| Bagher-Ebadian, Hassan; Jafari-Khouzani, Kourosh; Mitsias, Panayiotis D.; Lu, Mei; Soltanian-Zadeh, Hamid; Chopp, Michael; Ewing, James R. |
Predicting final extent of ischemic infarction using artificial neural network analysis of multi-parametric MRI in patients with stroke. |
2011 |
final infarct |
x |
✓ |
ns |
12 |
✓ |
✓ |
x |
x |
x |
x |
multilayer perceptions |
✓ |
x |
x |
x |
T1, T2, DWI, Proton-density weighted imaging |
|
Per voxel: 4 |
ns |
11 |
ns |
ns |
not reported |
leave-1-out |
none |
0.89 |
0.85 |
0.84 |
x |
x |
✓ |
x |
x |
x |
| Dharmasaroja, Permphan; Dharmasaroja, Pornpatr A. |
Prediction of intracerebral hemorrhage following thrombolytic therapy for acute ischemic stroke using multiple artificial neural networks. |
2012 |
sICH |
x |
✓ |
tPA |
194 |
x |
x |
x |
x |
x |
✓ |
Probabilistic Neural Networks |
x |
✓ |
✓ |
x |
Gender, Age, HTN, DM, Hyperlipidemia, Coronary artery disease, Previous ischemic stroke, Previous ICH, History of TIA, Peripheral artery disease, AFib, Smoking, Valvular heart disease, Alcohol use, Statin use, Antiplatelet use, Systolic BP, Diastolic BP, Platelet counts, Prothrombin time, INR, Blood sugar, LDL-C, Stroke subtype, Stroke location, CT findings, ASPECTS, NIHSS, Onset-to-treatment time |
|
29 |
90 |
ns |
10 |
ns |
ratio |
cross-validation |
none |
0.7875 |
0.3529 |
1 |
0.5217 |
x |
✓ |
x |
x |
x |
| Ho, King C.hung; Speier, William; El-Saden, Suzie; Liebeskind, David S.; Saver, Jeffery L.; Bui, Alex A. T; Arnold, Corey W. |
Predicting discharge mortality after acute ischemic stroke using balanced data |
2014 |
discharge mortality |
x |
✓ |
ns |
229 |
✓ |
✓ |
x |
x |
x |
x |
svm |
x |
x |
✓ |
x |
pre-NIHSS, age, platelet count, serum glucose, congestive heart failure, myocardial infarction |
|
6 |
90 |
ns |
ns |
39 |
absolute n |
cross-validation |
none |
x |
x |
x |
0.5 |
x |
✓ |
x |
x |
x |
| Bentley, Paul; Ganesalingam, Jeban; Carlton Jones, Anoma Lalani; Mahady, Kate; Epton, Sarah; Rinne, Paul; Sharma, Pankaj; Halse, Omid; Mehta, Amrish; Rueckert, Daniel |
Prediction of stroke thrombolysis outcome using CT brain machine learning |
2014 |
sICH |
x |
✓ |
tPA |
116 |
✓ |
✓ |
x |
x |
x |
x |
svm |
x |
✓ |
✓ |
x |
NCCT voxels, NIHSS |
|
per voxel: 1 + per patient: 1 |
ns |
106 |
ns |
ns |
not reported |
cross-validation |
none |
0.744 |
x |
x |
x |
x |
x |
x |
✓ |
x |
| Seiffge, D.J.; Gensicke, H.; Peters, N.; Bonati, L.H.; Kejda-Scharler, J.; Tranka, C.; Lyrer, P.A.; Engelter, S.T.; Karagiannis, A.; Strbian, D.; Kotisaari, K.; Leppa, M.; Tatlisumak, T.; Ginsbach, P. |
Simple variables predict miserable outcome after intravenous thrombolysis |
2014 |
90 day mrs |
x |
✓ |
tPA |
1984 |
✓ |
✓ |
x |
x |
x |
x |
logistic regression |
x |
✓ |
✓ |
x |
age, independence before stroke, normal Glasgow coma verbal score, able to lift arms and able to walk |
|
4 |
ns |
1346 |
ns |
638 |
absolute n |
none |
none |
0.786 |
x |
x |
x |
x |
x |
x |
✓ |
x |
| Ho, King Chung; Speier, William; El-Saden, Suzie; Arnold, Corey W. |
Classifying Acute Ischemic Stroke Onset Time using Deep Imaging Features. |
2017 |
onset time |
✓ |
x |
MCA |
105 |
✓ |
✓ |
x |
x |
x |
x |
stepwise multilinear regression |
✓ |
x |
x |
x |
DWI, ADC, FLAIR, PWI |
avg intensity in ROI of DWI, ADC, FLAIR, CBF, CBV, TTP, MTT + deep AE features |
Per voxel: 16 |
ns |
ns |
ns |
ns |
not reported |
nested cross-validation |
none |
0.683 |
0.687 |
0.591 |
0.765 |
x |
x |
✓ |
x |
✓ |
| McKinley, Richard; Hani, Levin; Gralla, Jan; El-Koussy, M.; Mattmann, Kaspar; Wiest, Roland; Bauer, S.; Reyes, Mauricio; Arnold, M.; Fischer, U.; Jung, S. |
Fully automated stroke tissue estimation using random forest classifiers (FASTER) |
2017 |
final infarct w or w/o reperfusion |
x |
✓ |
MCA |
80 |
✓ |
✓ |
x |
x |
x |
x |
random forest |
✓ |
x |
x |
x |
DWI, ADC, GRE, Standard dynamic susceptibility contrast-enhanced perfusion MRI, T2, Gd-MRA, T1C, TOF angiography |
ADC, Tmax, CBV, CBF, TTP, T1c, T2 |
Per voxel: 274 |
ns |
25 |
ns |
55 |
absolute n |
none |
none |
0.94/0.96 |
x |
x |
x |
x |
x |
x |
✓ |
x |
| Yu, Yannan; Guo, Danfeng; Liebeskind, David; Scalzo, Fabien; Lou, Min |
Prediction of hemorrhagic transformation severity in acute stroke from source perfusion MRI |
2018 |
sICH |
x |
✓ |
< 6h |
155 |
✓ |
✓ |
x |
x |
x |
x |
Kernel spectral regression |
✓ |
x |
x |
x |
PWI, DWI |
Per voxel: PWI, AIF, DWI |
ns |
ns |
ns |
ns |
ns |
not reported |
nested cross-validation |
none |
0.837 |
x |
x |
0.717 |
x |
✓ |
x |
x |
x |
| Livne, Michelle; Fiebach, Jochen B.; Sobesky, Jan; Boldsen, Jens K.; Mikkelsen, Irene K.; Mouridsen, Kim |
Boosted tree model reforms multimodal magnetic resonance imaging infarct prediction in acute stroke |
2018 |
final infarct |
x |
✓ |
ns |
170 |
✓ |
✓ |
x |
x |
x |
x |
xgboost |
✓ |
x |
x |
x |
DSC-MRI for perfusion-weighted imaging, T2-FLAIR, and DWI |
DWI, T2-FLAIR, pMTT, oTmax, pTmax, oCBF, pCBF, CTH, TTP, oMTT, RTH, pCBV |
Per voxel: 12 |
80 |
ns |
ns |
ns |
not reported |
cross-validation |
none |
0.92 |
x |
x |
x |
x |
✓ |
x |
x |
x |
| Nielsen, Anne; Hansen, Mikkel Bo; Tietze, Anna; Mouridsen, Kim |
Prediction of Tissue Outcome and Assessment of Treatment Effect in Acute Ischemic Stroke Using Deep Learning. |
2018 |
final infarct |
x |
✓ |
tPA |
222 |
✓ |
x |
x |
x |
x |
x |
cnn |
✓ |
x |
x |
x |
PWI, T2-FLAIR, DWI |
mean capillary transit time, cerebral blood volume, cerebral blood flow, cerebral metabolism of oxygen, relative transit time heterogeneity, delay, TRACE DWI, ADC, and T2-FLAIR |
64x64x9 |
85 |
158 |
15 |
29 |
ratio, absolute n |
