scholarly journals Accuracy and Prognostic Role of NCCT-ASPECTS Depend on Time from Acute Stroke Symptom-onset for both Human and Machine-learning Based Evaluation

2021 ◽  
Author(s):  
A. Potreck ◽  
C. S. Weyland ◽  
F. Seker ◽  
U. Neuberger ◽  
C. Herweh ◽  
...  
Keyword(s):  
Machine Learning ◽  
Clinical Outcome ◽  
Ct Angiography ◽  
Symptom Onset ◽  
Learning Algorithm ◽  
Intraclass Correlation ◽  
Onset Time ◽  
Hyperacute Stroke ◽  
Favorable Clinical Outcome

Abstract Purpose We hypothesize that the detectability of early ischemic changes on non-contrast computed tomography (NCCT) is limited in hyperacute stroke for both human and machine-learning based evaluation. In short onset-time-to-imaging (OTI), the CT angiography collateral status may identify fast stroke progressors better than early ischemic changes quantified by ASPECTS. Methods In this retrospective, monocenter study, CT angiography collaterals (Tan score) and ASPECTS on acute and follow-up NCCT were evaluated by two raters. Additionally, a machine-learning algorithm evaluated the ASPECTS scale on the NCCT (e-ASPECTS). In this study 136 patients from 03/2015 to 12/2019 with occlusion of the main segment of the middle cerebral artery, with a defined symptom-onset-time and successful mechanical thrombectomy (MT) (modified treatment in cerebral infarction score mTICI = 2c or 3) were evaluated. Results Agreement between acute and follow-up ASPECTS were found to depend on OTI for both human (Intraclass correlation coefficient, ICC = 0.43 for OTI < 100 min, ICC = 0.57 for OTI 100–200 min, ICC = 0.81 for OTI ≥ 200 min) and machine-learning based ASPECTS evaluation (ICC = 0.24 for OTI < 100 min, ICC = 0.61 for OTI 100–200 min, ICC = 0.63 for OTI ≥ 200 min). The same applied to the interrater reliability. Collaterals were predictors of a favorable clinical outcome especially in hyperacute stroke with OTI < 100 min (collaterals: OR = 5.67 CI = 2.38–17.8, p < 0.001; ASPECTS: OR = 1.44, CI = 0.91–2.65, p = 0.15) while ASPECTS was in prolonged OTI ≥ 200 min (collaterals OR = 4.21,CI = 1.36–21.9, p = 0.03; ASPECTS: OR = 2.85, CI = 1.46–7.46, p = 0.01). Conclusion The accuracy and reliability of NCCT-ASPECTS are time dependent for both human and machine-learning based evaluation, indicating reduced detectability of fast stroke progressors by NCCT. In hyperacute stroke, collateral status from CT-angiography may help for a better prognosis on clinical outcome and explain the occurrence of futile recanalization.

Abstract WP406: Training and Validation of Deepmedic Machine Learning Tool for Automated Hematoma Segmentation and Volume Analysis on CT Using Multicenter Data

Stroke ◽  
2020 ◽  
Vol 51 (Suppl_1) ◽  
Author(s):  
Sean Nurmsoo ◽  
Alessandro Guida ◽  
Alex Wong ◽  
Richard I Aviv ◽  
Andrew Demchuk ◽  
...  
Keyword(s):  
Machine Learning ◽  
Validation Cohort ◽  
Learning Algorithm ◽  
Intraclass Correlation ◽  
Ct Scans ◽  
Dice Similarity Coefficient ◽  
Machine Learning Algorithm ◽  
Manual Segmentation ◽  
Training Cohort ◽  
Volume Calculation

