AI-Driven Approaches in Neuro-Oncology: From Imaging to Treatment

Abstract

The integration of artificial intelligence (AI) into neuro-oncology has transformed the methods used for diagnosing, predicting outcomes, and treating brain tumors. This review explores both current and emerging AI-based techniques in the neuro-oncology field, with a focus on imaging, clinical decision-making, and personalized therapies. Advanced machine learning (ML) and deep learning (DL) algorithms are enhancing the interpretation of neuroimaging by enabling precise tumor segmentation, classification, and progression monitoring. Radiomics and radio genomics further link imaging characteristics with molecular profiles, offering noninvasive diagnostic and prognostic insights. In terms of treatment planning, AI supports surgical navigation, optimizes radiation therapy, and aids in the discovery of targeted drugs. Additionally, predictive models help forecast outcomes and adaptively adjust therapies. Despite these promising developments, challenges such as data standardization, model generalizability, and ethical concerns remain. Continued interdisciplinary collaboration is essential to fully leverage AI's potential in delivering more precise, efficient, and personalized care in neuro-oncology.

Keywords

Introduction

Fig. 1: The diagnosis and treatment approaches have been significantly enhanced by the integration of artificial intelligence (AI), leading to improved patient outcomes and quality of life.

TABLE 1:  AI MODELS USED IN CLASSIFICATION OF BRAIN TUMORS

SR. NO	TYPE OF BRAIN TUMOURS	ML AND DL ALGORITHMS
1.	meningioma	Optimized ANN; Logistic regression (LR); SVM; CNN
2.	Astrocytoma	Learning Vector Quantization (LVQ); Naive Bayes; SVM
3.	Brain stem gliomas	ML; DL; k-means clustering for segmentation; SVM
4.	Ependymomas	Radiomics; CNN; ML; RF; ANN (artificial neural network); Tree bagger (TB); Gradient boosting (GB)
5.	Oligodendrogliomas	T2WI-only network (MGMT-net); 3D U-Net architecture; CNN; Random survival forest (RSF); SVM; Network-based statistics (NBS)
6.	Optic pathway gliomas	DL; U-Net+ResNet architecture
7.	oligo-astrocytomas	CNN (conventional neural network); DL (deep learning)
8.	Primitive neuroectodermal	2-phase deep learning algorithm; LR; RF; Decision tree
9.	Pituitary tumours	CRNN (DenseNet+ResNet); Support vector machine; Deep neural network (DNN); Naive Bayes; LR; RF; XGboost
10.	Pineoblastoma	radiomics; Gaussian Naïve Bayes; Support Vector Machine Classifier (SVC); Multilayer Perceptron (MLP); CatBoost
11.	Choroid plexus	DL; ChP architecture
12.	Medulloblastoma	CNN; Topographic independent component analysis (TICA); SVM

Fig. 1: Various applications of machine learning

The capacity of tailored therapy to influence particular genetic and molecular changes present in brain tumors has drawn interest in neuro-oncology in recent years. These treatments aim to stop the growth of tumors by targeting important proteins, gene mutations, or signaling pathways that contribute to the development of cancer. A number of targeted medications are now being investigated or utilized in clinical settings to treat brain tumors, including gliomas. In neuro-oncology, the main targeted therapies, their molecular targets, tumor types, mechanisms of action, and clinical status are summarized in Table 3.

3.1 personalized treatment protocols

3.1.1. Molecular and Genetic Profiling in Gliomas

Understanding the molecular and genetic details of gliomas is crucial for personalized treatment in brain cancer care. This helps doctors diagnose more accurately, predict disease progression, and select the best treatments. Key biomarkers include IDH mutations, 1p/19q co-deletion, MGMT promoter methylation, and EGFR amplification. These markers help classify gliomas and predict treatment response. IDH Mutations Changes in the IDH1 and IDH2 genes are common in lower-grade gliomas and secondary glioblastomas. Tumors with these mutations usually grow slower, and patients often live longer than those with normal IDH genes. These tumors produce a chemical called D-2-hydroxyglutarate, which affects cancer development but also offers new treatment options. Studies show that patients with IDH mutations respond better to treatments like radiation and chemotherapy, especially with drugs like temozolomide(26)1p/19q Co-deletion When both the short arm of chromosome 1 (1p) and the long arm of chromosome 19 (19q) are missing in tumor cells, it is called a 1p/19q co-deletion. This is a key feature of oligodendrogliomas. Patients with this marker respond well to combined radiation and chemotherapy and often live longer. If a tumor has both IDH mutations and this co-deletion, it provides doctors with better information about the tumor type and likely outcomes.(27) EGFR Amplification In glioblastoma multiforme (GBM), having too many copies of the EGFR gene is common. This makes the tumor grow and spread more aggressively and is linked to a worse outlook for the patient. EGFR is important in how these tumors behave and are categorized. Because of its significance, scientists are studying EGFR as a potential target for new cancer treatments.(28)

