treated with aspirin. UIA growth occurred in 31 (11.4%) patients. In the multivariate Cox analysis, specific aneurysm locations (anterior communicating artery, posterior communicating artery, or middle cerebral artery; hazard ratio, 2.89 [95% CI, 1.22-6.88]; P=0.016) and a UIA size of 5 to <7 mm (hazard ratio, 7.61 [95% CI, 3.02-19.22]; P<0.001) were associated with a high risk of UIA growth, whereas aspirin and well-controlled blood pressure were associated with a low risk of UIA growth (hazard ratio, 0.29 [95% CI, 0.11-0.77]; P=0.013 and hazard ratio, 0.25 [95% CI, 0.10-0.66]; P=0.005, respectively). The cumulative annual growth rates were as high as 40.0 and 53.3 per 100 person-years in the high-risk patients (>1 risk factor) with and without aspirin, respectively. CONCLUSIONS: Aspirin therapy and well-controlled blood pressure are associated with a low risk of UIA growth; the incidence of UIA growth in high-risk patients in the first year is high, warranting intensive surveillance in this patient group. Registration: URL: https://www.clinicaltrials.gov. Unique identifier: NCT02846259.