• Al-Holou, Wajd
• Adapa, Arjun
• Kondepudi, Akhil
• Sagher, Oren
• Heth, Jason
• Reinecke, David
• Widhalm, Georg
• Golfinos, John
• Aabedi, Alexander
• Von Spreckelsen, Niklas
• Berger, Mitchell
• Hervey-Jumper, Shawn
• Wadiura, Lisa I.
• Hollon, Todd
• Freudiger, Christian
• Lowenstein, Pedro
• Castro, Maria
• Lee, Honglak
• Snuderl, Matija
• Camelo-Piragua, Sandra
• Nasir-Moin, Mustafa
• Chowdury, Asadur
• Neuschmelting, Volker
• Jiang, Cheng

Abstract

<jats:title>Abstract</jats:title><jats:p>Molecular classification has transformed the management of brain tumors by enabling more accurate prognostication and personalized treatment. However, timely molecular diagnostic testing for brain tumor patients is limited [1–3], complicating surgical and adjuvant treatment and obstructing clinical trial enrollment [4]. Here, we developed DeepGlioma, a rapid (&lt;90 seconds), AI-based diagnostic screening system to streamline the molecular diagnosis of diffuse gliomas. DeepGlioma is trained using a multimodal dataset that includes stimulated Raman histology (SRH), a rapid, label-free, non-consumptive, optical imaging method [5–7], and large-scale, public genomic data. In a prospective, multicenter, international testing cohort of diffuse glioma patients (N = 153) who underwent real-time SRH imaging, we demonstrate that DeepGlioma can predict the molecular alterations used by the World Health Organization (WHO) to define the adult-type diffuse glioma taxonomy (IDH mutation, 1p19q co-deletion, ATRX mutation) [8], achieving a mean molecular classification accuracy of 93.3 (±1.6)%. Our results represent how artificial intelligence and optical histology can be used to provide a rapid and scalable adjunct to wet lab methods for the molecular diagnosis of diffuse glioma patients.</jats:p>

Topics

• impedance spectroscopy