3.1.2. Advanced Neuroimaging Integration in Personalized Glioma Treatment Planning

New imaging techniques have greatly improved how doctors plan personalized treatment for people with gliomas. Tools like MRI, fMRI, DTI, and PET scans give detailed information about the tumor—such as its size, location, and how it affects important areas of the brain. Using these methods together helps doctors carefully plan surgery and radiotherapy based on each patient’s unique condition.

Magnetic Resonance Imaging (MRI)

MRI is the most commonly used imaging method in brain tumor care. It shows the size and shape of the tumor. More advanced types of MRI—like DWI, PWI, and SWI—help doctors check how dense the tumor is, how much blood it gets, and the condition of its blood vessels. These details are useful in telling whether a tumor is growing or if the changes are due to side effects of treatment like swelling or dead tissue (29)

Functional MRI (fMRI)

fMRI looks at blood flow changes in the brain to find areas involved in key functions such as speaking or moving. This helps doctors avoid damaging those areas during surgery. It also supports radiotherapy planning by making sure important parts of the brain aren’t harmed by radiation (30)

Diffusion Tensor Imaging (DTI)

DTI tracks the movement of water in the brain to show white matter pathways—like the ones used for movement and thinking. This is especially helpful for planning surgery near important brain structures. Including DTI in treatment planning helps protect the patient’s ability to move and think properly (31)

Positron Emission Tomography (PET)

PET scans look at how active the tumor cells are by using special chemicals (called tracers) that highlight living tumor tissue. This helps doctors see which parts of the tumor are still active, especially when MRI images are unclear. Using PET in planning can improve both surgery and radiation treatment by targeting the tumor more precisely (32)

Combining Multiple Imaging Methods

Using different imaging tools together gives a more complete picture of the tumor and its effect on the brain. For example, combining fMRI and DTI helps map both how the brain works and how it’s connected. PET and MRI together can more accurately show tumor borders, which is very helpful for safer and more effective treatment planning (33)

3.1.3 Biomarker-Guided Chemotherapy in Gliomas

In recent years, doctors have started using molecular biomarkers to help decide which chemotherapy treatments will work best for glioma patients. Two of the most important biomarkers are MGMT promoter methylation and IDH gene mutations. These markers not only guide treatment choices but also help predict how well a patient might respond and how long they might survive. MGMT Promoter Methylation and Response to Temozolomide: The MGMT gene plays a role in fixing damaged DNA in cells. When the promoter region of this gene is methylated (a type of chemical change), its ability to repair DNA is reduced. As a result, tumor cells with methylated MGMT are more likely to be damaged and killed by chemotherapy drugs like temozolomide (TMZ). However, MGMT methylation does not always guarantee a better response to TMZ. A study by Egaña et al. (2020) found that some patients with methylated MGMT still did not respond well to TMZ, which suggests that other biological factors might also play a role in how effective the treatment is (34)

IDH Mutations and Their Impact on Prognosis

Mutations in the IDH1 and IDH2 genes are often found in low-grade gliomas and in secondary glioblastomas. These genetic changes are usually linked to better outcomes, including longer survival and stronger responses to chemotherapy. According to Houillier et al. (2010), patients with IDH-mutant gliomas responded better to TMZ and had improved survival compared to those without the mutation (35). When both IDH mutation and MGMT promoter methylation are found in the same tumor, the prediction of treatment response becomes even more accurate. Christensen et al. (2011) showed that this combination helps doctors better estimate how patients with secondary glioblastomas will respond to TMZ therapy (26). Using molecular biomarkers like MGMT methylation and IDH mutations allows for a more personalized approach to glioma treatment. These markers help identify which patients are more likely to benefit from chemotherapy, especially temozolomide, and help improve overall treatment planning.

