Article Text

Download PDFPDF
P31 Optimising trajectories in computer assisted planning for cranial laser interstitial thermal therapy: a machine learning approach
  1. K Li1,
  2. VN Vakharia1,
  3. R Sparks2,
  4. LGS França1,
  5. A McEvoy3,
  6. A Miserocchi3,
  7. S Ourselin4,
  8. J Duncan1
  1. 1Department of Clinical and Experimental Epilepsy, University College London, London, UK
  2. 2Wellcome EPSRC Centre for Interventional and Surgical Sciences (WEISS), University College London, London, UK
  3. 3National Hospital for Neurology and Neurosurgery, London, UK
  4. 4School of Biomedical Engineering and Imaging Sciences, St Thomas’ Hospital, King’s College London, London, UK


Objectives Optimal trajectory planning for cranial laser interstitial thermal therapy (cLITT) in drug resistant focal mesial temporal lobe epilepsy (MTLE).

Design A composite ablation score of ablated AHC minus ablated PHG volumes were calculated and normalised. Random forest and linear regression were implemented to predict composite ablation scores and determine the optimal entry and target point combinations to maximize this.

Subjects Ten patients with hippocampal sclerosis were included.

Methods Computer Assisted Planning (CAP) cLITT trajectories were generated using entry regions that include the inferior occipital gyri (IOG), middle occipital gyri (MOG), inferior temporal gyri (ITG) and middle temporal gyri (MTG). Target points were varied by sequential erosions and transformations of the centroid of the amygdala. In total 760 trajectory combinations were generated per patient and ablation volumes were calculated based on a conservative 15 mm maximum ablation diameter.

Results Linear regression was superior to random forest predictions. Linear regression indicated that maximal composite ablation scores were associated with entry points that clustered around the junction of the IOG, MOG and MTG. The optimal target point was a translation of the centroid of the amygdala anteriorly and medially.

Conclusions Machine learning techniques accurately predict composite ablation scores with linear regression outperforming the random forest approach. Optimal CAP entry points for cLITT maximize ablation of the AHC and spare the PHG.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.