Establishment and Evaluation of Multimodal Image Recognition System of Glioma Based on Deep Learning

Research purposes: 1. To obtain the metabolic characteristics of glioma molecular imaging through a multimodal image recognition system. 2. To determine whether molecular imaging metabolic parameters can characterize the molecular typing of glioma by analyzing the relationship between metabolic parameters and tumor subtypes 3. To get metabolic classification based on metabolic parameters of glioma molecular imaging, and to identify the relationship between metabolic subtypes and surgical resection, radiotherapy and chemotherapy, and prognosis and further refine the molecular classification of glioma. Research Background: Glioma is the most common primary intracranial malignant tumor, accounting for 80% of central nervous system malignant tumors. It is highly invasive, with a surgical recurrence rate of up to 90%. The prognosis is extremely poor, which has caused a great burden. There are different molecular subtypes of glioma with distinct molecular biological characteristics, resulting in various prognosis of patients. With the continuous development of basic and clinical research of glioma and the advent of various new drugs and treatment technologies, molecular pathological diagnosis based on the individual level of glioma patients is particularly important. Clarifying the molecular pathology type before surgery will help the clinical diagnosis and prognostic judgment of glioma, and is of great significance for the optimization of treatment options. Based on the establishment of glioma molecular typing system, the project team use noninvasive molecular imaging technology to clarify the characteristics of molecular subsets of glioma based on the tumor metabolic parameters. Through combining deep learning-based target detection and image recognition with big data analysis, it has great potential in the clinical research of glioma diagnosis, prognosis and treatment options, which could provide a scientific basis for the establishment and promotion of glioma molecular analysis and recognition system.

Sex

All

Minimum age

18 Years

Maximum age

70 Years

Inclusion Criteria:

1. glioma patients confirmed by postoperatively pathology ;
2. the lesion is non-diffuse, and the tumor body, edema and surrounding normal tissue are clearly delimited;
3. capacity to give informed consent and follow study procedures.

Exclusion Criteria:

1. patients with previous treatment of glioma;
2. lack of clinical and image data or data inability to meet research needs;
3. severe cardiac dysfunction: acute decompensated heart failure and/or chronic heart failure functional class III or IV (New York Heart Association classification);
4. patients who gave up halfway

• BACKGROUNDChen W, Zheng R, Baade PD, Zhang S, Zeng H, Bray F, Jemal A, Yu XQ, He J. Cancer statistics in China, 2015. CA Cancer J Clin. 2016 Mar-Apr;66(2):115-32. doi: 10.3322/caac.21338. Epub 2016 Jan 25. PMID 26808342
• BACKGROUNDCeccarelli M, Barthel FP, Malta TM, Sabedot TS, Salama SR, Murray BA, Morozova O, Newton Y, Radenbaugh A, Pagnotta SM, Anjum S, Wang J, Manyam G, Zoppoli P, Ling S, Rao AA, Grifford M, Cherniack AD, Zhang H, Poisson L, Carlotti CG Jr, Tirapelli DP, Rao A, Mikkelsen T, Lau CC, Yung WK, Rabadan R, Huse J, Brat DJ, Lehman NL, Barnholtz-Sloan JS, Zheng S, Hess K, Rao G, Meyerson M, Beroukhim R, Cooper L, Akbani R, Wrensch M, Haussler D, Aldape KD, Laird PW, Gutmann DH; TCGA Research Network; Noushmehr H, Iavarone A, Verhaak RG. Molecular Profiling Reveals Biologically Discrete Subsets and Pathways of Progression in Diffuse Glioma. Cell. 2016 Jan 28;164(3):550-63. doi: 10.1016/j.cell.2015.12.028. PMID 26824661
• BACKGROUNDChaumeil MM, Lupo JM, Ronen SM. Magnetic Resonance (MR) Metabolic Imaging in Glioma. Brain Pathol. 2015 Nov;25(6):769-80. doi: 10.1111/bpa.12310. PMID 26526945
• BACKGROUNDla Fougere C, Suchorska B, Bartenstein P, Kreth FW, Tonn JC. Molecular imaging of gliomas with PET: opportunities and limitations. Neuro Oncol. 2011 Aug;13(8):806-19. doi: 10.1093/neuonc/nor054. Epub 2011 Jul 13. PMID 21757446