Development of a Diagnostic Tool for Identifying Genetic, Metabolic and Histopathologic Properties of Glial Brain Tumors

Principal Investigator: Alp Dinçer

Co-Investigators: Esin Öztürk Işık, Koray Özduman, Alpay Özcan, Özge Can

Researchers: Abdullah Baş, Asım Samlı, Ayhan Gürsan, Banu Saçlı Bilmez, Buse Buz Yaluğ, Cansu Akın Levi, Ceren Aslı Kaykayoğlu, Esra Sümer, Gökçe Hale Hatay, Hande Halilibrahimoğlu, Korhan Polat, Oğuzhan Aslan, Öyküm Nağme Renkli, Seda Keskin, Sena Azamat

Abstract
For increasing life quality of patients with glioma, which on average has 15 months of survival rate when diagnosed with the worst type, the most important factor is deciding with accuracy on the therapeutic path that will cause minimal discomfort to the patient. Magnetic resonance imaging (MRI) with its high soft tissue contrast characteristics is the most effective technique in determining the cancer type without a need for invasive approaches like biopsy. However, single MR modalities remain insufficient when faced with the complex structure of the disease. Accordingly, for taking full advantage of the soft tissue contrast information provided by multimodal MR, the aim of this project is to combine multimodal information for applying machine learning methodologies with the aim of differentiating glioblastoma types. For this purpose, MR, genetic and metabolomic data obtained from a prospective cohort were analyzed and machine learning methods were developed for predicting survival rate by only using the MR data. The genetic and metabolomic data obtained from biopsy were also included in the ground truth information.

Keywords: Glioma, Magnetic Resonance Imaging, Magnetic Resonance Spectroscopy, Genomics, Metabolomics, Machine Learning

Website of the tool: https://github.com/Computational-Imaging-LAB/IRIS-MRS-AI

Objectives:

  • Optimize MR imaging methods, including pre- and post-contrast T1 and T2 weighted anatomical imaging, spectroscopy, diffusion tensor imaging, and perfusion imaging, to predict genetic disorders in tumor tissue and levels of predetermined metabolites correlated with disease behavior (e.g., IDH, TERTp, ATRX, EGFR).
  • Establish a comprehensive patient sample for reliable identification of genetic glioma subgroups using IDH and TERTp markers.
  • Determine radiologically measured metabolites (choline, creatinine, N-acetyl aspartate (NAA), glutamine, glutamate, 2-hydroxyglutarate (2HG), lactate) in gliomas through MR spectroscopy (MRS) examination using mass spectrometry from intraoperative tissue biopsies, a novel approach not seen in existing literature.
  • Optimize the MRS technique for measuring the above-mentioned metabolites, increasing reproducibility, objectivity, and reliability. Standardization should be done based on in-vitro (phantom-based) and in-vivo (measured in patients' tumors) metabolites, measuring repeatability and reliability.
  • Quantify "biological criteria" (e.g., vessel number, mitotic index, necrosis, mitosis) of findings in MRI modalities other than MRS (e.g., diffusion tensor imaging, perfusion MRI, anatomical MRI) in histopathological studies of the entire patient cohort.
  • Use machine learning algorithms and bioinformatics methods to integrate optimized MRI methods in a multimodal approach that effectively demonstrates the biological behavior of the disease and patient survival.
  • Determine parameter groups with the highest predictive power for disease survival and treatment response.
  • Multimodally integrate "biomarkers" from different MR modalities to effectively predict treatment response and patient survival.
  • Identify multimodal MR biomarker measurement "patterns" that correspond to different disease biology. Correlate these patterns with histopathological diagnoses and genetic groups.
  • In the final stpe, develop a tool that independently and automatically makes disease treatment more effective and safe, using the produced information, learns from each new clinical sample to increase reliability, and provides objective information about disease biology to guide clinicians in treatment decisions. The tool should function independently of the radiologist but be available for use by clinicians, assisting them with objective information for treatment planning.