Introduction: We sought to train and validate an automated machine learning algorithm for ICH segmentation and volume calculation using multicenter data. Methods: An open-source 3D deep machine learning algorithm “DeepMedic” was trained using manually segmented ICH from 208 CT scans (129 patients) from the multicenter PREDICT study. The algorithm was then validated with 125 manually segmented CT scans (48 patients) from the SPOTLIGHT study. Manual segmentation was performed with Quantomo semi-automated software. ABC/2 was measured for all studies by two neuroradiologists. Accuracy of DeepMedic segmentation was assessed using the Dice similarity coefficient. Analysis was stratified by presence of IVH. Intraclass correlation (ICC) with 95% confidence intervals (CI) assessed agreement between manual vs. DeepMedic segmentation volume; and manual segmentation and ABC/2 volume. Bland-Altman charts were analyzed for ABC/2 and DeepMedic vs. manual segmentation volumes. Results: DeepMedic demonstrated high segmentation accuracy in the training cohort (median Dice 0.96; IQR 0.95 - 0.97) and in the validation cohort (median Dice 0.91; IQR 0.86 - 0.94). Dice coefficients were not significantly different between patients with IVH in the training cohort; however was significantly worse in the validation cohort in patients with IVH (Wilcoxon p<0.001). Agreement was significantly better between DeepMedic and manual segmentation (PREDICT: ICC 0.99 [95%CI 0.99 -1.00]; SPOTLIGHT: ICC 0.98 [95%CI 0.97 - 0.99]) than between ABC/2 and manual segmentation (PREDICT: ICC 0.92 [95%CI 0.89 - 0.95]; SPOTLIGHT: ICC 0.95 [95%CI 0.93-0.97]). Improved accuracy of DeepMedic was demonstrated in Bland-Altman charts (Fig 1). Conclusion: ICH machine learning segmentation with DeepMedic is feasible and accurate; and demonstrates greater agreement with manual segmentation compared to ABC/2 volumes. Accuracy of the machine learning algorithm however is limited in patients with IVH.

scholarly journals Automated Prediction of Ischemic Brain Tissue Fate from Multi-Phase CT-Angiography in Patients with Acute Ischemic Stroke Using Machine Learning

2020 ◽  
Author(s):  
Wu Qiu ◽  
Hulin Kuang ◽  
Johanna Ospel ◽  
Michael D Hill ◽  
Andrew Demchuk ◽  
...  
Keyword(s):  
Machine Learning ◽  
Ischemic Stroke ◽  
Acute Ischemic Stroke ◽  
Correlation Coefficient ◽  
Ct Angiography ◽  
Tissue Perfusion ◽  
Dice Similarity Coefficient ◽  
Derivation Cohort ◽  
Multiphase Ct

Background: Multiphase CT-Angiography (mCTA) provides time variant images of the pial vasculature supplying brain in patients with acute ischemic stroke (AIS). To develop a machine learning (ML) technique to predict infarct, penumbra and tissue perfusion from mCTA source images. Methods: 284 patients with AIS were included from the PRoveIT study. All patients had non-contrast CT, mCTA and CTP imaging at baseline and follow up MRI/NCCT imaging. Of the 284 patient images, 140 patient images were randomly selected to train and validate three ML models to predict infarct, penumbra, and perfusion parameter on CTP, respectively. The remaining unseen 144 patient images independent of the derivation cohort were used to test the derived ML models. The predicted infarct, penumbra, and perfusion volume from ML models was spatially and volumetrically compared to manually contoured follow up infarct and time-dependent Tmax thresholded volume (CTP volume), using Bland-Altman plots, concordance correlation coefficient (CCC), intra-class correlation coefficient (ICC), and Dice similarity coefficient (DSC). Results: Within the test cohort, Bland-Altman plots showed that the mean difference between the mCTA predicted infarct and follow up infarct was 21.7 mL (limit of agreement (LoA): -41.0 to 84.3mL) in the 100 patients who had acute reperfusion (mTICI 2b/2c/3), and 3.4mL (LoA: -66 to 72.9mL) in the 44 patients who did not achieve reperfusion (mTICI 0/1). Amongst reperfused subjects, CCC was 0.4 [95%CI: 0.15-0.55, P<.01] and ICC 0.42 [95% CI: 0.18-0.50, P<.01]; in non-reperfused subjects CCC was 0.52 [95%CI: 0.2-0.6, P<.001] and ICC 0.6 [95% CI: 0.37-0.76, P<.001]. No difference was observed between the mCTA and CTP predicted infarct volume for the overall test cohort (P=.67). Conclusion: Multiphase CT Angiography is able to predict infarct, penumbra and tissue perfusion, comparable to CT perfusion imaging.

