Artificial intelligence for management of patients with intracranial neoplasms

Although rare, intracranial neoplasms are associated with high morbidity and often poor prognosis. After a brief epidemiologic introduction, artificial intelligence (AI) applications in the fields of neurosurgery and neuro-oncology are reviewed, with a focus on machine learning (ML). AI can play an important role in the diagnostic process of brain tumors, through imaging-related applications, from segmentation to prediction of clinical features and patient outcome. AI decision support systems are promising to aid the physician in defining the best treatment options, based on predicted outcomes. Important technological advances have provided neurosurgeons with innovative equipment that can assist in surgical resection of brain lesions: while neuronavigation is now standard for most procedures, new systems can help differentiate neoplastic tissue from normal brain parenchyma and robotics has found specific applications. Assessment of prognosis through ML algorithms has been shown to be at least as accurate as normally used prognostic indices and the opinion of clinical experts. Although extremely promising, AI applications in neurosurgical practice still present several limitations—from quantity and quality of training data, to concerns of ethical implications—highlighting the need for continued research in this growing field. This chapter provides an overview of the applications AI and ML in the habitual steps of clinical management of a patient with an intracranial neoplasm, discussing the present and future AI tools available to assist diagnosis, treatment, and prognosis.