3.1.4 Liquid Biopsies and Monitoring in Glioma Treatment:

 Liquid biopsy is a new and less risky way to learn about brain tumors. Instead of using surgery to get tumor tissue, doctors can test body fluids like blood or cerebrospinal fluid (CSF) to find tumor-related substances such as DNA or RNA. This method is helpful because it can be done multiple times to keep track of the tumor and how it responds to treatment.

Evaluating Treatment Effectiveness Doctors can check ctDNA levels regularly to see whether a treatment is working or if it needs to be changed (36). Detecting Early Recurrence This method can sometimes find signs that the tumor is coming back before it shows up on traditional scans (37). Why CSF Might Be Better Than Blood: It is because although drawing blood is easier, CSF samples usually give more accurate results in brain tumors like gliomas. This is because the blood-brain barrier limits how much tumor DNA can move into the blood. Since CSF is closer to the brain, it often contains more tumor material (38).. Even though liquid biopsy is a promising tool, there are still some issues—such as improving the sensitivity of tests and making the methods consistent across different labs. Scientists are actively working on these problems so that liquid biopsies can become a regular part of glioma treatment in the future (39).

3.2 Artificial Intelligence in Neuro-Oncology Radiotherapy

AI in Tumor Mapping and Brain Function Preservation Artificial intelligence is significantly improving the accuracy and efficiency of identifying tumor regions for radiotherapy. Traditionally, this process was time-consuming and varied clinicians. With AI, tumor boundaries can be outlined more consistently and quickly, reducing errors and enhancing treatment planning.

Whole brain radiotherapy (WBRT) is commonly used for patients with multiple brain tumors. However, it can unintentionally affect healthy regions such as the hippocampus, a critical area for memory. Radiation to the hippocampus may lead to memory loss. To address this, deep learning algorithms have been developed to design treatment plans that avoid irradiating the hippocampus. These AI-generated plans can be created in under 10 minutes, improving both the speed and safety of care (40)

Personalized Radiotherapy for Glioblastoma:

Glioblastoma is an aggressive brain tumor that spreads rapidly into nearby tissues, making it difficult to treat using traditional methods. AI tools now combine advanced imaging techniques, such as MRI and FET-PET, with tumor growth prediction models. This integration allows doctors to better understand how a tumor may evolve and to create personalized radiation plans that effectively target the tumor while preserving surrounding healthy brain tissue. (41)

AI in Dose Prediction and Quality Assurance:

Accurate radiation dosing is crucial in cancer therapy. Deep learning models like the 3D Dense Dilated U-Net can predict how radiation doses will be distributed using CT images and existing treatment plans. These systems also help detect low-quality or potentially unsafe plans at an early stage, enabling healthcare providers to make informed adjustments (42). For quality assurance, AI has streamlined the verification process, making it faster and more reliable. Research indicates that AI tools can save around 16 minutes during plan review and detect more than half of the safety issues before treatment begins (43).

AI in Brachytherapy (Internal Radiotherapy):

AI is also being applied in brachytherapy, a form of internal radiation therapy. It assists in tumor outlining, dose calculation, and outcome prediction. These tools help reduce manual effort and accelerate the overall process. However, widespread clinical use requires thorough validation and constant human supervision to ensure patient safety (44).

Importance of Human Oversight:

Despite the many advantages AI brings to radiotherapy, it should not replace human expertise. AI systems may carry biases from their training data and lack the ability to understand a patient’s unique health needs or personal values. Human judgment remains essential to ensure that treatment plans are safe, ethical, and tailored to each individual. Radiation oncologists provide the necessary experience, insight, and empathy that AI cannot replicate (45).

