{
  "abstract": "Glioblastoma, one of the most aggressive forms of brain cancer affecting the elderly and children indescriminately requires advanced computational techniques to achieve accurate and efficient classification for improved patient outcomes. This study explores the application of Quantum Boltzmann Machines (QBMs), a quantum-enhanced variant of classical Boltzmann Machines, to classify glioblastoma from multimodal datasets including imaging, spectroscopy, and molecular profiles. QBMs leverage quantum computing’s ability to represent and sample from complex probability distributions, offering a computational advantage in optimizing high-dimensional data. This study demonstrates and presents QBMs and how they may outperform classical models in accuracy and training efficiency, particularly useful for distinguishing glioblastoma from normal and marginal tissues. Preliminary results highlight the potential of quantum-enhanced machine learning to revolutionize cancer diagnostics by enabling faster and more precise classifications. Our future work will involve scaling QBMs to larger datasets, integrating multimodal inputs, and validating performance in clinical workflows, paving the way for quantum technologies in precision oncology.jdavids@ic.ac.uk",
  "authors": [
    {
      "affiliations": [
        "Imperial College London"
      ],
      "name": "Davids Joseph"
    },
    {
      "affiliations": [
        "Imperial College London"
      ],
      "name": "Alalade Andrew"
    },
    {
      "affiliations": [
        "Imperial College London"
      ],
      "name": "Ashrafian Hutan"
    }
  ],
  "title": "246 Feasibility of quantum computing and quantum boltzmann machines for glioblastoma classification",
  "uid": "a9c0b98d-618d-5280-b1b7-03e9c235db3c"
}