none |
none |
0.88 |
x |
x |
x |
x |
x |
x |
✓ |
x |
| Giacalone, Mathilde; Rasti, Pejman; Debs, Noelie; Frindel, Carole; Cho, Tae-Hee; Grenier, Emmanuel; Rousseau, David |
Local spatio-temporal encoding of raw perfusion MRI for the prediction of final lesion in stroke. |
2018 |
final infarct |
x |
✓ |
anterior |
4 |
x |
x |
x |
x |
x |
✓ |
svm |
✓ |
x |
x |
x |
PWI |
DSC-PWI |
(60/3)x59 = 1180 |
99 |
ns |
ns |
ns |
not reported |
cross-validation |
none |
x |
0.96 |
0.947 |
x |
x |
✓ |
x |
x |
x |
| Bouteldja, Nassim; Heinrich, Mattias P.; Lucas, Christian; Kemmling, Andre; Aulmann, Linda F.; Mamlouk, Amir Madany |
Learning to predict ischemic stroke growth on acute CT perfusion data by interpolating low -dimensional shape representations |
2018 |
final infarct |
x |
✓ |
thrombectomy |
29 |
✓ |
✓ |
x |
x |
x |
x |
cnn |
x |
✓ |
x |
x |
CT perfusion |
CBV, Time To Drain (TTD) maps |
64x64x28x2 |
80 |
ns |
20 |
ns |
ratio |
cross-validation |
none |
x |
x |
x |
0.43 |
x |
✓ |
x |
x |
x |
| Van Os, Hendrikus J.A.; Kruyt, Nyika D.; Wermer, Marieke J.H.; Hilbert, Adam; Marquering, Henk A.; Ramos, Lucas A.; Olabarriaga, Silvia D.; Zwinderman, Koos H.; Van Leeuwen, Matthijs; Van Walderveen, Marianne A.A.; Dippel, Diederik W.J.; Steyerberg, Ewout W.; Lingsma, Hester F.; Venema, Esmee; Van Der Schaaf, Irene C.; Schonewille, Wouter J.; Majoie, Charles B.L.M. |
Predicting outcome of endovascular treatment for acute ischemic stroke: Potential value of machine learning algorithms |
2018 |
successful recan, 90 day mrs |
x |
✓ |
thrombectomy, anterior |
1383 |
✓ |
✓ |
x |
x |
x |
x |
random forest |
x |
✓ |
✓ |
x |
NIHSS, GCS, medical history (TIA, ischemic stroke, intracranial hemorrhage, subarachnoid hemorrhage, myocardial infarction, peripheral artery disease, DM, HTN, hypercholesterolemia), smoking, laboratory tests (blood glucose, INR, creatinine, thrombocyte count, CRP), blood pressure, medication (thrombocyte aggregation inhibitors, oral anticoagulant drugs, anti-hypertensive drugs, statins), modified Rankin Score (mRS) before stroke onset, administration of intravenous tPA (yes/no), stroke onset to groin time, transfer from another hospital, admitted during weekend or off hours clot burden score, clot location, collaterals, and presence of intracranial atherosclerosis |
|
53 |
80 |
ns |
20 |
ns |
ratio |
nested cross-validation |
grid |
0.55/0.79 |
x |
x |
x |
x |
✓ |
x |
x |
x |
| Monteiro, Miguel; Fonseca, Ana Catarina; Freitas, Ana Teresa; Pinho E Melo, Teresa; Francisco, Alexandre P.; Ferro, Jose M.; Oliveira, Arlindo L. |
Using Machine Learning to Improve the Prediction of Functional Outcome in Ischemic Stroke Patients. |
2018 |
90 day mrs |
x |
✓ |
tPA |
425 |
x |
x |
x |
x |
x |
✓ |
random forest |
x |
x |
✓ |
x |
demographic information, past history and risk factors, the time between stroke onset, arrival at the hospital and the start of treatment, NIHSS (discriminated by field, not just the final result), test results, type of treatment |
|
49 |
90 |
ns |
ns |
ns |
not reported |
cross-validation |
grid |
0.808 |
x |
x |
x |
x |
x |
x |
✓ |
x |
| Ho, King Chung; Speier, William; Zhang, Haoyue; Scalzo, Fabien; El-Saden, Suzie; Arnold, Corey W. |
A Machine Learning Approach for Classifying Ischemic Stroke Onset Time From Imaging. |
2019 |
onset time |
✓ |
x |
MCA |
131 |
✓ |
✓ |
x |
x |
x |
x |
logistic regression |
✓ |
x |
x |
x |
DWI, ADC, FLAIR, PWI perfusion parameters |
AE deep features, AIF |
131 |
90 |
ns |
10 |
ns |
ratio |
nested cross-validation |
none |
0.765 |
0.788 |
0.609 |
0.788 |
x |
x |
✓ |
x |
✓ |
| Tozlu, Ceren; Maucort-Boulch, Delphine; Ozenne, Brice; Cho, Tae-Hee; Nighoghossian, Norbert; Derex, Laurent; Hermier, Marc; Mikkelsen, Irene Klaerke; Berthezene, Yves; Pedraza, Salvador; Fiehler, Jens; Ostergaard, Leif; Baron, Jean-Claude |
Comparison of classification methods for tissue outcome after ischaemic stroke |
2019 |
final infarct |
x |
✓ |
anterior, <6/12h, tPA/conservative treatment |
55 |
✓ |
✓ |
x |
x |
x |
x |
Adaptive boosting |
✓ |
x |
x |
x |
T2 FLAIR, DWI, ADC, PWI, CBV, CBF, TTP, MTT, TMAX |
T2 FLAIR, DWI, ADC, CBV, CBF, TTP, MTT, TMAX |
Per voxel: 8 |
ns |
54 |
ns |
1 |
absolute n |
nested cross-validation |
grid |
0.88/0.29 |
0.49 |
0.97 |
0.28 |
x |
✓ |
x |
x |
x |
| Ho, King Chung; Sarma, Karthik V.; Scalzo, Fabien; Speier, William; El-Saden, Suzie; Arnold, Corey |
Predicting ischemic stroke tissue fate using a deep convolutional neural network on source magnetic resonance perfusion images |
2019 |
final infarct |
x |
✓ |
MCA |
48 |
✓ |
✓ |
x |
x |
x |
x |
cnn |
✓ |
x |
x |
x |
FLAIR, PWI |
PWI source |
2x25x25x64 |
90 |
ns |
ns |
ns |
not reported |
nested cross-validation |
none |
0.871 |
0.799 |
x |
0.347 |
x |
✓ |
x |
x |
x |
| Kasasbeh, Aimen S.; Christensen, Soren; Albers, Gregory W.; Lansberg, Maarten G.; Parsons, Mark W.; Campbell, Bruce |
Artificial Neural Network Computer Tomography Perfusion Prediction of Ischemic Core |
2019 |
Infarct core mapping |
✓ |
x |
anterior, <8h |
128 |
✓ |
x |
x |
x |
x |
x |
multilayer perceptions |
x |
✓ |
✓ |
x |
CTP parameters (rCBF, CBV, MTT, and Tmax), clinical parameters available at time of presentation (NIHSS, age, sex, and time from symptom onset to CTP) |
CBF, CBV, TTP, Tmax + NIHSS, age, sex, and time from symptom onset to CTP |
ns |
60 |
ns |
20 |
ns |
ratio |
cross-validation |
none |
0.87 |
0.91 |
0.66 |
0.43 |
x |
✓ |
x |
x |
x |
| Lopez-Rivera, Victor; Lee, Songmi; Sheth, Sunil A.; Savitz, Sean I.; Barman, Arko; Grotta, James C.; Yoo, Albert J.; Inam, Mehmet E.; Giancardo, Luca |