Predicting imminent disease progression in advanced colorectal cancer by a machine-learning algorithm.

2019 ◽  
Vol 37 (4_suppl) ◽  
pp. 645-645
Author(s):  
Yuri Kogan ◽  
Shmuel Shannon ◽  
Eldad Taub ◽  
Marina Kleiman ◽  
Moran Elishmereni ◽  
...  
Keyword(s):  
Colorectal Cancer ◽  
Machine Learning ◽  
Disease Progression ◽  
Tumor Markers ◽  
Learning Algorithm ◽  
Diagnostic Algorithm ◽  
Predictive Values ◽  
New Approach ◽  
Single Marker

645 Background: In advanced cancers, predicting disease progression just before its clinical manifestation enables an earlier switch to the next treatment line, preventing deterioration in the patient's state and potentially improving survival. Yet, given the ambiguity of current tumor markers in alerting to progression, physicians are unable to forecast this key event. We developed a diagnostic algorithm for announcing an approaching disease progression in late-stage colorectal cancer (CRC) patients by processing continuous carcinoembryonic antigen (CEA) input. Methods: Longitudinally measured CEA data of advanced CRC patients treated by standard 1st line chemotherapies, collected from 2 clinical trials (projectdatasphere.org), served for algorithm development by machine-learning and training assisted by receiver-operating-characteristic (ROC) analysis and correlation tests. Performance was validated by cross-validation techniques. Results: CEA and response evaluations of 489 CRC patients (median follow-up time: 168 days) were processed by the algorithm, predicting disease progression with 57% sensitivity (100/175 progression events) and 88% specificity (21/175 false positives). Positive and negative predictive values, accuracy and Cohen’s kappa were 64%, 84%, 79% and 0.46, respectively. The algorithm’s predictive power was superior to that of standard statistical analyses of these CEA data (e.g., ROC). Conclusions: Our study offers a new approach to using tumor markers as prognosticators. The algorithm-amplified ability of CEA to predict progression in CRC complements our recent findings in lung cancer, where integration of CEA and 4 other markers provided 66% sensitivity in predicting progression, surpassing the low capacity of each separate marker. Conceivably, future algorithm-integration of multiple markers in CRC may also exceed the limited signal of a single marker. Clinical use of our algorithm, amplifying weak marker signals of imminent progression, should allow physicians to reliably harness tumor markers for improving treatment and potentially extending survival in cancer patients.

Stent assisted coiling versus non-stent assisted coiling for the management of ruptured intracranial aneurysms: a meta-analysis and systematic review

2019 ◽  
Vol 11 (5) ◽  
pp. 489-496 ◽  
Author(s):  
Xiaoxi Zhang ◽  
Qiao Zuo ◽  
Haishuang Tang ◽  
Gaici Xue ◽  
Pengfei Yang ◽  
...  
Keyword(s):  
Clinical Outcome ◽  
Recurrence Rate ◽  
Complication Rate ◽  
Intracranial Aneurysms ◽  
Meta Analysis ◽  
Lower Recurrence Rate ◽  
Significant Difference ◽  
Favorable Clinical Outcome ◽  
Ruptured Intracranial Aneurysms

