medtech in everyday life

catching a temporal lobe seizure on the Empatica E4?

I often put the E4 on as I am going to bed. But I fall asleep *really* fast, so it is rare for me to have a seizure as I am sliding into sleep.

But this time, the seizure woke me up.

Just as I was drifting off to sleep, I woke up with that uneasy feeling that comes before a focal seizure, followed by the intense, familiar and ominous chain-of-thought that I can never remember after it has happened.

The feeling afterwards is not pleasant.

Here’s the wide view of the data. The seizure happened at 11.58 on the 9th of Feb 2017. I am writing this post some days later. (Click for larger image).

You can see some EDA peaks and troughs before the seizure – not terribly big in amplitude, but focal seizures have smaller magnitude EDA changes than grand mal seizures [Poh 2011;p115].

Looking in a bit closer, I see the BVP pulsing every 7 seconds or so in the period leading up to the seizure (starts 23.50), accompanies by EDA changes. I dont know enough about BVP yet to say whether this just happens all the time….  I think about my hunch that the artery squeezed between my medulla oblongata and low-lying cerebellar tonsils is causing seizures (see scans and research paper below). I wonder if this pulsing BVP is related? But then again, maybe this pulsing is just a normal pattern.

The phenomenon of herniated cerebellar tonsils causing both scoliosis and epilepsy has been described in the literature by Narasimhan et al (10.13070/rs.en.1.818). In the case study described, the authors say: “It is debatable that if left untreated, the natural history of scoliosis would have inevitably lead to curve progression, disability from back pain, cardiopulmonary problems and psychosocial concerns.” Elsewhere more controversial research has suggested that in patients with collagen disorders, the cause of progressively worsening Chiari can be filum syndrome due to occult tethered cord. See (10.1007/s00701-005-0482-y)

The two scans above, taken 4 years apart, appear to show the cerebellar tonsils descending lower into – or towards – the foramen magnum.

Although the EEG report from 2011 says ‘epilepsy proven on telemetry’, the fact that my symptoms were judged to be ‘MRI negative’ has caused a lot of umming and ahhing over the years. However I dont think the radiologist looked at my MRI very carefully. The hippocampus appears in just two slices (I think) but to me, the second slice clearly shows one of the signs of mesial temporal scleroris: a relatively hyperintense signal on one of the hippocampi in the FLAIR image (click image to make it larger):

and possible asymmetry, both of which is are indicators of mesial temporal sclerosis associated with temporal lobe epilepsy. (10.1007/s00234-014-1397-0)

(doi.org/10.3988/jcn.2008.4.1.1) See also International League Against Epilepsy Classification.

I reflect on the likely CSF path given the asymmetrical CSF flow implied above in my case.

See research paper by Urbach et al (014): “Is the type and extent of mesial temporal sclerosis measureable?”

I look back into the E4 data. I wonder if there is anything more to see where movement artefacts are at a minimum.

This snapshot of the data, taken closer in, shows a great deal of artefact at first, but something odd happening at the time I pressed the E4 button.

I look closer in.

Um, is that my BVP amplitude dropping to zero for 2 or 3 seconds?

Next stop: I’ll be reading Garbage in, Garbage OutIdentify Blood Volume Pulse (BVP) Artifacts Before Analyzing and Interpreting BVP, Blood Volume Pulse Amplitude, and Heart Rate/Respiratory Sinus Arrhythmia Data by Peper et al (2010)  (10.5298/1081-5937-38.1.19)

The link to the data in the cloud is hereE4 worn on left arm.








Your email address will not be published. Required fields are marked *