Machine Learning-Enabled Automated Determination of Acute Ischemic Core From Computed Tomography Angiography |
2019 |
LVO + Infarct Core mapping |
✓ |
x |
anterior |
224 |
✓ |
✓ |
✓ |
x |
x |
x |
cnn |
x |
✓ |
x |
x |
CTA |
|
ns |
90 |
ns |
ns |
ns |
not reported |
cross-validation |
none |
0.88/0.9/0.844 |
x |
x |
x |
x |
✓ |
x |
x |
x |
| Qiu, W.; Kuang, H.; Nair, J.; Assis, Z.; Najm, M.; McDougall, C.; McDougall, B.; Chung, K.; Wilson, A. T.; Goyal, M.; Hill, M. D.; Demchuk, A. M.; Menon, B. K. |
Radiomics-Based Intracranial Thrombus Features on CT and CTA Predict Recanalization with Intravenous Alteplase in Patients with Acute Ischemic Stroke. |
2019 |
succesful lys |
x |
✓ |
tPA, ICA/M1 MCA |
67 |
x |
x |
x |
x |
x |
x |
svm |
x |
✓ |
x |
x |
nCCT and CTA and manually extracted thrombus |
NCCT, CTA, and radiomics change features |
12 |
80 |
ns |
ns |
ns |
not reported |
cross-validation |
none |
0.85 |
x |
x |
x |
x |
✓ |
x |
x |
x |
| Hilbert, A.; Barros, R.S.; Strijkers, G.J.; Ramos, L.A.; Olabarriaga, S.D.; Zwinderman, A.H.; van Os, H.J.A.; Wermer, M.J.H.; van der Schaaf, I.; Dippel, D.; Roos, Y.B.W.E.M.; van Zwam, W.H.; Yoo, A.J.; Tolhuisen, M.L.; Emmer, B.J.; Majoie, C.B.L.M.; Marquering, H.A.; Lycklama a Nijeholt, G.J. |
Data-efficient deep learning of radiological image data for outcome prediction after endovascular treatment of patients with acute ischemic stroke |
2019 |
successful recan, 90 day mrs |
x |
✓ |
thrombectomy |
1301 |
✓ |
✓ |
x |
x |
x |
x |
cnn |
x |
✓ |
x |
x |
CTA |
Maximum Intensity Projections |
368x432 |
75 |
ns |
ns |
ns |
not reported |
cross-validation |
none |
0.71/0.65 |
x |
x |
x |
x |
✓ |
x |
x |
x |
| Heo, JoonNyung; Yoon, Jihoon G.; Park, Hyungjong; Kim, Young Dae; Nam, Hyo Suk; Heo, Ji Hoe |
Machine Learning-Based Model for Prediction of Outcomes in Acute Stroke. |
2019 |
90 day mrs |
x |
✓ |
no recanalization therapy (not specified) |
2604 |
✓ |
✓ |
x |
x |
x |
x |
multilayer perceptions |
x |
x |
✓ |
x |
Age, Sex, Smoking status, Time from onset to admission, individual NIHSS, TOAST classification, Systolic BP, Diastolic BP, Pre-Stroke mRS, HTN, Diabetes, Hypercholesterolemia, Metabolic syndrome, AFib, Myocardial infarction, Chronic heart failure, Cancer, Previous stroke, Statin, Antiplatelet, Anticoagulation, Hemoglobin, White blood cell count, Platelet count, Prothrombin time, Glucose level |
|
38 |
67 |
1744 |
ns |
ns |
not reported |
single split |
none |
0.888 |
0.367 |
0.984 |
x |
x |
x |
x |
✓ |
x |
| Winder, Anthony J.; Siemonsen, Susanne; Flottmann, Fabian; Fiehler, Jens; Thomalla, Gotz; Forkert, Nils D. |
Technical considerations of multi-parametric tissue outcome prediction methods in acute ischemic stroke patients |
2019 |
final infarct |
x |
✓ |
thrombectomy, tPA, distal ICA/M1 of MCA |
100 |
✓ |
✓ |
x |
x |
x |
x |
random forest |
✓ |
x |
✓ |
x |
PWI, DWI, Clinical: NIHSS, age, sex, time from symptom onset |
ADC, distance to ischemic core, tissue type, anatomical location, CBV, MTT, Tmax, CBF, NIHSS, age, sex, time from symptom onset |
Per voxel: 8 + per patient: 4 |
ns |
99 |
ns |
ns |
not reported |
leave-1-out |
none |
x |
0.544 |
0.997 |
0.464 |
x |
✓ |
x |
x |
x |
| Xie, Yuan; Jiang, Bin; Li, Ying; Zhu, Guangming; Wintermark, Max; Zaharchuk, Greg; Gong, Enhao; Michel, Patrik |
Use of gradient boosting machine learning to predict patient outcome in acute ischemic stroke on the basis of imaging, demographic, and clinical information |
2019 |
90 day mrs |
x |
✓ |
anterior |
512 |
x |
x |
x |
x |
x |
✓ |
Gradient Boosting Machines |
x |
✓ |
✓ |
x |
Sex, Age, Total ASPECTS, Specific ASPECTS, Collaterals status, Hyperdense MCA sign present, Infarct volume on CTP, penumbra volume on CTP, Side of stroke, Occluded vessels, Baseline TIMI score, NASCET degree of stenosis (left/right), NIHSS baseline |
|
23 |
80 |
ns |
ns |
ns |
not reported |
cross-validation |
none |
0.748 |
0.655 |
0.692 |
x |
x |
✓ |
x |
x |
x |
| Alawieh, Ali; Zaraket, Fadi; Alawieh, Mohamed Baker; Chatterjee, Arindam Rano; Spiotta, Alejandro |
Using machine learning to optimize selection of elderly patients for endovascular thrombectomy. |
2019 |
90 day mrs |
x |
✓ |
thrombectomy, elderly |
146 |
✓ |
✓ |
✓ |
x |
x |
x |
M5P - regression decision tree |
x |
✓ |
✓ |
x |
age, gender, race, pre-stroke modified Rankin Scale (mRS) score, presence of diabetes, hyperlipidemia, HTN, AFib, ASPECT, ivtPA |
|
10 |
ns |
110 |
ns |
36 |
absolute n |
cross-validation |
none |
x |
x |
x |
x |
✓ |
✓ |
x |
x |
x |
| Smith, Wade S.; Keenan, Kevin J.; Lovoi, Paul A. |
A Unique Signature of Cardiac-Induced Cranial Forces During Acute Large Vessel Stroke and Development of a Predictive Model |
2020 |
LVO |
✓ |
x |
ns |
42 |
✓ |
✓ |
x |
x |
x |
x |
Unknown |
x |
x |
x |
✓ |
headpulse from cranial accelerometer, electrocardiogram outputs |
measures of trace-by-trace variation from the ECG-signal-averaged mean |
156 |
80 |
ns |
ns |
ns |
not reported |
cross-validation |
none |
0.79 |
0.737 |
0.87 |
x |
x |
✓ |
x |
x |
x |
| Chan, Lung; Hong, Chien-Tai; Chung, Chen-Chih; Bamodu, Oluwaseun Adebayo; Chiu, Hung-Wen |
Artificial neural network based prediction of postthrombolysis intracerebral hemorrhage and death |
2020 |
sICH, 90-day-mortality |
x |
✓ |
tPA |
331 |
✓ |
✓ |
x |
x |
x |
x |
multilayer perceptions |
x |
x |
✓ |
x |
Diastolic BP, LDL, Hyperlipidemia, AFib, Heart disease, Age, tPA total dose (mg), Baseline NIHSS, DM, Ischemic heart disease |
pre-treatment parameters |
5/6 |
80 |
326/285 |
ns |
ns |
not reported |
cross-validation |
