Display Settings:

Items per page
We are sorry, but NCBI web applications do not support your browser and may not function properly. More information

Results: 8

1.
Fig. 4

Fig. 4. From: Quickest detection of drug-resistant seizures: An optimal control approach.

Sequence of first singular values σ1 and correspondent first singular vectors v1 around an ictal period (gray background) in patient 1 (seizure s1, A), 2 (s1, B), 3 (s3, C), and 4 (s3, D).

Sabato Santaniello, et al. Epilepsy Behav. 2011 December;22(7-5):S49-S60.
2.
Fig. 3

Fig. 3. From: Quickest detection of drug-resistant seizures: An optimal control approach.

(A) Schematic of an m = 2 state HMM. (B) Sampling probability distribution function of the observations zk in state xk = 0 (nonictal) and xk = 1 (ictal). Data collected from patient 2. (C) Seizure onset detection formulated as a feedback control problem.

Sabato Santaniello, et al. Epilepsy Behav. 2011 December;22(7-5):S49-S60.
3.
Fig. 2

Fig. 2. From: Quickest detection of drug-resistant seizures: An optimal control approach.

(A) Three nodes in a network are loosely connected and the correspondent connectivity matrix A is close to identity with full rank. (B) Two of three nodes are strongly connected and the connectivity matrix B is close to losing rank. (C) Average direction and amplification of the first singular vectors corresponding to matrices A and B.

Sabato Santaniello, et al. Epilepsy Behav. 2011 December;22(7-5):S49-S60.
4.
Fig. 8

Fig. 8. From: Quickest detection of drug-resistant seizures: An optimal control approach.

Sensitivity analysis. The average detection delay and false-positive rate for patients 1 and 2 (A), 3 and 4 (B) are computed for γ ranging from 0.01 to 0.999 on the validation data. Gray circles denote the average performance achieved with the value of γ used in Tables 2 and Table 3.

Sabato Santaniello, et al. Epilepsy Behav. 2011 December;22(7-5):S49-S60.
5.
Fig. 5

Fig. 5. From: Quickest detection of drug-resistant seizures: An optimal control approach.

(A–D) Maximum likelihood estimate of parameters Θx (black line) in (6) with 95% confidence bounds (gray background) for nonictal (A, C) and ictal (B, D) state. (E) History-dependent estimate of the probability q0(zk|Hk) (black line) and q1(zk|Hk) (gray dotted line) around a specific hand-annotated seizure (gray background). Plots refer to patient 2. Probabilities in (E) refer to seizure s3 (validation data).

Sabato Santaniello, et al. Epilepsy Behav. 2011 December;22(7-5):S49-S60.
6.
Fig. 7

Fig. 7. From: Quickest detection of drug-resistant seizures: An optimal control approach.

QD on validation data. The electrographic onset (dashed vertical black line), the correspondent QD estimation (circles), and threshold Fk (gray and dash-dot black lines) for patients 3 (A) and 4 (B). For each patient, two QD estimations are computed (gray and dash-dot black) for different values of parameter γ. Plots refer to seizures s3 (patient 3) and s3 (patient 4), respectively. In (A) and (B) (bottom plot) the horizontal dashed black line denotes the threshold for the BE detector.

Sabato Santaniello, et al. Epilepsy Behav. 2011 December;22(7-5):S49-S60.
7.
Fig. 6

Fig. 6. From: Quickest detection of drug-resistant seizures: An optimal control approach.

QD on validation data. The electrographic onset (dashed vertical black line), the correspondent QD estimation (circles), and threshold Fk (gray and dash-dot black lines) for patients 1 (A) and 2 (B). For patient 2, two QD estimations are computed for different values of parameter γ (black circle γ = 0.52; gray circle γ = 0.01). Plots refer to seizures s1 (patient 1) and s3 (patient 2), respectively. In (B) (bottom plot) the horizontal dashed black line denotes the threshold for the BE detector.

Sabato Santaniello, et al. Epilepsy Behav. 2011 December;22(7-5):S49-S60.
8.
Fig. 1

Fig. 1. From: Quickest detection of drug-resistant seizures: An optimal control approach.

Three-dimensional reconstructions of the brain for patients 1 (A), 2 (B), 3 (C), and 4 (D) with the electrode grids superimposed. Red-squared boxes denote the hand-annotated seizure foci. In (A) to (D), yellow circles are electrodes visible in their proper location on the cortex; green circles are electrodes that would be visible where the underlying cortex rendered properly; cyan circles are electrodes occluded by cortex (rendered or not) because they wrap around underneath or behind the temporal lobe. In patient 2, recordings from electrodes 1, 4, 6, 23, and 33–37 were not included in this study. In patient 3, recordings from electrode 34–64 were not included in this study. Electrodes not included in this study showed neglegible modulation during the ictal periods.

Sabato Santaniello, et al. Epilepsy Behav. 2011 December;22(7-5):S49-S60.

Display Settings:

Items per page

Supplemental Content

Recent activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...
Write to the Help Desk