3.3 AI-Assisted Surgical Planning in Brain Tumor Treatment

Surgery for brain tumors, especially glioblastoma, requires careful planning. If surgery won't improve patient outcomes, it might cause more harm than good, so doctors must assess each case individually. AI and Machine Learning tools now help doctors make these decisions by offering personalized insights beyond traditional scoring systems. ML models predict survival chances based on age, tumor type, and tumor count(46).  1. Predicting Survival and Tumor Progression AI tools are improving survival and tumor behavior prediction: Deep learning and radiomics estimate post-surgery survival in glioblastoma with 74.5% accuracy. ML models predict tumor recurrence using pre-surgery scans. Connectome-based AI predicts genetic features like IDH mutation from MRI scans, with AUC scores of 0.76-0.94. Voxel-based ML tools predict tumor regrowth locations for better surgical planning(47).  2. Assessing Whether Tumor Can Be Removed For complex cases like glioblastoma, AI systems evaluate tumor removal feasibility through MRI images. Marcus et al. developed an AI tool to rate tumor complexity, while Kommers et al. created GSI-RADS for consistent tumor descriptions(48).  3. AI for Tumor Segmentation and Resection Assessment Zanier et al. used U-Net deep learning to analyze glioma scans, mapping tumors and assessing removal extent. The RANO Resect Group advocates AI for measuring remaining tumor volume. Fathi Kazerooni et al. used AI to differentiate tumor areas for personalized removal strategies(49).  4. Using Brain Network Maps (Connectome) for Surgery Quicktome™ maps brain functions for safer surgeries. Luckett et al.'s 3D CNN models map language and movement areas with 96% accuracy. ML tools help surgeons identify critical brain areas, even when displaced by tumors(50)

4. AI FOR PRE-SURGICAL PLANNING

Once surgery is an option, AI tools make the planning process faster and more accurate:

CNN-based tools can outline important brain parts much faster than humans—up to 30 times quicker. AI can also help plan the direction and angle of radiation therapy by modelling how tissues will move during treatment. The ROBOCAST project uses AI to create detailed surgical roadmaps, already tested in procedures like brain biopsies. In summary, AI is transforming how brain tumor surgeries are planned and performed. It helps doctors predict patient outcomes, plan safe surgeries, and protect critical brain functions. As the technology grows, AI is expected to become an essential tool in providing safer and more personalized care for brain tumor patients.

5. AI IN NEUROSURGICAL ONCOLOGY

Neurosurgical oncology is a specialized discipline that integrates the expertise of neurosurgeons, oncologists, radiologists, and other specialists to deliver comprehensive care for patients with benign and malignant tumors of the brain and spine. Its primary objectives include maximizing safe tumor resection, alleviating symptoms, improving quality of life and survival outcomes, and advancing research and innovation. Surgical intervention is indicated for tissue diagnosis, cytoreduction, relief of mass effect with potential neurological improvement, and enhancement of survival and quality of life. In glioma management, MGMT gene methylation status has emerged as an important prognostic biomarker predicting responsiveness to alkylating agents. Additionally, the Response Assessment in Neuro-Oncology (RANO) criteria were developed to objectively differentiate treatment-related changes from true tumor progression in both low- and high-grade gliomas. Overall, neurosurgical oncology plays a pivotal role in glioma care through diagnosis, symptom control, and enabling direct therapeutic delivery that bypasses the blood–brain barrier.(51)

Laser interstitial thermal therapy (LITT) is an adjuvant treatment for intracranial lesions that are treatment refractory or in deep or eloquent brain. Initial studies of LITT in surgical neuro-oncology are limited in size and follow-up. The result shows A total of 91 patients underwent 100 LITT procedures; 61% remain alive with 72% local control at a median 7.2-mo follow-up. (52)

Artificial intelligence (AI) has been explored for its potential applications in the field of neurosurgery. Recent research indicates that machine learning (ML) offers considerable promise in neuro-oncological care, spine surgery, epilepsy management, and other neurosurgical domains. It is imperative for clinicians to familiarize themselves with the various forms of AI, as a comprehensive understanding of these technologies, rather than blind trust, is crucial for their appropriate and safe integration into patient care (53).

Parduzi et al. investigated artificial intelligence–based intraoperative neurophysiological monitoring (IONM) to improve preservation of motor pathways during complex neurosurgical procedures. Using machine learning and deep learning models trained on large multicenter datasets of motor-evoked potentials (MEPs), the study demonstrated high accuracy in muscle classification, with random forest and convolutional neural networks showing strong performance. Explainable AI techniques revealed that frequency components and peak latencies are key features for accurate MEP interpretation. These findings suggest that AI-assisted IONM may reduce muscle mislabeling, support the development of standardized warning criteria, and ultimately enhance intraoperative patient safety.

(54).