none |
0.941/0.976 |
0.857/0.944 |
0.925/0.955 |
x |
x |
x |
x |
✓ |
x |
| Chiu, Hung-Wen; Chung, Chen-Chih; Chen, You-Chia; Hong, Chien-Tai; Hu, Chaur-Jong; Hu, Han-Hwa; Chan, Lung; Chi, Nai-Fang |
Artificial neural network-based analysis of the safetand efficacy of thrombolysis for ischemic stroke in older adults in Taiwan |
2020 |
successful lys, 90 day mrs |
x |
✓ |
elderly, <3h |
80 |
✓ |
✓ |
x |
x |
x |
x |
multilayer perceptions |
x |
x |
✓ |
x |
age, sex, history of HTN, history of T2DM, AF status, history of previous stroke, initial NIHSS score, onset-to-hospital time, tPA received |
|
9 |
80 |
ns |
ns |
ns |
not reported |
cross-validation |
none |
0.974 |
0.667 |
0.875 |
x |
x |
✓ |
x |
x |
x |
| You, Jia; Yu, Philip L. H.; Tsang, Anderson C. O.; Leung, Gilberto K. K.; Tsui, Eva L. H.; Woo, Pauline P. S.; Lui, Carrie S. M. |
Automated Hierarchy Evaluation System of Large Vessel Occlusion in Acute Ischemia Stroke |
2020 |
LVO |
✓ |
x |
ns |
300 |
✓ |
✓ |
x |
x |
x |
x |
xgboost |
x |
✓ |
✓ |
x |
demographic information, clinical symptoms, vital signs, imaging features obtained from NCCT scans extracted by U-Net |
clinical data, NCCT features |
1049 |
ns |
200 |
ns |
100 |
absolute n |
cross-validation |
none |
0.847 |
0.953 |
0.684 |
0.804 |
x |
✓ |
x |
x |
x |
| Qiu, W.; Kuang, H.; Ospel, J.; Hill, M. D.; Demchuk, A.; Goyal, M.; Menon, B. |
Automated prediction of ischemic brain tissue fate from multi-phase CT-Angiography in patients with acute ischemic stroke using machine learning |
2020 |
final infarct |
x |
✓ |
ns |
284 |
✓ |
✓ |
x |
x |
x |
x |
random forest |
x |
✓ |
x |
x |
multiphase CT angiography |
Per voxel: (1) average and standard deviation of HUs across 3-phase CTA images; (2) coefficient of variance of HUs in 3-phase CTA images; (3) changing slopes of HUs between any two phases; (4) peak of |
6 |
ns |
140 |
ns |
144 |
absolute n |
cross-validation |
none |
x |
x |
x |
0.225 |
x |
x |
x |
✓ |
x |
| Stib, Matthew T.; Yao, Anthony D.; Boxerman, Jerrold L.; Baird, Grayson L.; Vasquez, Justin; Dong, Mary P.; Kim, Yun Ho; Subzwari, Sumera S.; Triedman, Harold J.; Wang, Amy; Wang, Hsin-Lei Charlene; Eickhoff, Carsten; Cetintemel, Ugur; Jayaraman, Mahesh; McTaggart, Ryan A. |
Detecting large vessel occlusion at multiphase CT angiography by using a deep convolutional neural network |
2020 |
LVO |
✓ |
x |
ns |
540 |
✓ |
✓ |
x |
x |
x |
x |
cnn |
x |
✓ |
x |
x |
multiphase CT angiography |
3 phase MIP |
ns |
80 |
ns |
10 |
ns |
ratio |
single split |
none |
0.89 |
1 |
0.77 |
x |
x |
✓ |
x |
x |
x |
| Meijs, Midas; Meijer, Frederick J.A.; Prokop, Mathias; Ginneken, Bram van; Manniesing, Rashindra |
Image-level detection of arterial occlusions in 4D-CTA of acute stroke patients using deep learning |
2020 |
LVO |
✓ |
x |
anterior |
584 |
✓ |
✓ |
x |
x |
x |
x |
cnn |
x |
✓ |
x |
x |
NCCT, CTA, 4D-CTA, derived perfusion maps |
3D normalized TTS map |
68x68x68 |
ns |
184 |
ns |
279 |
absolute n |
single split |
none |
0.98 |
0.95 |
0.92 |
x |
x |
✓ |
x |
x |
x |
| Wang, Feng; Huang, Yuanhanqing; Zhang, Wei; Xia, Yong; Fang, Kun; Cheng, Xin; Zhou, Xiaoyu; Liu, Xueyuan; Yu, Xiaofei; Li, Gang; Wang, Xiaoping; Luo, Guojun; Wu, Danhong; Campbell, Bruce C.V.; Dong, Qiang; Zhao, Yuwu |
Personalized risk prediction of symptomatic intracerebral hemorrhage after stroke thrombolysis using a machine-learning model |
2020 |
sICH |
x |
✓ |
tPA |
2237 |
✓ |
✓ |
x |
x |
x |
x |
multilayer perceptions |
x |
x |
✓ |
x |
age, AFib, glucose level, NIHSS score, door to needle time |
|
5 |
70 |
ns |
30 |
ns |
ratio |
cross-validation |
none |
0.82 |
x |
x |
x |
x |
✓ |
x |
x |
x |
| Wang, Kai; Shou, Qinyang; Ma, Samantha J.; Liebeskind, David; Qiao, Xin J.; Saver, Jeffrey; Salamon, Noriko; Kim, Hosung; Yu, Yannan; Xie, Yuan; Zaharchuk, Greg; Scalzo, Fabien; Wang, Danny J. J. |
Deep Learning Detection of Penumbral Tissue on Arterial Spin Labeling in Stroke. |
2020 |
Penumbral Tissue Mapping |
✓ |
x |
thrombectomy, anterior |
149 |
x |
x |
x |
x |
x |
✓ |
cnn |
✓ |
x |
x |
x |
3-dimensional pseudo-continuous arterial spin labeling (pCASL), DWI |
CBF, ADC |
2x48x48x48 |
90 |
ns |
ns |
12 |
ratio |
cross-validation |
none |
0.958/0.957/0.942 |
x |
x |
0.47/0.43 |
x |
✓ |
x |
x |
x |
| Yu, Yannan; Xie, Yuan; Thamm, Thoralf; Gong, Enhao; Ouyang, Jiahong; Huang, Charles; Christensen, Soren; Marks, Michael P.; Lansberg, Maarten G.; Albers, Gregory W.; Zaharchuk, Greg |
Use of Deep Learning to Predict Final Ischemic Stroke Lesions From Initial Magnetic Resonance Imaging. |
2020 |
final infarct |
x |
✓ |
thrombectomy, anterior |
182 |
✓ |
✓ |
x |
x |
x |
x |
cnn |
✓ |
x |
x |
x |
DWI, PWI, ADC, Tmax, CBF, CBV, MTT |
|
image dim (not specified) x5x8 |
60 |
ns |
20 |
ns |
ratio |
cross-validation |
none |
0.92 |
0.66 |
0.97 |
0.53 |
x |
x |
x |
✓ |
x |
| Rava, Ryan A.; Ionita, Ciprian N.; Mokin, Maxim; Snyder, Kenneth V.; Waqas, Muhammad; Siddiqui, Adnan H.; Levy, Elad I.; Davies, Jason M. |
Performance of angiographic parametric imaging in locating infarct core in large vessel occlusion acute ischemic stroke patients |
2020 |
Infarct core mapping |
✓ |
x |
thrombectomy, successful reca |
25 |
✓ |
✓ |
x |
x |
x |
x |
svm |
x |
x |
x |
✓ |
DSA (anterior-posterior, lateral) |
MTT, AUC |
Per voxel: 2 |
ns |
24 |
ns |
1 |
absolute n |
cross-validation |
none |
0.9038 |
x |
x |
x |
x |
✓ |
x |
x |
x |
| Suetens, Paul; Robben, David; Boers, Anna M.M.; Marquering, Henk A.; Roos, Yvo B.W.E.M.; Majoie, Charles B.L.M.; van Oostenbrugge, Robert J.; van Zwam, Wim H.; Dippel, Diederik W.J.; van der Lugt, Aad; Langezaal, Lucianne L.C.M.; Lemmens, Robin |