Results:

  • Anatomical MR Imaging

    By employing the selected features obtained from the feature selection algorithm, rather than using all features, the estimation of isocitrate dehydrogenase (IDH), telomerase reverse transcriptase promoter (TERTp), ATRX, PTEN mutation states, and 1p/19q codeletion exhibited improved accuracy rates. Notably, the highest accuracy rate of 87.2% was achieved in predicting the IDH mutation status, using only 8 features

  • Diffusion Tensor Imaging

    The estimation of IDH-TERT subgroups, IDH mutation status, and TERT mutation status was conducted based on diffusion anisotropy and eigenvalue distributions. Additionally, classifications were performed using distribution parameters to observe their impact on the classification process. Remarkably, all distributions significantly improved the classification performance. The accuracy rates in the IDH-TERT classification, which initially ranged from 50% to 60%, improved to the 70% range with the use of feature selection. Similarly, for IDH and TERT estimations, which initially showed accuracy rates in the 70-80% range without feature selection, the rates further increased up to 90%.

  • Perfusion MRI

    In our study, rCBV maps formed by the differentiation of blood volume in the brain were examined by using masks obtained from both post contrast T1 and T2 weighted imaging. The IDH mutation status was classified based on the features extracted from the perfusion values. The analysis revealed that all features, along with various methods, had an impact on the classification. The accuracy rates in classifying IDH mutation using perfusion MRI ranged from 70% to 80%. By employing a high-rate sampling method to balance the number of patients with and without IDH mutation, the accuracy rate increased to 86%. 

  • Magnetic Resonance Spectroscopy
    • Similar metabolic profiles were observed in TERTp only mutated gliomas, and TERTp-wt and IDH-wt gliomas. However, TERTp-mutant gliomas displayed significantly higher levels of glutamate (Glu) and glutathione (GSH). In addition to the previously reported 2HG, our study identified glycine (Glyc), GSH, Glu, and glutamine (Gln) peaks as important differentiators in distinguishing between IDH and TERTp subgroups. 
    • Differences between short echo-time point resolution spectroscopy (Point RESolved Spectroscopy (PRESS)) and 68 ms Mescher–Garwood PRESS (MEGA-PRESS) in terms of IDH mutation detection in gliomas were investigated. Our results showed that PRESS with short echo time outperformed IDH mutation classification results based on MR spectral profiles from the two spectral data acquisition techniques.
    • In another study, IDH and TERTp mutations were tried to be identified by deep learning methods. As a result, the one-dimensional convolutional neural network (1D-CNN) architecture, which gives good results in time series, has shown higher success than Long Short Term Memory (LSTM) and Recurrent Neural Network (RNN) architectures applied to other time series problems. Hyperparameter optimization methods were used to reach the full potential of the models. Finally, IDH mutation was detected with an accuracy of 94.11%, TERTp mutations were detected with an accuracy of 76.92%, and patients with only TERTp mutations were detected with an accuracy of 82.05%, regardless of the IDH mutation status.

 

 Figure. IDH-wt / TP53-wt GBM. Voxel selection on T2-weighted MRI (a), MR spectroscopic data (b) and IDH (c), Ki-67 (d) and hematoxylin and eosin (e) staining with LCModel results.

Figure. MEGA-PRESS difference (a) and short TE PRESS (c) spectra and LCModel analysis of voxel locations (b, d) for an IDH-mut grade III glioma.

  • Comparison of Mass Spectroscopy and MRS Results

    The correlation values of the metabolites examined in the mass spectrometry study, and it was observed that the 2-hydroxyglutarate (2HG), creatine (Cr), myo-inostol (Ins), lactate (Lac) and N-acetyl aspartic acid (NAA) values obtained by both methods were correlated. While the IDH mutation could be detected with an accuracy of 92.42% by mass spectroscopy, the accuracy rate was found to be 82.94% in classification with MR spectroscopy.