Deep brain stimulation (DBS) has emerged as a key neurosurgical therapy for movement and certain psychiatric disorders. Traditionally, optimal DBS target localization relies on either a direct method, which visualizes the target nucleus on MRI, or an indirect method based on geometric calculations using the anterior commissure–posterior commissure (AC–PC) reference system. However, current evidence is limited by small sample sizes, preventing definitive conclusions about the reliability and clinical superiority of AI-assisted approaches. Although AI applications in DBS are still at an early stage and lack statistically robust validation, emerging tools, including large language models such as ChatGPT and Gemini, show promise in text-based tasks. Their role in image-based analysis—critical for neurosurgical planning and target localization—remains underexplored and requires further investigation.(55)

5.1 Robot-assisted neurosurgery:

Robot-assisted neurosurgery, commonly referred to as robotic surgery, allows physicians to execute a diverse array of intricate procedures with enhanced precision, flexibility, and control compared to traditional techniques. Typically, robotic surgery is performed through small incisions to enable minimally invasive procedures within the brain, although it is occasionally utilized in open surgeries as well. Several robotic systems are employed, including

PUMA-200 (Programmable Universal Manipulation Arm or Programmable Universal Machine Assembly). The PUMA 200 was also employed for holding and manipulating surgical retractors while achieving radical excision of thalamic astrocytomas in six children in 1991.

da Vinci: This is the most widely used master-controlled console-based robotic system. Its application in neurosurgery has yet to become widespread due to factors such as the system's large size and the limited number of instruments.

SOCRATES: This was the first tele-mentoring system in neurosurgery, wherein a robotic arm is controlled by a mentor from another institution.

Robotic Stereotactic Assistance (ROSA): Also known as a robotic operating surgical assistant, this single-arm robot provides excellent dexterity and accuracy, thereby reducing operative time in procedures such as DBS, stereo electroencephalography (SEEG), and repetitive nerve stimulation (RNS), among others (56). These various systems are employed in neurosurgery to enhance the quality of surgery and improve success rates.

6. CHALLENGERS AND FUTURE DIRECTIONS

Artificial intelligence (AI) in neuro-oncology encounters several challenges, including issues related to data quality and diversity, model interpretability, ethical considerations, and the necessity for robust legal frameworks. Ethical considerations in the integration of AI for brain tumor diagnosis, prognosis, and treatment encompass data privacy, algorithmic fairness, legal liability, and social implications. This focus includes diagnostic AI models for key genomic markers, predictive AI models for response assessment before and after therapy, and the differentiation of true disease progression from treatment-related changes, which remains a significant challenge in current clinical neuro-oncology practice (7,57). Ethically, it is imperative to balance technological advancement with patient consent, dignity, and privacy, ensuring the responsible use of AI tools. Legally, AI in healthcare must navigate complex regulations governing patient data and medical decision-making. Clinicians require substantial training to interpret AI results, and these systems must undergo rigorous testing for accuracy and reliability using extensive datasets. Ultimately, the advancement of AI technology depends on continuous development and innovation to enhance its efficacy in the diagnosis, treatment, management, and research of brain tumors and other malignancies. There is an urgent need for further exploration and the development of novel algorithms and tools across various domains, including imaging techniques, computational models, and data fusion strategies (58).

7. CONCLUSION:

Artificial intelligence is transforming the detection, diagnosis, and treatment of brain cancers by improving imaging interpretation, enabling early tumor detection, and supporting personalized treatment planning. Through machine learning and deep learning, AI enhances accuracy in surgery and radiation therapy, improves safety with robotic assistance, and accelerates drug discovery by identifying patients most likely to benefit from specific therapies. While AI offers major advances in precision and efficiency of brain cancer care, its full potential requires further research, robust validation, appropriate infrastructure, and careful attention to data privacy, fairness, and transparency.