Prediction of final infarct volume from native CT perfusion and treatment parameters using deep learning |
2020 |
final infarct |
x |
✓ |
thrombectomy, anterior, <=6h |
188 |
✓ |
✓ |
x |
x |
x |
x |
cnn |
x |
✓ |
x |
x |
Time-attenuation curves and manual AIF + time between stroke onset and CTP imaging |
CTP image, downsampled CTP, arterial input function, metadata |
CTP dim (not specified), +1+1 |
80 |
ns |
ns |
ns |
not reported |
cross-validation |
experimentally set on single split |
0.54 |
x |
x |
0.47 |
x |
✓ |
x |
x |
x |
| Li, X.; Sun, C.; Zhao, Z.; Wang, F.; Zheng, X.; Zou, J.; Wu, M.; Ge, W.; Zhou, J. |
Using machine learning to predict stroke-associated pneumonia in Chinese acute ischaemic stroke patients |
2020 |
post-stroke pneumonia |
x |
✓ |
ns |
3160 |
✓ |
✓ |
x |
x |
x |
x |
xgboost |
x |
x |
✓ |
x |
age, premorbid mRS, NIHSS at admission, FBG, sex, and history of AF |
|
6 |
80 |
ns |
20 |
ns |
ratio |
cross-validation |
grid |
0.841 |
0.81 |
0.733 |
x |
x |
x |
x |
✓ |
x |
| Burgermeister, Simon; Bernava, Gianmarco; Rosi, Andrea; Carrera, Emmanuel; Hofmeister, Jeremy; Vargas, Maria Isabel; Montet, Xavier; Poletti, Pierre-Alexandre; Platon, Alexandra; Lovblad, Karl-Olof; MacHi, Paolo |
Clot-Based Radiomics Predict a Mechanical Thrombectomy Strategy for Successful Recanalization in Acute Ischemic Stroke |
2020 |
first time reca, num passages |
x |
✓ |
thrombectomy |
136 |
✓ |
✓ |
x |
x |
x |
x |
svm |
x |
✓ |
x |
x |
NCCT, CT-angiography |
thrombus-related radiomic features |
9 |
ns |
109 |
ns |
47 |
absolute n |
nested cross-validation |
none |
0.88 |
0.5 |
0.971 |
x |
x |
✓ |
x |
x |
x |
| Nishi, Hidehisa; Ishii, Akira; Ono, Isao; Okawa, Masakazu; Miyamoto, Susumu; Oishi, Naoya; Ogura, Takenori; Chihara, Hideo; Sunohara, Tadashi; Fukumitsu, Ryu; Imamura, Hirotoshi; Sadamasa, Nobutake; Nakahara, Ichiro; Sakai, Nobuyuki; Yamana, Norikazu; Hatano, Taketo |
Deep Learning-Derived High-Level Neuroimaging Features Predict Clinical Outcomes for Large Vessel Occlusion |
2020 |
90 day mrs |
x |
✓ |
thrombectomy, anterior |
324 |
✓ |
✓ |
x |
x |
x |
x |
cnn |
✓ |
x |
x |
x |
DWI |
|
128×128×32 |
ns |
200 |
ns |
74 |
absolute n |
cross-validation |
none |
0.73 |
0.722 |
0.602 |
x |
x |
x |
x |
✓ |
x |
| Alaka, Shakiru A.; Brobbey, Anita; Williamson, Tyler; Demchuk, Andrew M.; Sajobi, Tolulope T.; Menon, Bijoy K.; Goyal, Mayank; Hill, Michael D. |
Functional Outcome Prediction in Ischemic Stroke: A Comparison of Machine Learning Algorithms and Regression Models |
2020 |
90 day mrs |
x |
✓ |
ns |
1121 |
✓ |
✓ |
x |
x |
x |
x |
svm |
x |
x |
✓ |
x |
age, sex, NIHSS, treatment received, blood glucose, systolic blood pressure, diastolic blood pressure, HTN, diabetes, AFib, international normalize ratio, creatinine, platelet count, hematocrit |
|
14 |
ns |
614 |
ns |
507 |
absolute n |
cross-validation |
grid |
0.71 |
0.65 |
0.77 |
x |
x |
✓ |
x |
x |
x |
| Fu, Bowen; Qi, Shouliang; Tao, Lin; Xu, Haibin; Kang, Yan; Yao, Yudong; Chen, Huisheng; Yang, Benqiang; Duan, Yang |
Image Patch-Based Net Water Uptake and Radiomics Models Predict Malignant Cerebral Edema After Ischemic Stroke |
2020 |
edema |
x |
✓ |
MCA |
116 |
✓ |
✓ |
x |
x |
x |
x |
random forest |
x |
✓ |
✓ |
x |
NCCT + age, gender, and NIHSS score |
Image patch-based net water uptake + standard deviation, slope, entropy, skewness, kurtosis + age, gender, and NIHSS score |
16 |
ns |
115 |
ns |
ns |
not reported |
cross-validation |
grid |
0.96 |
0.85 |
0.95 |
0.91 |
x |
✓ |
x |
x |
x |
| Grosser, Malte; Gellißen, Susanne; Borchert, Patrick; Sedlacik, Jan; Nawabi, Jawed; Fiehler, Jens; Forkert, Nils Daniel |
Improved multi-parametric prediction of tissue outcome in acute ischemic stroke patients using spatial features. |
2020 |
final infarct |
x |
✓ |
first time, anterior, NIHSS > 4, <12hrs |
99 |
✓ |
✓ |
x |
x |
x |
x |
xgboost |
✓ |
x |
x |
x |
DWI, PWI |
ADC, CBF, CBV, MTT, Tmax, MNI x,y,z coordinates |
Per voxel: 8 |
ns |
98 |
ns |
ns |
not reported |
leave-one-out |
grid |
0.893 |
x |
x |
0.387 |
x |
✓ |
x |
x |
x |
| Lee, Hyunna; Lee, Eun-Jae; Ham, Sungwon; Lee, Han-Bin; Lee, Ji Sung; Kwon, Sun U.; Kim, Jong S.; Kim, Namkug; Kang, Dong-Wha |
Machine Learning Approach to Identify Stroke Within 4.5 Hours. |
2020 |
onset time |
✓ |
x |
ns |
355 |
✓ |
✓ |
x |
x |
x |
x |
random forest |
✓ |
x |
x |
x |
MRI DWI and FLAIR, infarct segmentation |
infarct volume, ADC, DWI, FLAIR: intensity, gradient, texture information, local binary patterns within the infarct regions, statistical measurements (ie, mean, SD, skewness, and kurtosis) |
3x89 |
85 |
ns |
ns |
ns |
not reported |
single split |
none |
0.851 |
0.758 |
0.826 |
x |
x |
x |
✓ |
x |
✓ |
| Bacchi, Stephen; Patel, Sandy; Zerner, Toby; Kleinig, Timothy; Jannes, Jim; Oakden-Rayner, Luke |
Deep Learning in the Prediction of Ischaemic Stroke Thrombolysis Functional Outcomes: A Pilot Study |
2020 |
90 day mrs, dichotomized NIHSS at 24h |
x |
✓ |
ns |
204 |
x |
x |
x |
x |
x |
x |
cnn |
x |
✓ |
✓ |
x |
NCCT and clinical (not specified) |
|
128x128x12 + (not specified) |
85 |
ns |
15 |
ns |
ratio |
cross-validation |
grid |
0.75/0.7 |
0.56/0.93 |
0.93/0.53 |
0.69/0.74 |
x |
x |
x |
✓ |
x |
| Brugnara, Gianluca; Neuberger, Ulf; Mahmutoglu, Mustafa A.; Foltyn, Martha; Herweh, Christian; Heiland, Sabine; Ulfert, Christian; Bendszus, Martin; Mohlenbruch, Markus A.; Pfaff, Johannes A.R.; Vollmuth, Philipp; Nagel, Simon; Schonenberger, Silvia; Ringleb, Peter Arthur |