A graph with red and black lines</p> <p>Description automatically generated

Figure. Box plot of metabolite concentrations determined by mass spectroscopy of the IDH-mut and IDH-wt groups

A graph with red and black linesDescription automatically generated

Figure. Box plot of metabolite concentrations determined by MRS of the IDH-mut and IDH-wt groups

  • Glioma Genetic Mutation Classification Tool

    Glioma Genetic Mutation Classification Tool aims to identify IDH and TERTp genetic mutations in gliomas using non-invasive MR data. This innovative software fills the gap in utilizing machine learning for interpreting MR modalities in clinical processes. With the support and follow-up provided, the program enables genetic mutation prediction before surgery. Beyond pre-defined models for glioma genetic mutation classification, the software includes an expert user section. In this section, users can train models with various classical machine learning algorithms using their own data, import them, and perform new classifications. The expert user section also offers auxiliary methods such as feature selection and synthetic data generation, all efficiently handled by a comprehensive trainer. The user-friendly interface ensures that users can produce effective models with minimal effort. This adaptability guarantees the software's viability and responsiveness to changing conditions or demands.

Figure. MRS module

Figure. MRS module interface

Figure. Visualization outputs of the tool.

DISSEMINATION

Organized Workshops

‘Magnetic Resonance Spectroscopy Training’, organized with Gokce Hale Hatay, Banu Sacli Bilmez and Abdullah Bas. Bogazici University, March 19, 2022

Publications

  1. Hangel G, Schmitz-Abecassis B, Sollmann N, Pinto J, Arzanforoosh F, Barkhof F, Booth T, Calvo-Imirizaldu M, Cassia G, Chmelik M, Clement P, Ercan E, Fernández-Seara MA, Furtner J, Fuster-Garcia E, Grech-Sollars M, Guven NT, Hatay GH, Karami G, Keil VC, Kim M, Koekkoek JAF, Kukran S, Mancini L, Nechifor RE, Özcan A, Ozturk-Isik E, Piskin S, Schmainda KM, Svensson SF, Tseng CH, Unnikrishnan S, Vos F, Warnert E, Zhao MY, Jancalek R, Nunes T, Hirschler L, Smits M, Petr J, Emblem KE. (2023) Advanced MR Techniques for Preoperative Glioma Characterization: Part 2. J Magn Reson Imaging. 2023 Jun;57(6):1676-1695. doi: 10.1002/jmri.28663.
  2. Hirschler L, Sollmann N, Schmitz-Abecassis B, Pinto J, Arzanforoosh F, Barkhof F, Booth T, Calvo-Imirizaldu M, Cassia G, Chmelik M, Clement P, Ercan E, Fernández-Seara MA, Furtner J, Fuster-Garcia E, Grech-Sollars M, Guven NT, Hatay GH, Karami G, Keil VC, Kim M, Koekkoek JAF, Kukran S, Mancini L, Nechifor RE, Özcan A, Ozturk-Isik E, Piskin S, Schmainda K, Svensson SF, Tseng CH, Unnikrishnan S, Vos F, Warnert E, Zhao MY, Jancalek R, Nunes T, Emblem KE, Smits M, Petr J, Hangel G. (2023) Advanced MR Techniques for Preoperative Glioma Characterization: Part 1. J Magn Reson Imaging. 2023 Jun;57(6):1655-1675. doi: 10.1002/jmri.28662. 
  3. Sacli-Bilmez, B., Danyeli, A. E., Yakicier, M. C., Aras, F. K., Pamir, M. N., Özduman, K., Dincer, A., & Ozturk-Isik, E. (2023). Magnetic resonance spectroscopic correlates of progression free and overall survival in “glioblastoma, IDH-wildtype, WHO grade-4”. Frontiers in Neuroscience, 2023 Jun 29;17:1149292. doi: 10.3389/fnins.2023.1149292. 
  4. Halilibrahimoğlu H., Polat K., Keskin S., Genç O., Aslan O., Ozturk-Isik E., Yakıcıer C., Danyeli A. E., Pamir M. N., Özduman K., Dinçer A., & Özcan A. (2023) Associating IDH and TERT Mutations in Glioma with Diffusion Anisotropy in Normal-Appearing White Matter. AJNR American Journal of Neuroradiology. 2023 May; 44(5):553-561. doi: 10.3174/ajnr.A7855.
  5. Sacli-Bilmez, B., Firat, Z., Topcuoglu, O. M., Yaltirik, K., Ture, U., & Ozturk-Isik, E. (2023). Identifying overall survival in 98 glioblastomas using VASARI features at 3T. Clinical Imaging, 93, 86-92. doi:https://doi.org/10.1016/j.clinimag.2022.10.011. 
  6. Ozturk-Isik, E., Cengiz, S., Ozcan, A., Yakicier, C., Danyeli, AE., Pamir, M. N., et al.. (2019). Identification of IDH and TERTp mutation status using 1H-MRS in 112 hemispheric diffuse gliomas. Journal of Magnetic Resonance Imaging. 51(6), 1799-1809. Retrieved from https://doi.org/10.1002/jmri.26964