REFERENCES

1. National cancer insitute. Cancer Stat Facts: Brain and Other Nervous System Cancer. 2025. Surveillance, Epidemiology, and End Results Program. Available from: https://seer.cancer.gov/statfacts/html/brain.html#risk
2. Yuen CA, Zheng M, Saint-Germain MA, Kamson DO. Meningioma: Novel Diagnostic and Therapeutic Approaches. Biomedicines. 2025;13(3):659.
3. Power P, Straehla JP, Fangusaro J, Bandopadhayay P, Manoharan N. Pediatric neuro-oncology: Highlights of the last quarter-century. Neoplasia [Internet]. 2025;59:101098. Available from: https://www.sciencedirect.com/science/article/pii/S1476558624001398
4. Rees JH. Diagnosis and treatment in neuro-oncology: an oncological perspective. Br J Radiol. 2011;84(special\_issue\_2):S82--S89.
5. Grand S, Nedunchelian M, Charara S, Demaison R, Jean C, Galloux A, et al. Tumor or not a tumor: Pitfalls and differential diagnosis in neuro-oncology. Rev Neurol (Paris) [Internet]. 2023;179(5):378–93. Available from: https://www.sciencedirect.com/science/article/pii/S0035378723008937
6. Baker CR, Pease M, Sexton DP, Abumoussa A, Chambless LB. Artificial intelligence innovations in neurosurgical oncology: a narrative  review. J Neurooncol. 2024 Sep;169(3):489–96.
7. Khalighi S, Reddy K, Midya A, Pandav KB, Madabhushi A, Abedalthagafi M. Artificial intelligence in neuro-oncology: advances and challenges in brain tumor diagnosis, prognosis, and precision treatment. NPJ Precis Oncol. 2024;8(1):80.
8. Kale M, Wankhede N, Pawar R, Ballal S, Kumawat R, Goswami M, et al. AI-driven innovations in Alzheimer’s disease: Integrating early diagnosis, personalized treatment, and prognostic modelling. Ageing Res Rev [Internet]. 2024;101:102497. Available from: https://www.sciencedirect.com/science/article/pii/S1568163724003155
9. Khalifa M, Albadawy M. Artificial intelligence for clinical prediction: exploring key domains and essential functions. Comput Methods Programs Biomed Updat. 2024;100148.
10. Aamir A, Iqbal A, Jawed F, Ashfaque F, Hafsa H, Anas Z, et al. Exploring the current and prospective role of artificial intelligence in disease diagnosis. Ann Med Surg. 2024;86(2):943–9.
11. Neal RD, Tharmanathan P, France B, Din NU, Cotton S, Fallon-Ferguson J, et al. Is increased time to diagnosis and treatment in symptomatic cancer associated with poorer outcomes? Systematic review. Br J Cancer. 2015;112(1):S92--S107.
12. Hale AT, Stonko DP, Wang L, Strother MK, Chambless LB. Machine learning analyses can differentiate meningioma grade by features on magnetic resonance imaging. Neurosurg Focus. 2018;45(5):E4.
13. Laukamp KR, Thiele F, Shakirin G, Zopfs D, Faymonville A, Timmer M, et al. Fully automated detection and segmentation of meningiomas using deep learning on routine multiparametric MRI. Eur Radiol [Internet]. 2019;29(1):124–32. Available from: https://doi.org/10.1007/s00330-018-5595-8
14. Segato A, Marzullo A, Calimeri F, De Momi E. Artificial intelligence for brain diseases: A systematic review. APL Bioeng. 2020;4(4).
15. Chen C, Guo X, Wang J, Guo W, Ma X, Xu J. The Diagnostic Value of Radiomics-Based Machine Learning in Predicting the Grade of Meningiomas Using Conventional Magnetic Resonance Imaging: A Preliminary Study. Front Oncol [Internet]. 2019;Volume 9-. Available from: https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2019.01338
16. Moon CM, Lee YY, Kim DY, Yoon W, Baek BH, Park JH, et al. Preoperative prediction of Ki-67 and p53 status in meningioma using a multiparametric MRI-based clinical-radiomic model. Front Oncol [Internet]. 2023;Volume 13. Available from: https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2023.1138069
17. Chen B, Chen C, Wang J, Teng Y, Ma X, Xu J. Differentiation of low-grade astrocytoma from anaplastic astrocytoma using radiomics-based machine learning techniques. Front Oncol. 2021;11:521313.