PurposeTo compare the safety and efficiency of stent assisted coiling (SAC) with non-SAC for the management of ruptured intracranial aneurysms.MethodsA meta-analysis that compared SAC with coiling alone and balloon assisted coiling was conducted by database searching. The primary outcomes of this study were immediate occlusion and progressive thrombosis rate, overall perioperative complication rate, and angiographic recurrence. Secondary outcomes included mortality at discharge, hemorrhagic and ischemic complications, and favorable clinical outcome at discharge and at follow-up.ResultsEight retrospective cohort studies with 1408 ruptured intracranial aneurysms (SAC=499; non-SAC=909) were included. The SAC group tended to show a lower immediate complete occlusion rate than the non-SAC group (54.3% vs 64.2%; RR 0.90; 95% CI 0.83 to 0.99; I2=17.4%) and achieved a significantly higher progressive complete rate at follow-up (73.4% vs 61.0%; RR 1.30; 95% CI 1.16 to 1.46; I2=40.5%) and a lower recurrence rate (4.8% vs 16.6%; RR 0.28; 95% CI 0.16 to 0.50; I2=0.0%). With respect to safety concerns, overall perioperative complications in the SAC group were significantly higher (20.2% vs 13.1%; RR 1.70; 95% CI 1.36 to 2.11; I2=0.0%). However, no significant difference was found for mortality rate at discharge (6.3% vs 6.2%; RR 1.29; 95% CI 0.86 to 1.94; I2=0.0%), or favorable clinical outcome rate at discharge (73.4% vs 74.2%; RR 0.95; 95% CI 0.88 to 1.02; I2=12.1%) and at follow-up (85.6% vs 87.9%; RR 0.98; 95% CI 0.93 to 1.02; I2=0.0%; P=0.338).ConclusionsSAC has a lower recurrence rate than non-SAC. Nevertheless, further validation by well designed prospective studies is warranted for determining whether stents improve angiographic outcome without an increased complication rate or unfavorable clinical outcome.

scholarly journals Clinical presentation of COVID-19 – a model derived by a machine learning algorithm

2021 ◽  
Vol 18 (1) ◽  
pp. 3-8 ◽  
Author(s):  
Malik Yousef ◽  
Louise C. Showe ◽  
Izhar Ben Shlomo
Keyword(s):  
Machine Learning ◽  
Clinical Presentation ◽  
Learning Algorithm ◽  
Area Under The Curve ◽  
Risk Groups ◽  
Verbal Description ◽  
Machine Learning Algorithm ◽  
Proper Selection ◽  
Selection Of

Abstract COVID-19 pandemic has flooded all triage stations, making it difficult to carefully select those most likely infected. Data on total patients tested, infected, and hospitalized is fragmentary making it difficult to easily select those most likely to be infected. The Israeli Ministry of Health made public its registry of immediate clinical data and the respective status of infected/not infected for all viral DNA tests performed up to Apr. 18th, 2020 including almost 120,000 tests. We used a machine-learning algorithm to find out which immediate clinical elements mattered the most in identifying the true status of the tested persons including age or gender matter, to enable future better allocation of surveillance policy for those belonging to high-risk groups. In addition to the analyses applied on the first batch of the available data (Apr. 11th), we further tested the algorithm on the independent second batch (Apr. 12th to 18th). Fever, cough and headache were the most diagnostic, differing in degree of importance in different subgroups. Higher percentage of men were found positive (9.3 vs. 7.3%), but gender did not matter for the clinical presentation. The prediction power of the model was high, with accuracy of 0.84 and area under the curve 0.92. We provide a hand-held short checklist with verbal description of importance for the leading symptoms, which should expedite the triage and enable proper selection of people for further follow-up.

Clinical outcome and imaging follow-up in acute stroke patients with normal perfusion CT and normal CT angiography

2010 ◽  
Vol 53 (2) ◽  
pp. 79-88 ◽  
Author(s):  
Bernd Eckert ◽  
Tobias Küsel ◽  
Andreas Leppien ◽  
Peter Michels ◽  
Axel Müller-Jensen ◽  
...  
Keyword(s):  
Clinical Outcome ◽  
Acute Stroke ◽  
Ct Angiography ◽  
Perfusion Ct ◽  
Stroke Patients

Determination of Stent Frame Displacement After Endovascular Aneurysm Sealing

2017 ◽  
Vol 25 (1) ◽  
pp. 52-61 ◽  
Author(s):  
Ruben van Veen ◽  
Kim van Noort ◽  
Richte C. L. Schuurmann ◽  
Jan Wille ◽  
Cornelis H. Slump ◽  
...  
Keyword(s):  
Ct Angiography ◽  
Lateral Displacement ◽  
Intraclass Correlation ◽  
Ct Scans ◽  
Anatomical Landmarks ◽  
3 Dimensional ◽  
Type Ia ◽  
Conventional Ct

