Background
Methods
Patients and EEG recordings
Etiology and outcome
Electroencephalographic and clinical features
Name | Type | Categories/values |
---|---|---|
EEG features | ||
EEG background continuity | Ordinal | 1. Continuous/nearly continuous |
2. Discontinuous | ||
3. Burst suppression | ||
4. Suppressed | ||
Background amplitude | Ordinal | 1. < 10 µV |
2. 10–20 µV | ||
3. > 20 µV | ||
Background frequency | Ordinal | 1. 1–3.5 Hz |
2. 4–7.5 Hz | ||
3. 8–13 Hz | ||
4. > 13 Hz | ||
Background reactivity | Binary | Absent/present |
Background symmetry | Binary | Absent/present |
Stage II sleep transients | Binary | Absent/present |
Sporadic epileptiform discharges | Binary | Absent/present |
Rhythmic or periodic patterns ("main term 2") | Categorical | 1. Periodic discharges |
2. Rhythmic delta activity | ||
3. Rhythmic spike waves/sharp wave | ||
4. No rhythmic or periodic pattern | ||
Clinical features | ||
Age | Continuous | Real number |
Gender | Binary | Female/male |
Glasgow coma scale | Ordinal | 3–15 |
C-reactive protein | Continuous | Real number |
Etiology | Categorical | 1. Stroke |
2. Trauma/neurosurgery | ||
3. Toxic/metabolic/infectious | ||
4. Hypoxic ischemic encephalopathy | ||
5. No etiology available |
Classification model
Results
Patients
Etiology | N | Age [IQR] | Female (%) | Survival (%) | Favorable outcome (%) | EEG delay [IQR] |
---|---|---|---|---|---|---|
Stroke | 82 | 67 [55 78] | 41 (50.0) | 37 (45.1) | 21 (25.6) | 69 [37 122] |
TBI/NS | 50 | 63 [40 73] | 13 (26.0) | 33 (66.0) | 24 (48.0) | 75 [43 115] |
TBI | 44 | 63 [40 73] | 12 (27.3) | 30 (68.2) | 21 (47.7) | 71 [43 112] |
MIII | 47 | 62 [51 73] | 14 (29.8) | 26 (55.3) | 22 (46.8) | 139 [48 268] |
HIE | 110 | 66 [54 75] | 30 (27.3) | 42 (38.2) | 34 (30.9) | 24 [17 52] |
No etiology available at recruitment time | 75 | 68 [59 77] | 25 (33.3) | 49 (65.3) | 38 (50.7) | 101 [33 192] |
All | 364 | 67 [55 75] | 123 (33.8) | 187 (51.4) | 139 (38.2) | 59 [24 138] |
Performance of the general classifiers
AUC | Accuracy | Sensitivity | Specificity | PPV | NPV | |
---|---|---|---|---|---|---|
Predicting survival | ||||||
EEG features | .812 | .736 | .833 | .639 | .694 | .796 |
[.721 .874] | [.657 .814] | [.739 .928] | [.519 .760] | [.588 .801] | [.683 .909] | |
Clinical features | .643 | .596 | .617 | .574 | .587 | .603 |
[.535 .737] | [.508 .683] | [.494 .740] | [.450 .698] | [.466 .709] | [.478 .729] | |
All features | .806 | .752 | .717 | .787 | .768 | .739 |
[.721 .871] | [.675 .829] | [.603 .831] | [.684 .890] | [.657 .878] | [.632 .845] | |
Predicting | ||||||
Favorable outcome | ||||||
EEG features | .790 | .703 | .848 | .613 | .574 | .868 |
[.693 .862] | [.621 .784] | [.744 .952] | [.503 .724] | [.456 .691] | [.777 .959] | |
Clinical features | .641 | .628 | .587 | .653 | .509 | .721 |
[.537 .736] | [.542 .714] | [.445 .729] | [.546 .761] | [.375 .644] | [.614 .827] | |
All features | .777 | .703 | .565 | .787 | .619 | .747 |
[.687 .852] | [.621 .784] | [.422 .709] | [.694 .879] | [.472 .766] | [.651 .843] |
Stroke | TBI/NS | TBI | MIII | HIE | No etiology | |
---|---|---|---|---|---|---|
Predicting survival | ||||||
General classifier | .695 | .955 | .944 | .697 | .958 | .737 |
[.433 .878] | [.734 1.000] | [.628 1.000] | [.423 .894] | [.827 .996] | [.442 .936] | |
Specific classifiers | .666 | .694 | .800 | .452 | .833 | .646 |
[.534 .769] | [.469 .851] | [.571 .907] | [.290 .628] | [.741 .902] | [.488 .775] | |
Predicting | ||||||
Favorable outcome | ||||||
General classifier | .706 | .819 | .866 | .742 | .921 | .678 |
[.474 .880] | [.472 .972] | [.500 1.000] | [.475 .909] | [.762 .979] | [.404 .876] | |
Specific classifiers | .662 | .651 | .723 | .396 | .865 | .652 |
[.510 .776] | [.480 .803] | [.547 .851] | [.240 .568] | [.781 .923] | [.512 .765] |