Multimodal Predictive Modeling of Endovascular Treatment Outcome for Acute Ischemic Stroke Using Machine-Learning |
2020 |
90 day mrs |
x |
✓ |
thrombectomy, anterior |
246 |
✓ |
✓ |
x |
x |
x |
x |
Gradient Boosting |
x |
✓ |
✓ |
x |
Clinical variables (sex, age, premorbid mRS score, NIHSS score on admission, presence of wake-up stroke, interval from symptom onset to imaging and to groin puncture, concomitant treatment with intravenous r-tPA, comorbidities [HTN, hyperlipidemia, coronary heart disease, diabetes including Hba1c and glucose level]), NCCT (acute ischemic volume, eASPECTS), CTA (site of vessel occlusion on CTA, e-CTA collateral score), CTP parameters (volumes of ischemic penumbra, infarct core, and penumbra to core ratio) |
|
20 |
ns |
ns |
ns |
ns |
not reported |
.632 bootstrap |
.632 bootstrap |
0.747 |
x |
x |
x |
x |
✓ |
x |
x |
x |
| Ramos, Lucas A.; Zwinderman, Aeilko H.; Olabarriaga, Silvia D.; Kappelhof, Manon; Van Kranendonk, Katinka; Strijkers, Gustav J.; Majoie, Charles B. L. M.; Marquering, Henk A.; van Os, Hendrikus J. A.; Kruyt, Nyika D.; Wermer, Mariekke J. H.; Chalos, Vicky; van der Lugt, Aad; Roos, Yvo B. W. E. M.; van Zwam, Wim H.; van der Schaaf, Irene C.; van Walderveen, Marianne A. A. |
Predicting Poor Outcome Before Endovascular Treatment in Patients With Acute Ischemic Stroke |
2020 |
90 day mrs |
x |
✓ |
thrombectomy, anterior |
1526 |
✓ |
✓ |
x |
x |
x |
x |
multilayer perceptions |
x |
✓ |
✓ |
x |
previous stroke, myocardial infarction, peripheral artery disease, DM, HTN, AFib, Hypercholesterolemia, Antiplatelet use, DOAC use, Coumarin use, Heparin use, Blood pressure medication, Statin use, HAS on baseline NCCT, Relevant (new) ischemia/hypodensity, hemorrhagic transformation, Leukoaraiosis, Old infarcts in same ASPECTS region, Intracranial atherosclerosis on CTA scored by core lab, Sex, Most proximal occlusion segment on CTA scored by core lab, based on CBS, Smoking, Inclusion on weekday or weekend, Admission between 17.00-08-00 (weekday)/ weekend or holiday, Transfer from other hospital, Intravenous alteplase treatment, No abnormalities at symptomatic carotid bifurcation, 50% or more atherosclerotic stenosis at symptomatic carotid bifurcation, Atherosclerotic occlusion/Floating thrombus/Pseudo-occlusion/Carotid dissection/Occlusion side on CTA scored by core lab, In-hospital stroke, Contraindications for IVT, Second occlusion in other territory present, Collateral score, Pre-stroke mRS, ASPECTS, CBS, NIHSS; GLucose level, Systolic BP, Diastolic BP, INR, Thrombocyte count, CRP, Age, Total Glasgow Coma Scale, Onset-to-groin-time, onset-two-IVT |
|
51 |
90 |
ns |
10 |
ns |
ratio |
nested cross-validation |
grid |
0.81 |
0.53 |
0.89 |
x |
x |
✓ |
x |
x |
x |
| Kim, Yoon-Chul; Kim, Hyung Jun; Chung, Jong-Won; Kim, In Gyeong; Seo, Woo-Keun; Kim, Gyeong-Moon; Bang, Oh Young; Seong, Min Jung; Kim, Keon Ha; Jeon, Pyoung; Nam, Hyo Suk |
Novel estimation of Penumbra zone based on infarct growth using machine learning techniques in acute ischemic stroke |
2020 |
final infarct with and without successful recanalization |
x |
✓ |
MCA, thrombectomy |
92 |
✓ |
✓ |
x |
x |
x |
x |
random forest |
✓ |
x |
x |
x |
DWI, PWI, ADC |
ADC and rTTP features |
per voxel: 48 |
ns |
53 |
ns |
39 |
absolute n |
cross-validation |
random |
x |
x |
x |
0.49 |
x |
✓ |
x |
x |
x |
| Zihni, Esra; Madai, Vince Istvan; Livne, Michelle; Galinovic, Ivana; Khalil, Ahmed A.; Fiebach, Jochen B.; Frey, Dietmar |
Opening the black box of artificial intelligence for clinical decision support: A study predicting stroke outcome. |
2020 |
90 day mrs |
x |
✓ |
supratentorial |
314 |
✓ |
✓ |
x |
x |
x |
x |
multilayer perceptions |
x |
x |
✓ |
x |
Age, Sex, NIHSS, Cardiac history, DM, Hypercholesterolemia, Thrombolysis |
|
7 |
75 |
ns |
25 |
ns |
ratio |
nested cross-validation |
grid |
0.83 |
x |
x |
x |
x |
✓ |
x |
x |
x |
| Wu, MingRu; Li, Xiang; Wang, FuSang; Zheng, XiaoHan; Sun, Chao; Zou, JianJun; Pan, XiDing; Zhu, YuBing; Zhao, Zheng; Jiang, ChunLian; Liu, YuKai; Zhou, JunShan; Yang, Jie; Wang, ShiHao |
Predicting 6-Month Unfavorable Outcome of Acute Ischemic Stroke Using Machine Learning |
2020 |
six-month dichotomized mRS |
x |
✓ |
ns |
1735 |
✓ |
✓ |
x |
x |
x |
x |
random forest |
x |
x |
✓ |
x |
age, sex, onset-to-admission delay > 4,5h, HTN, DM, hyperlipidemia, coronary artery disease, AFib, previous cerebral infarction, valvular heart disease, smoking (never, former, current), drinking (never, former, current), premorbid mRS, NIHSS at admission, BMI, pulse, systolic BP, diastolic BP, platelet count, urea nitrogen, creatinine, blood fasting glucose, triglycerides, total cholesterol, LDL-cholesterol, HDL cholesterol, IV thrombolysis |
|
21 |
80 |
ns |
20 |
ns |
ratio |
cross-validation |
grid |
0.874 |
0.5 |
0.95 |
x |
x |
x |
x |
✓ |
x |
| Nishi, Hidehisa; Ono, Isao; Okawa, Masakazu; Miyamoto, Susumu; Ishii, Akira; Oishi, Naoya; Ogura, Takenori; Chihara, Hideo; Hatano, Taketo; Sunohara, Tadashi; Fukumitsu, Ryu; Imamura, Hirotoshi; Sakai, Nobuyuki; Nakahara, Ichiro; Yamana, Norikazu; Sadamasa, Nobutake |
Predicting Clinical Outcomes of Large Vessel Occlusion before Mechanical Thrombectomy Using Machine Learning |
2020 |
90 day mrs |
x |
✓ |
thrombectomy, anterior |
502 |
✓ |
✓ |
x |
x |
x |
x |
regularized LR |
x |
✓ |
✓ |
x |
age, sex, premorbid mRS, HTN, DM, AFib, antithrombotic drug intake, stroke onset pattern, NIHSS, blodd glucose level on arrival, site of occlusion pretreatment, side of occlusion, ASPECTS, iv-tPA administration, last-seen-well-time, time at hospital arrival |