Invited Speeches

  1. Ozturk-Isik E. Metabolites by MRS/I. ISMRM 2023. Educational session: ‘Quantifying Spins from Head to Toe’. June 3, 2023 (invited talk)
  2. Ozturk-Isik E. Imaging Physics - MRI of Gliomas. SPARK ACADEMY AFRICA-BRATS BRAINHACK 2023. May 24, 2023 (invited talk)
  3. Ozturk-Isik E. Detection of Genetic Mutations in Brain Tumors using Advanced Magnetic Resonance Imaging-based Machine Learning. Turkish Health Institutes (TÜSEB), Turkish Cancer Institute (TKE), Innovative Diagnostic Methods in Cancer Seminar. April 6, 2023. (invited talk)
  4. Ozturk-Isik E. Advanced MRI Techniques for Glioma Imaging. The Collaborative Glioma Imaging and Clinical Management Symposium for African Scientists and Clinicians. February 13, 2023 (invited talk)
  5. Ozturk-Isik E. The Role of Machine Learning in Brain Tumor Classification and Detection of Genetic Mutations. 18th Congress of Neurosurgery. Antalya, Turkey, October 27, 2022 (invited talk)
  6. Ozturk-Isik E. Pearls of Dublin I: Genomics – Integrating Genomics into Neuro-oncology. GliMR 2.0 3rd Annual Meeting. Kuşadası, Izmir, Turkey, September 28, 2022 (invited talk)
  7. Ozturk-Isik E. Genomics – Integrating Genomics into Neuro-oncology. GliMR Training School – Artificial Intelligence in Neuro-Oncology. Dublin, Ireland, July 25, 2022 (invited talk)
  8. Sümer E., Radiogenomics applied (Hands-on)–Pyradiomics, GliMR Training School, July 25 2022, Dublin, Ireland. 
  9. Sümer E, A Beginner's Guide to Radiomic Analysis, TMRD 2022, May 27 2022. (invited talk)
  10. Ozturk-Isik E. How do Radiologists and Engineers Collaborate? An Engineer's Perspective. TMRD 2022. May 26, 2022. (invited talk)
  11. Ozturk-Isik E. How Can AI Help for MRS? ISMRM 2022. Educational session: ‘Steady State MRS’. London, UK, May 8, 2022 (invited talk)
  12. Ozturk-Isik E. Innovative Approaches in Medical Imaging. March 30, 2022. Isik University, Biomed IV. (invited talk)
  13. Ozturk-Isik E. Machine Learning on MRI Images for the Detection of Genetic Mutations in Brain Tumors. TÜSEB Artificial Intelligence in Healthcare Mini Symposium Series – 6. December 29, 2021 (invited talk)
  14. Ozturk-Isik E. Machine Learning-Based Classification of Genetic Mutations in Brain Tumors for Determining Prognosis: MR Spectroscopic Imaging Analysis. Acıbadem Üniversitesi Nörolojik Bilimler Toplantısı. January 25, 2020. (invited talk)
  15. Ozturk-Isik E. The Role of Magnetic Resonance Spectroscopic Imaging in Glioma Diagnosis. Turkish Neurosurgery Society NOVA 2019. Istanbul, Turkey, December 21, 2019. (invited talk)
  16. Ozturk-Isik E. MRS in brain tumours: The technical perspective. Glioma MR imaging 2.0 Action meeting. Malta. December 12, 2019. (invited talk)

