Jamie Norris
@jamnor23
Followers
189
Following
2K
Media
6
Statuses
68
PhD student @cdt_ai_health investigating epilepsy networks and brain stimulation with machine learning 🤖🧠
London, England
Joined March 2020
Take on the 2025 Seizure Detection Challenge and contribute to the future of epilepsy research! Submission deadline: February 16, 2025. Hosted by: https://t.co/8kwhHlZXB3 Competition details: https://t.co/2oDpSKZPwL
#AI #seizure #epilepsy #competition @AnphyLab
0
7
12
Presenting this at @mlforhc in Toronto this morning! Come and see poster 182 :) Paper:
Excited to share this preprint along with @roschkoenig, @aswinchari and the rest of the team! This work focuses on two considerations when localising the Seizure Onset Zone (SOZ) from responses to Single-Pulse Electrical Stimulation (SPES). 1/n
0
1
7
A huge thank you to @DorienBlooijs, @dora_hermes and the rest of the team for making the dataset openly available ( https://t.co/syDpg4WaZs) as well as to every patient involved Project code can be found here:
github.com
Localising the Seizure Onset Zone from Single-Pulse Electrical Stimulation Responses with a Transformer - norrisjamie23/Localising_SOZ_from_SPES
0
0
2
There’s still a long way to go before these models can be clinically useful. Future work will include: - Exploring pre-training techniques - Focusing on outcome rather than the SOZ alone - Model explainability - Generalising across centres/recording modalities …and more! n/n
1
0
2
We also found that the Transformers didn’t just benefit from seeing more channels during inference, requiring only 4 channels to match the performance of the CNN (which used 37). This demonstrated the effectiveness of cross-channel attention. 5/n
1
0
1
Results: - In this cohort of patients that underwent 0.2 Hz SPES, the divergent paradigm performed poorly. - The convergent paradigm led to better results, as the literature suggested. - Switching to a Transformer led to a statistically significant improvement. 4/n
1
0
2
2) Generalising to unseen patients: most techniques assume that the input is a fixed number of consistently ordered channels, which isn’t the case in iEEG. For this, we implemented encoder-only Transformers to extract features across arbitrary electrode numbers/placements. 3/n
1
0
1
1) Analysis paradigm: shall we classify a site from the responses that stimulating it evokes elsewhere (divergent), or from the responses at this site upon stimulation elsewhere (convergent)? We implemented an existing CNN approach and compared both paradigms. 2/n
1
0
1
Excited to share this preprint along with @roschkoenig, @aswinchari and the rest of the team! This work focuses on two considerations when localising the Seizure Onset Zone (SOZ) from responses to Single-Pulse Electrical Stimulation (SPES). 1/n
Localising the Seizure Onset Zone from Single-Pulse Electrical Stimulation Responses with a Transformer.
1
3
10
1/n Doctors spend too long doing paperwork We set out to reduce this using (shock) LLMs prompted by clinical guidelines. Our recently accepted paper focuses on generating hospital discharge summaries https://t.co/4LL4PR2HBw
#aaai2024 #springsymposiusm #clinicalfoundationmodels
2
11
44
Our #EpilepsySurgery research group @GOSH_NeuroSurg & @DNP_ICH has been alive for many years but so far, nameless and faceless. We’re excited to now have a name - caesar - and a website at https://t.co/2WYNcHCYhR. Have a look around, find out what we’re up to, get in touch!
2
5
48
So pleased I have the chance to get involved with this - excited about thinking through national strategies to maximise the use of novel technologies for epilepsy research - get in touch if you want to get involved!
The #EpilepsyResearchInstitute is pleased to announce the appointment of the theme leads who will drive the Institute’s research strategy for our Enabling Technologies and Reproduction & Hormones programmes: https://t.co/q6kpkgRNML
3
4
16
The other day I made some charts 📈 for the @polycrisis project. They were for a great piece on the globalisation debate, written by @AntheaERoberts and @nicolas_lamp. I wrote up the steps to make the chart below over here https://t.co/TTFCuxk14u 🧵Thread
3
23
66
Normative brain mapping using scalp EEG and potential clinical application https://t.co/fGORAgZzqf
0
2
5
Interested in learning about applying computational approaches to questions in clinical neurology? We are organising a one day workshop in Newcastle this October 5th. https://t.co/Cx6ikNGaKh
0
10
16
Stimulation to probe, excite, and inhibit the epileptic brain https://t.co/o9m0MbBqX1
#electricalstimulation #diagnosis #treatment #deepbrainstimulation #DBS #epilepsy #epilepsyresearch #ILAE @IlaeWeb @epilepsiajourn @WileyNeuro
1
22
27
Passive and active markers of cortical excitability in epilepsy https://t.co/NyIws8uAOa
#epilepsy #EEG #biomarkers #HFOspikes #corticalexcitability #ILAE @IlaeWeb @yesILAE @epilepsiajourn @WileyNeuro
0
13
23
🚨 STOP FIRE AND REHIRE REDUNDANCIES AT UCL 🚨 @UCL and @BidvestNoonan have told outsourced security guards that they will all be fired-and-rehired on worse conditions, with 40 guards made redundant. 🧵 A thread on what's happening and how you can help https://t.co/RoDSDbWYXH
actionnetwork.org
Outsourced UCL workers are fighting against fire-and-rehire & redundancies. We need your help. UCL and subcontractor Bidvest Noonan have told outsourced security officers that they will be facing...
3
234
228
Registration for the PitVis EndoVis 2023 MICCAI Challenge is open! The challenge pertains to workflow recognition in endoscopic pituitary surgery. Register at: https://t.co/vlvduEIiaq.
@endo_vis @MICCAI_Society @WEISS_UCL @SRV_Group @BRAIN_UCL @TouchSurgery #MICCAI2023
2
20
43
Last week @LSE_UCU published a damning report on academic casualisation with many excellent charts📈 I saw some people saying they'd like to see similar analysis for their own institution. So I went away and did that with interactive charts. https://t.co/lHwUGNjCt3 THREAD🧵:
There's a crisis of academic casualisation and precarity at the LSE. Insecure and short-term contracts are rife. We're social scientists, so we decided to research how serious the problem is. Here's a thread with our key findings, 28-page report + code repository. @ucu #ucuRISING
11
99
216