Purpose: To describe and validate a new methodology for visualizing and quantifying 3-dimensional (3D) displacement of the stent frames of the Nellix endosystem after endovascular aneurysm sealing (EVAS). Methods: The 3D positions of the stent frames were registered to 5 fixed anatomical landmarks on the post-EVAS computed tomography (CT) scans, facilitating comparison of the position and shape of the stent frames between consecutive follow-up scans. Displacement of the proximal and distal ends of the stent frames, the entire stent frame trajectories, as well as changes in distance between the stent frames were determined for 6 patients with >5-mm displacement and 6 patients with <5-mm displacement at 1-year follow-up. The measurements were performed by 2 independent observers; the intraclass correlation coefficient (ICC) was used to determine interobserver variability. Results: Three types of displacement were identified: displacement of the proximal and/or distal end of the stent frames, lateral displacement of one or both stent frames, and stent frame buckling. The ICC ranged from good (0.750) to excellent (0.958). No endoleak or migration was detected in the 12 patients on conventional CT angiography at 1 year. However, of the 6 patients with >5-mm displacement on the 1-year CT as determined by the new methodology, 2 went on to develop a type Ia endoleak in longer follow-up, and displacement progressed to >15 mm for 2 other patients. No endoleak or progressive displacement was appreciated for the patients with <5-mm displacement. Conclusion: The sac anchoring principle of the Nellix endosystem may result in several types of displacement that have not been observed during surveillance of regular endovascular aneurysm repairs. The presented methodology allows precise 3D determination of the Nellix endosystems and can detect subtle displacement better than standard CT angiography. Displacement >5 mm on the 1-year CT scans reconstructed with the new methodology may forecast impaired sealing and anchoring of the Nellix endosystem.

scholarly journals Stent-Assisted Coil Embolization of Wide-Necked Intracranial Aneurysms Using a Semi-Deployment Technique: Angiographic and Clinical Outcomes in 31 Consecutive Patients

2010 ◽  
Vol 16 (4) ◽  
pp. 385-393 ◽  
Author(s):  
X. Gao ◽  
G. Liang ◽  
Z. Li ◽  
H. Qu ◽  
X. Wei
Keyword(s):  
Clinical Outcome ◽  
Intracranial Aneurysms ◽  
Stent Migration ◽  
Delayed Complication ◽  
Favorable Clinical Outcome ◽  
Incomplete Occlusion ◽  
Wide Neck ◽  
Stent Stenosis ◽  
Related Mortality

We describe a modified stent-assisted coiling technique, named the semi-deployment technique, in the endovascular treatment of wide-neck aneurysms. Thirty-one consecutive patients with 31 wide-necked or fusiform intracranial aneurysms were treated with the semi-deployment technique. The technical feasibility of the procedure, procedure-related complications, angiographic results, clinical outcome and follow-up angiography were evaluated. In every case, the semi-deployment technique was successfully deployed. Immediate angiography demonstrated complete occlusion in 24 cases (77.4%), neck remnant in four cases (12.9%), and incomplete occlusion in three cases (9.7%). Procedural-related morbidity occurred in one patient (3.2%) but no procedural-related mortality. A favorable clinical outcome (Modified Ran-kin Scale score 0–2) was observed in 90.3% of the patients (average follow-up time, 23.1 months). No rehemorrhage of treated aneurysms occurred. Angiography follow-up was obtained in 22 cases (71.0 %). Three aneurysms (13.6 % of the follow-up angiograms) demonstrated recanalization. No delayed coil or stent migration was found. One patient had in-stent stenosis as a delayed complication. We found that the semi-deployment technique was helpful in the treatment of wide-neck aneurysms.

Sign in / Sign up

Export Citation Format

Share Document