Conference Proceedings

International Proceedings:

  1. Sacli-Bilmez B, Danyeli AE, Yakicier C, Pamir MN, Ozduman K, Dincer A, Ozturk-Isik E.  Magnetic Resonance Spectroscopic Correlates of Survival in IDH-wt, TERTp-mut Gliomas. International Society for Magnetic Resonance in Medicine. Toronto, Canada, June 3-8, 2023 (digital poster)
  2. Azamat S, Buz-Yalug B, Ozcan A, Ersen Danyeli A, Pamir MN, Dinçer A, Ozduman K, Ozturk-Isik E. Boosting The Deep Learning Performance in Predicting IDH Mutation in Gliomas Using Multiparametric MRI Including SWI, FLAIR and CE-T1WI. International Society for Magnetic Resonance in Medicine. Toronto, Canada, June 3-8, 2023 (digital poster)
  3. Azamat S, Buz-Yalug B, Ozcan A, Ersen Danyeli A, Pamir MN, Dinçer A, Ozduman K, Ozturk-Isik E. A Pretrained CNN Model Using Multiparametric MRI to Identify WHO Tumor Grade of Meningiomas. International Society for Magnetic Resonance in Medicine. Toronto, Canada, June 3-8, 2023 (oral presentation)
  4. Sümer E, Danyeli A. E., Pamir MN, Özduman K, Dinçer A, Ozturk- Isik E. Prediction of Astrocytoma Pathological Grade Using Radiomics Extracted from Pre-operative Multiparametric MRI. International Society for Magnetic Resonance in Medicine. Toronto, Canada, June 3-8, 2023 (digital poster)
  5. Buz-Yalug B, Danyeli AE, Yakicier C, Pamir MN, Ozduman K, Dincer A, Ozturk-Isik E. Relative Cerebral Blood Volume Differences Between Adult-Type Diffuse Glioma Subtypes According to WHO 2021. International Society for Magnetic Resonance in Medicine. Toronto, Canada, June 3-8, 2023 (digital poster)
  6. Bas A, Sacli-Bilmez B, Buz-Yalug B, Sumer E, Azamat S, Hatay GH,  Danyeli AE, Can O, Ozduman K, Dincer A, and Ozturk-Isik E. IRIS-DL: A Deep Learning Software Tool for Identifying Genetic Mutations in Gliomas and Meningiomas. International Society for Magnetic Resonance in Medicine. Toronto, Canada, June 3-8, 2023 (oral presentation)
  7. Bas A., Sacli-Bilmez B, Danyeli A. E., Can O., Ozduman K., Dincer A., Ozturk-Isik E. Attention Deep-Shallow Network (ADSN): A Deep Learning Model for IDH and TERTp Mutation Detection in Gliomas using 1H-MRS. International Society for Magnetic Resonance in Medicine. Toronto, Canada, June 3-8, 2023 (digital poster)
  8. Azamat S, Buz-Yalug B, Ozcan A, Danyeli AE, Pamir MN, Dincer A, Ozduman K, Ozturk-Isik E. Can Deep Learning Based on Multiparametric MRI Distinguish IDH Mutation in Gliomas? GliMR 3rd Annual Meeting. Kuşadası, Turkey, September 28-30, 2022 (oral presentation)
  9. Guven NT, Dinçer A, Tanrıkulu B, Ozturk-Isik E. Non-Invasive Prediction of Survival Times of Pediatric Diffuse Midline Glioma Patients with H3K27M Mutation Using MRI Radiomics Features. GliMR 3rd Annual Meeting. Kuşadası, Turkey, September 28-30, 2022 (oral presentation)
  10. Buz-Yalug B, Dindar SS, Danyeli AE, Yakicier MN, Pamir MN, Ozduman K, Dincer A, Ozturk-Isik E. Identification of IDH and TERTp mutations of Gliomas Using DSC-MRI  and Deep Learning. GliMR 3rd Annual Meeting. Kuşadası, Turkey, September 28-30, 2022 (oral presentation)