|
16 |
ns |
387 |
ns |
115 |
absolute n |
cross-validation |
grid |
0.9 |
x |
x |
x |
x |
x |
x |
✓ |
x |
| Su, Emily Chia-Yu; Chiu, Hung-Wen; Chung, Chen-Chih; Hong, Chien-Tai; Huang, Yao-Hsien; Chan, Lung; Hu, Chaur-Jong |
Predicting major neurologic improvement and long-term outcome after thrombolysis using artificial neural networks |
2020 |
90 day mrs, >=8 point NIHSS improvement at 24h |
x |
✓ |
tPA, age 18-80, NIHSS 4-25 |
196 |
✓ |
✓ |
x |
x |
x |
x |
multilayer perceptions |
✓ |
✓ |
✓ |
x |
age, baseline systolic BP, diastolic BP, heart rate, glucose level at admission, GCS, NIHSS score, DM, HbA1c level, stroke subtype |
|
10 |
80 |
ns |
ns |
54 |
absolute n |
cross-validation |
none |
x |
0.824/0.783 |
0.533/0.583 |
x |
x |
✓ |
x |
x |
x |
| Kang, Yoon Jung; Cho, Han Jin; Kim, Nae Ri; Lee, Suk Min; Choi, Byung Kwan; Sung, Sang Min; Cho, Giphil |
Prediction of early neurological deterioration in acute minor ischemic stroke by machine learning algorithms |
2020 |
worsening of NIHSS within 3 days |
x |
✓ |
NIHSS => 3 |
739 |
✓ |
✓ |
x |
x |
x |
x |
tree boosting |
✓ |
✓ |
✓ |
x |
age, sex, HTN, diabetes, hyperlipidemia, smoking, coronary heart disease, AFib, previous TIA/ischemic stroke/hemorrhagic stroke, location and arterial territory, stroke subtypes, hemorrhagic transformation, NIHSS, relevant arterial steno-occlusive lesions |
|
17 |
ns |
ns |
ns |
ns |
not reported |
none |
none |
0.934 |
0.833 |
x |
0.8 |
x |
✓ |
x |
x |
x |
| Heo, Tak Sung; Kim, Yu Seop; Choi, Jeong Myeong; Jeong, Yeong Seok; Seo, Soo Young; Lee, Jun Ho; Jeon, Jin Pyeong; Kim, Chulho |
Prediction of stroke outcome using natural language processing-based machine learning of radiology report of brain MRI |
2020 |
90 day mrs |
x |
✓ |
first time |
1840 |
✓ |
✓ |
x |
x |
x |
x |
cnn |
✓ |
x |
x |
x |
MRI Radiology Reports |
|
ns |
70 |
ns |
30 |
ns |
ratio |
cross-validation |
grid |
0.805 |
x |
x |
x |
x |
✓ |
x |
x |
x |
| Matsumoto, Koutarou; Kamouchi, Masahiro; Soejima, Hidehisa; Yonehara, Toshiro; Nohara, Yasunobu; Nakashima, Naoki |
Stroke Prognostic Scores and Data-Driven Prediction of Clinical Outcomes after Acute Ischemic Stroke |
2020 |
90 day mrs, in-hospital mortality |
x |
✓ |
ns |
3445 |
✓ |
✓ |
x |
x |
x |
x |
Gradient boosting |
✓ |
✓ |
✓ |
x |
Clinical data: Age, Men, HTN, DM, Dyslipidemia, AFib, Current smoker, BMI, Peripheral artery disease, coronary artery disease, Congestive heart failure, Kidney disease on dialysis, Stroke, Cancer, pre-admission mRS, Onset-to-admission time, stroke subtype (Small vessel occlusion, Large artery atherosclerosis, Cardioembolic, Other undetermined, Tranisent ischemic attack, Systolic BP, Diastolic BP, Heart rate, Body temperature, NIHSS score\n\nLaboratory data: Hemoglobin, White Blood Cell Count, Albumin, Albumin/globulin ratio, ALT, AST, Lactate dehydrogenase, gamma-GT, total Bilirubin, BNP, CK, CRP, D-Dimers, HbA1c, Glucose, BUN, Creatinine, HDL, LDL, Triglycerides, Uric acid, Potassium, Sodium, Chloride |
|
49 |
80 |
ns |
20 |
ns |
ratio |
nested cross-validation |
none |
0.84/0.92 |
x |
x |
x |
x |
x |
x |
✓ |
x |
| Grosser, Malte; Gellißen, Susanne; Borchert, Patrick; Sedlacik, Jan; Nawabi, Jawed; Fiehler, Jens; Forkert, Nils D. |
Localized prediction of tissue outcome in acute ischemic stroke patients using diffusion- and perfusion-weighted MRI datasets. |
2020 |
final infarct |
x |
✓ |
anterior, <12h |
99 |
x |
x |
x |
x |
x |
✓ |
logistic regression |
✓ |
x |
x |
x |
DWI, PWI |
ADC, CBV, CBF, MTT, Tmax + lesion probabilites |
Per voxel: 6 |
ns |
98 |
ns |
ns |
not reported |
leave-one-out |
none |
0.872 |
0.444 |
0.955 |
0.348 |
x |
✓ |
x |
x |
x |
| Velagapudi, Lohit; Mouchtouris, Nikolaos; Schmidt, Richard F.; Khanna, Omaditya; Sweid, Ahmad; Al Saiegh, Fadi; Gooch, M. Reid; Jabbour, Pascal; Rosenwasser, Robert H.; Tjoumakaris, Stavropoula; Vuong, David; Sadler, Bryan |
A Machine Learning Approach to First Pass Reperfusion in Mechanical Thrombectomy: Prediction and Feature Analysis |
2021 |
first time reca |
x |
✓ |
thrombectomy |
220 |
x |
✓ |
x |
x |
x |
x |
random forest |
x |
x |
✓ |
✓ |
age > 65, age > 80, BMI (categorical low, medium, high based on standard deviation), sex, history of Cerebrovascular accident, Coronary Artery Disease, HTN, T2DM, Hhyperlipidemia, or AFib, smoking status, blood thinner use, statin use, NIHSS on presentation, anterior or posterior stroke, proximal or distal stroke, use of tPA before procedure, time between symptom onset to catheterization, use of stent retriever, and use of aspiration |
|
20 |
66 |
ns |
33 |
ns |
ratio |
cross-validation |
none |
0.659 |
0.615 |
0.702 |
x |
x |
✓ |
x |
x |
x |
| Zhang, Haoyue; Polson, Jennifer S.; Nael, Kambiz; Salamon, Noriko; Yoo, Bryan; El-Saden, Suzie; Scalzo, Fabien; Speier, William; Arnold, Corey W. |
Intra-Domain Task-Adaptive Transfer Learning to Determine Acute Ischemic Stroke Onset Time |
2021 |
onset time |
✓ |
x |
ns |
422 |
✓ |
✓ |
x |
x |
x |
x |
cnn |
✓ |
x |
x |
x |
FLAIR, DWI, ADC, T2 |
stacked (flair, dwi, t2) slices |
ns |
64 |
ns |
20 |
ns |
ratio |
single split |
none |
0.74 |
0.7 |
0.81 |
x |
x |
x |
✓ |
x |
✓ |
| Silva, Carlos A.; Pinto, Adriano; Pereira, Sergio; Alves, Victor; Reyes, Mauricio; Meier, Raphael; Wiest, Roland |
Combining unsupervised and supervised learning for predicting the final stroke lesion |
2021 |
final infarct |
x |
✓ |
thrombectomy |
75 |
x |
x |
x |
x |
x |
✓ |
Restricted Boltzmann Machines + CNN |
✓ |
x |
x |
x |
ADC, MTT, TTP, Tmax, rCBF, rCBV |
ADC, MTT, TTP, Tmax, rCBF, rCBV |
6x84x84 |
