{
  "abstract": "Glioblastoma is the most common malignant primary brain tumor making up 54% of all gliomas and 16% of all primary brain tumours. As a consequence effective glioblastoma management requires precision in tumour identification and resection to maximize therapeutic outcomes while preserving healthy tissue. This study presents a robotic-assisted dual-modality framework integrating probe-based confocal laser endomicroscopy (pCLE) and Raman spectroscopy for real-time glioblastoma characterization. By combining high-resolution enhanced modality cellular imaging from p-CLE with molecular fingerprinting from Raman spectroscopy, the system enables comprehensive intraoperative tumour assessment. Artificial intelligence algorithms further enhance this approach, providing automated tissue classification and margin delineation for guided tumour resection in a go-no-go paradigm. The robotic platform ensures precise navigational trajectory planning and stability, critical for the accurate acquisition of multimodal data in complex surgical environments. Results demonstrate the framework’s capability to distinguish glioblastoma from normal and marginal tissues with high sensitivity and specificity. This innovation promises to advance neurosurgical practice by enabling AI-driven, data-centric tumour resection, ultimately improving patient outcomes and reducing recurrence rates. Future work focuses on clinical validation and scalability for widespread surgical adoption.jdavids@ic.ac.uk",
  "authors": [
    {
      "affiliations": [
        "Imperial College London"
      ],
      "name": "Davids Joseph"
    },
    {
      "affiliations": [
        "Imperial College London"
      ],
      "name": "Hill Ciaran"
    },
    {
      "affiliations": [
        "National Hospital for Neurology and Neurosurgery"
      ],
      "name": "Thorne Lewis"
    },
    {
      "affiliations": [
        "Imperial College London"
      ],
      "name": "Cartucho Jiao"
    },
    {
      "affiliations": [
        "Imperial College London"
      ],
      "name": "Alalade Andrew"
    },
    {
      "affiliations": [
        "Imperial College London"
      ],
      "name": "Ashrafian Hutan"
    },
    {
      "affiliations": [
        "Imperial College London"
      ],
      "name": "Darzi Ara"
    }
  ],
  "title": "242 Robotic assisted dual-modality characterisation of glioblastoma for AI-guided tumour resection using p-CLE and Raman Spectroscopy",
  "uid": "d89de369-6671-5be1-9140-c29a451a877c"
}