  11. Bas A, Sacli-Bilmez B, Hatay GH, Ozcan A, Levi C, Danyeli AE, Can O, Yakicier C, Pamir MN, Ozduman K, Dincer A, Ozturk-Isik E. IRIS-MRS-AI and IRIS-DL. GliMR 3rd Annual Meeting. Kuşadası, Turkey, September 28-30, 2022 (oral presentation) 
  12. Çetin Aİ, Danyeli AE, Pamir MN, Ozduman K, Dincer A, Ozturk-Isik E. Tumor Segmentation in Glioma with Deep Learning on T2-Weighted MRI. GliMR 3rd Annual Meeting. Kuşadası, Turkey, September 28-30, 2022 (oral presentation)
  13. Turhan G, Ozturk-Isik E. Classification of Histopathological Grades of Glial Tumors on Arterial Spin Labeling MRI Using Spatio-temporal Convolutional Neural Networks. GliMR 3rd Annual Meeting. Kuşadası, Turkey, September 28-30, 2022 (oral presentation) 
  14. Samli A, Sümer E, Danyeli AE, Pamir MN, Dincer A, Ozduman K, Ozturk-Isik E. The Effect of Super Resolution on Radiomics Analysis in Gliomas. GliMR 3rd Annual Meeting. Kuşadası, Turkey, September 28-30, 2022 (oral presentation)
  15. Sacli-Bilmez B, Bas A, Danyeli AE, Pamir MN, Ozduman K, Dincer A, Ozturk-Isik E. 1D-CNN models for IDH and TERTp Mutation Detection in Diffuse Gliomas using Proton Magnetic Resonance Spectroscopy. GliMR 3rd Annual Meeting. Kuşadası, Turkey, September 28-30, 2022 (oral presentation)
  16. Sümer E, Ozturk-Isik E, A radiomics pipeline for neuro-oncological research. GliMR 3rd Annual Meeting. Kuşadası, Turkey, September 28-30, 2022 (oral presentation)
  17. Sümer E, Danyeli AE, Pamir MN, Dinçer A, Özduman K, and Ozturk-Isik E, Radiomics for Prediction of Grades of Intracranial Astrocytomas from T2-weighted MRI. GliMR 3rd Annual Meeting. Kuşadası, Turkey, September 28-30, 2022 (oral presentation)
  18. Hatay GH, Dincer A, Tanrikulu B, Ozturk-Isik E. Survival Analysis of Pediatric DMG Patients using MRSI. GliMR 3rd Annual Meeting. Kuşadası, Turkey, September 28-30, 2022 (oral presentation)
  19. Buz-Yalug B, Ersen Danyeli A, Yakicier C, Pamir MN, Ozduman K, Dincer A, Ozturk-Isik E. Differentiation of IDH and TERTp mutations in Glioma Using Dynamic Susceptibility Contrast MRI with Machine Learning at 3T. International Society for Magnetic Resonance in Medicine. London, UK, May 7-12, 2022 (digital poster)
  20. Halilibrahimoglu H, Kaykayoglu A, Bas A, Ozduman K,Yakicier C, Ersen Danyeli A, Pamir MN, Dincer A, Ozturk–Isik E, Ozcan A, Using Transfer Learning for IDH Mutation Prediction in Gliomas Using Whole Brain Diffusion Anisotropy Indices. The European Society for Magnetic Resonance in Medicine and Biology. Virtual Meeting, Oct 7-9, 2021, p.183-184. (digital poster)
  21. Halilibrahimoglu H, Polat K, Keskin S, Ozduman K,Yakicier C, Ersen Danyeli A, Pamir MN, Dincer A, Ozturk–Isik E, Ozcan A, Predicting Glioma Genotype Using n-Component Gaussian Distributions of Diffusion Anisotropy Distributions of Normal-Appearing White Matter. The European Society for Magnetic Resonance in Medicine and Biology. Virtual Meeting, Oct 7-9, 2021, p.192-193. (digital poster)