ns |
36 |
ns |
32 |
absolute n |
cross-validation |
none |
x |
x |
x |
0.38 |
x |
✓ |
x |
x |
x |
| Debs, Noelie; Frindel, Carole; Cho, Tae-Hee; Nighoghossian, Norbert; Rousseau, David; Berthezene, Yves; Eker, Omer; Ovize, Michel; Buisson, Marielle; Mechtouff, Laura |
Impact of the reperfusion status for predicting the final stroke infarct using deep learning |
2021 |
final infarct |
x |
✓ |
thrombectomy |
109 |
✓ |
✓ |
x |
x |
x |
x |
cnn |
✓ |
x |
x |
x |
DWI, FLAIR, PWI, CBF, CBV, MTT, Tmax, TTP |
DWI, ADC, CBV, CBF, TMAX |
5x192x192 |
80 |
ns |
ns |
ns |
not reported |
cross-validation |
independently in fixed search space |
0.87/0.81 |
0.43/0.63 |
x |
0.43/0.44 |
x |
x |
x |
✓ |
x |
| Hakim, Arsany; Wiest, Roland; Lansberg, Maarten G.; Winzeck, Stefan; Parsons, Mark W.; Lucas, Christian; Robben, David; Reyes, Mauricio; Zaharchuk, Greg; Christensen, Soren |
Predicting Infarct Core From Computed Tomography Perfusion in Acute Ischemia With Machine Learning: Lessons From the ISLES Challenge |
2021 |
Infarct Core Mapping |
✓ |
x |
thrombectomy, anterior |
103 |
✓ |
x |
x |
x |
x |
x |
cnn |
x |
✓ |
x |
x |
dynamic CTP, postprocessed perfusion maps (CBF, CBV, MTT, Tmax) calculated by conventional thresholding method (RAPID) |
|
ns |
ns |
63 |
ns |
40 |
absolute n |
none |
none |
x |
0.55 |
x |
0.51 |
x |
x |
x |
✓ |
x |
| Yu, Y.; Xie, Y.; Thamm, T.; Gong, E.; Ouyang, J.; Christensen, S.; Marks, M.P.; Lansberg, M.G.; Zaharchuk, G.; Albers, G.W. |
Tissue at Risk and Ischemic Core Estimation Using Deep Learning in Acute Stroke |
2021 |
Tissue at risk, ischemic core |
✓ |
x |
ns |
237 |
✓ |
✓ |
x |
x |
x |
x |
cnn |
✓ |
x |
x |
x |
DWI, ADC, Tmax, MTT, CBF, CBV, and thresholded masks of Tmax and ADC |
|
image dim (not specified) x5x8 |
60 |
ns |
20 |
ns |
ratio |
cross-validation |
none |
0.92/0.94 |
x |
x |
0.6/0.57 |
x |
x |
x |
✓ |
x |
| Kappelhof, N.; Ramos, L.A.; Kappelhof, M.; van Kranendonk, K.R.; Majoie, C.B.L.M.; Marquering, H.A.; van Os, H.J.A.; Kruyt, N.D.; Wermer, M.J.H.; Dippel, Diederik; Lingsma, Hester; Chalos, V.; Roos, Y.B.W.E.M.; van Zwam, W.H.; van der Schaaf, I.C.; van Walderveen, M.A.A.; van Oostenbrugge, R.J. |
Evolutionary algorithms and decision trees for predicting poor outcome after endovascular treatment for acute ischemic stroke |
2021 |
90 day mrs, dichotomized mRS after recanalization |
x |
✓ |
thrombectomy |
1363 |
x |
x |
x |
x |
x |
✓ |
fuzzy decision tree |
x |
✓ |
✓ |
x |
age, sex, ASPECTS, Clot Burden Score, Creatinine, CRP, GCS, Glucose, INR, thrombolysis, prior use of antiplatelets/coumarin/antihypertensive, heparin use, NOAC use, statin use, NIHSS, pre-stroke-mRS, history of AFib/Diabetes/Hypercholesterinemia/HTN/Myocardial infarction/peripheral artery disease/previous stroke/previos stroke in the same vascular area, transfer from other hospital, BP at baseline, smoking, onset-to-thrombolysis,imaging-to-puncture, imaging-to-thrombolysis, ER-to-imaging, ER-to-puncture, onset-to-imaging, onset-to-intervention, hospital-to-intervention-hospital (intervention hospital, first hospital, referral hospital), off-hours/weekend admission, onset-to-groin/CTA (scored by radiologist): occluded segment, collateral score, hyperdense artery sign, hemorrhagic transformation, new ischemia, Leukoaraiosis, previous infarctions, carotid dissection, floating thrombus, normal carotid bifurcation, amount of aortic stenosis, atherosclerotic occlusion, lesions at bifurcation, carotid pseudo-occlusion, carotid web, intracranial atherosclerosis, intracranial vascular malformation, most proximal occluded segment, occlusion side, |
|
86 |
ns |
1090 |
ns |
ns |
not reported |
cross-validation |
none |
x |
x |
x |
x |
✓ |
✓ |
x |
x |
x |
| Hamann, Janne; Luft, Andreas R.; Wegener, Susanne; Herzog, Lisa; Sick, Beate; Wehrli, Carina; Bink, Andrea; Piccirelli, Marco; Stippich, Christoph; Dobrocky, Tomas; Gralla, Jan; Wiest, Roland; Panos, Leonidas; Fischer, Urs; Arnold, Marcel; Kaesmacher, Johannes |
Machine-learning-based outcome prediction in stroke patients with middle cerebral artery-M1 occlusions and early thrombectomy |
2021 |
90 day mrs |
x |
✓ |
MCA, thrombectomy, M1, <6 hrs |
222 |
✓ |
✓ |
x |
x |
x |
x |
random forest |
✓ |
x |
✓ |
x |
rCBV, rCBF, Tmax, time-to-peak, temporal maximum intensity projections, manually drawn regions of interest, automated ROI segmentations, patient characteristics (Age, Sex, Risk factors, Previous medication, On admission NIHSS, On admission vital signs, on admission lab parameters, infarct side, IVT, Onset to imaging, Onset to groin puncture, General anesthesia, Collateralization status, Perfusion imaging, After treatment TICI/NIHSS/ICH) |
age, NIHSS, systolic blood pressure, hypertension, diabetes, smoking, previous ischemic event, preceding IVT, onset to groin puncture time, collateralization status, perfusion value of the medial MCA |
12 |
80 |
ns |
20 |
ns |
ratio |
cross-validation |
none |
0.684 |
x |
x |
x |
x |
x |
x |
✓ |
x |
| Jiang, B.; Zhu, G.; Xie, Y.; Heit, J.J.; Chen, H.; Li, Y.; Zaharchuk, G.; Wintermark, M.; Ding, V.; Eskandari, A.; Michel, P. |
Prediction of clinical outcome in patients with large-vessel acute ischemic stroke: Performance of machine learning versus span-100 |
2021 |
90 day mrs |
x |
✓ |
anterior |
1431 |
✓ |
✓ |
x |
x |
x |
x |
xgboost |
x |
✓ |
✓ |
x |
baseline NIHSS, age, glucose at admission, ischemic core volume on CTP, penumbra volume on CTP, CTA clot burden score |
|
6 |
80 |
ns |
ns |
ns |
not reported |
cross-validation |
none |
0.8 |
0.722 |
0.74 |
x |
x |
x |
x |
✓ |
x |