  22. Bas A, Sacli-Bilmez B, Danyeli AE, Yakicier C, Pamir MN, Ozduman K, Dincer A, Ozturk-Isik E. The Effect of Cramer-Rao Lower Bound Thresholds on Classication of IDH and TERTp Mutation Status in Gliomas using 1H-MRS. International Society for Magnetic Resonance in Medicine. Vancouver, Canada May 15-20, 2021, p.952. (digital poster)
  23. Bas A, Sacli-Bilmez B, Danyeli AE, Yakicier C, Pamir MN, Ozduman K, Dincer A, Ozturk-Isik E. 1D-CNN for the Detection of IDH and TERTp Mutations in Diffuse Gliomas using Proton Magnetic Resonance Spectroscopy. International Society for Magnetic Resonance in Medicine. Vancouver, Canada May 15-20, 2021, p.957. (digital poster)
  24. Bas A, Sacli-Bilmez B, Hatay GH, Ozcan A, Levi C, Danyeli AE, Can O, Yakicier C, Pamir MN, Ozduman K, Dincer A, Ozturk-Isik E. Glioma Genetic Diagnosis Software for Detection of IDH and TERTp Mutations based on 1H MR Spectroscopy and Mass Spectrometry. International Society for Magnetic Resonance in Medicine. Vancouver, Canada May 15-20, 2021, p.932. (digital poster)
  25. Buz-Yalug B, Ersen Danyeli A, Yakicier C, Pamir MN, Ozduman K, Dincer A, Ozturk-Isik E. Identification of IDH and TERT Mutation Status in Glioma Patients using Dynamic Susceptibility Contrast MRI. International Society for Magnetic Resonance in Medicine. Virtual Meeting, May 15-20, 2021, p.1075. (digital poster)
  26. Sacli-Bilmez B, Ersen Danyeli A, Ekşi MŞ, Tan K, Can Ö, Yakicier C, Pamir MN, Dincer A, Ozduman K, Ozturk-Isik E. Correlations of Single Voxel 1H-MRS Findings with Tumor Biology in Meningiomas. International Society for Magnetic Resonance in Medicine. Virtual Meeting, May 15-20, 2021, p.954. (digital poster)
  27. Sacli-Bilmez B, Ersen Danyeli A, Yakicier C, Pamir MN, Ozduman K, Dincer A, Ozturk-Isik E. Lactate and glutathione levels detected with proton MR spectroscopy are associated with poor survival in IDH wild type TERTp mutant diffuse gliomas. SNO-NCI Joint Symposium: Targeting CNS Tumor Metabolism, Virtual Meeting, 6-7 April, 2021, p.BIMG-15. (oral presentation) 
  28. Sacli-Bilmez B, Akin-Levi C, Ersen Danyeli A, Yakicier C, Pamir MN, Ozduman K, Dincer A, Can Ö, Ozturk-Isik E.  Identification of IDH mutation status using proton MR spectroscopy and mass spectrometry: a study of 178 gliomas.  SNO-NCI Joint Symposium: Targeting CNS Tumor Metabolism, Virtual Meeting, 6-7 April, 2021, p.BIMG-14. (pre-recorded presentation)    
  29. Aras FK, Halilibrahimoglu H, Renkli N, Kaykayoglu A, Ozduman K, Ozturk-Isik E, Pamir N, Dincer A, Ozcan A. Investigating Distal ADC Distribution for Radiogenomic Classification of Gliomas via Machine Learning. European Society of Magnetic Resonance in Medicine and Biology Annual Conference. Virtual Meeting, September 30 – October 2, 2020. (digital poster)
  30. Ozturk-Isik E, Sacli-Bilmez B, Danyeli AE, Ozcan A, Yakicier C, Pamir MN, Ozduman, Dincer A. The effect of Tumor Grade within IDH Wild-Type and IDH Mutant Gliomas Assessed by Proton Magnetic Resonance Spectroscopy at 3T. International Society for Magnetic Resonance in Medicine. Virtual Meeting, August 8-14, 2020. (oral presentation)
  31. Sacli-Bilmez B, Gursan A, Danyeli AE, Yakicier C, Pamir MN, Ozduman, Dincer A, Ozturk-Isik E. MR Spectroscopic Differences of Low and High Grade TERTp-only Gliomas. International Society for Magnetic Resonance in Medicine. Virtual Meeting, August 8-14, 2020. (digital poster)
  32. Sacli-Bilmez B, Firat Z, Topcuoglu M, Yaltirik CK, Ture U, Ozturk-Isik E. Identifying Overall Survival in Glioblastoma Patients Using VASARI Features at 3T. International Society for Magnetic Resonance in Medicine. Virtual Meeting, August 8-14, 2020, p.422. (oral presentation)
  33. Gursan A, Sahin H, Altun B, Talas AT, Hatay GH, Kocaturk O, Garipcan B, Dincer A, Ozturk-Isik E. An MRS Phantom Design with Multiple Compartments for Mimicking IDH Mutant and IDH Wild-Type Brain Tumors. European Society of Magnetic Resonance in Medicine and Biology Annual Conference. Rotterdam, Netherlands, October 3-5, 2019. (digital poster)
  34. Gursan A, Hatay GH, Yakcer C, Pamir MN, Ozduman K, Dincer A, Ozturk-Isik E. Comparison of MEGA-PRESS and Short Echo Time PRESS on Classification of IDH Mutation Using Machine Learning at 3T. International Society for Magnetic Resonance in Medicine. Montreal, Canada, May 11-16, 2019. (digital poster)
  35. Halilibrahimoglu H, Polat K, Keskin S, Aslan O, Genc O, Ozduman K, Yakicier C, Ozturk-Isik E, Pamir MN, Dincer A, Ozcan A. Testing Machine Learning Algorithms using Anisotropy Indices of Normal Appearing White Matter as Predictors of Molecular Group of Gliomas. International Society for Magnetic Resonance in Medicine. Montreal, Canada, May 11-16, 2019. (digital poster)
  36. Ozturk-Isik E, Cengiz S, Ozcan A, Yakicier C, Pamir MN, Ozduman K, and Dincer A. Magnetic Resonance Spectroscopic Differences of Diffuse Glioma Groups Classified by IDH and TERT Promoter Mutations at 3T. International Society for Magnetic Resonance in Medicine. Paris, France, June 16-21, 2018. (oral presentation)
  37. Ozturk-Isik E, Cengiz S, Ozduman K, Ozcan A, Yakicier C, Pamir MN, Dincer A. Prediction of IDH-Mutation Status of Diffuse-Gliomas Based on Short- Echo Time Magnetic Resonance Spectroscopy at 3T. International Society of Magnetic Resonance in Medicine Annual Conference, Honolulu, HI, USA, 22-27 April 2017. (digital poster)

National Proceedings:

  1. Ozturk-Isik E. Determination of Biomarkers in Brain Diseases by MRI based Machine Learning Methods. Turkish Society of Radiology, Antalya, Turkey, 6-9 November, 2019. (oral presentation) 
  2. Sacli-Bimez B, Firat Z, Topcuoglu M, Yaltirik CK, Ture U, Ozturk-Isik E, Predicting Surival Time of Gliomblastoma Multiforme Patients using VASARI Features. Turkish Society of Radiology, Antalya, Turkey, 6-9 November, 2019. (oral presentation) 
  3. Ozturk-Isik E., Prediction of IDH mutation status of gliomas based on MR spectroscopic imaging in clinical settings. Workshop on Recent Advances in Magnetic Resonance Imaging, 20 June 2017.İstanbul, Turkey. (oral presentation)