18. Uddin S, Khan A, Hossain ME, Moni MA. Comparing different supervised machine learning algorithms for disease prediction. BMC Med Inform Decis Mak. 2019;19(1):1–16.
19. Zhi F, Shao N, Wang R, Deng D, Xue L, Wang Q, et al. Identification of 9 serum microRNAs as potential noninvasive biomarkers of human astrocytoma. Neuro Oncol. 2015;17 3:383–91.
20. Shirahata M, Ono T, Stichel D, Schrimpf D, Reuss DE, Sahm F, et al. Novel, improved grading system (s) for IDH-mutant astrocytic gliomas. Acta Neuropathol. 2018;136:153–66.
21. Potter N, Karakoula A, Phipps KP, Harkness W, Hayward R, Thompson DNP, et al. Genomic deletions correlate with underexpression of novel candidate genes at six loci in pediatric pilocytic astrocytoma. Neoplasia. 2008;10(8):757--IN9.
22. Li W, Wang C hua, Cheng G, Song Q. International conference on machine learning. Trans Mach Learn Res. 2023;
23. Hazarika H, Barua R, Gourisaria MK, Jena JJ, Das S, Patra SS. Detection of Glioma from Brain CT Scan images using Explainable AI based Ensemble Feature Extraction. Procedia Comput Sci. 2025;258:1877–87.
24. Chang P, Grinband J, Weinberg BD, Bardis M, Khy M, Cadena G, et al. Deep-learning convolutional neural networks accurately classify genetic mutations in gliomas. Am J Neuroradiol. 2018;39(7):1201–7.
25. Blanc-Durand P, Van Der Gucht A, Schaefer N, Itti E, Prior JO. Automatic lesion detection and segmentation of 18F-FET PET in gliomas: a full 3D U-Net convolutional neural network study. PLoS One. 2018;13(4):e0195798.
26. SongTao Q, Lei Y, Si G, YanQing D, HuiXia H, XueLin Z, et al. IDH mutations predict longer survival and response to temozolomide in secondary glioblastoma. Cancer Sci. 2012;103(2):269–73.
27. Bou Zerdan M, Assi HI. Oligodendroglioma: a review of management and pathways. Front Mol Neurosci. 2021;14:722396.
28. E Taylor T, B Furnari F, K Cavenee W. Targeting EGFR for treatment of glioblastoma: molecular basis to overcome resistance. Curr Cancer Drug Targets. 2012;12(3):197–209.
29. Martucci M, Russo R, Giordano C, Schiarelli C, D’Apolito G, Tuzza L, et al. Advanced magnetic resonance imaging in the evaluation of treated glioblastoma: A pictorial essay. Cancers (Basel). 2023;15(15):3790.
30. Lakhani DA, Sabsevitz DS, Chaichana KL, Quiñones-Hinojosa A, Middlebrooks EH. Current state of functional MRI in the presurgical planning of brain tumors. Radiol Imaging Cancer. 2023;5(6):e230078.
31. Manan AA, Yahya NA, Taib NHM, Idris Z, Manan HA. The assessment of white matter integrity alteration pattern in patients with brain tumor utilizing diffusion tensor imaging: a systematic review. Cancers (Basel). 2023;15(13):3326.
32. MiyaKe K, Ogawa D, OKaDa M, HaTaKeyaMa T, TaMiya T. Usefulness of positron emission tomographic studies for gliomas. Neurol Med Chir (Tokyo). 2016;56(7):396–408.
33. Schebesch KM, Rosengarth K, Brawanski A, Proescholdt M, Wendl C, Höhne J, et al. Clinical benefits of combining different visualization modalities in neurosurgery. Front Surg. 2019;6:56.
34. Egaña L, Auzmendi-Iriarte J, Andermatten J, Villanua J, Ruiz I, Elua-Pinin A, et al. Methylation of MGMT promoter does not predict response to temozolomide in patients with glioblastoma in Donostia Hospital. Sci Rep. 2020;10(1):18445.
35. Houillier C, Wang X, Kaloshi G, Mokhtari K, Guillevin R, Laffaire J, et al. IDH1 or IDH2 mutations predict longer survival and response to temozolomide in low-grade gliomas. Neurology. 2010;75(17):1560–6.
36. Mair R, Mouliere F. Cell-free DNA technologies for the analysis of brain cancer. Br J Cancer. 2022;126(3):371–8.
37. Soffietti R, Bettegowda C, Mellinghoff IK, Warren KE, Ahluwalia MS, De Groot JF, et al. Liquid biopsy in gliomas: a RANO review and proposals for clinical applications. Neuro Oncol. 2022;24(6):855–71.
38. De Mattos-Arruda L, Mayor R, Ng CKY, Weigelt B, Mart\’\inez-Ricarte F, Torrejon D, et al. Cerebrospinal fluid-derived circulating tumour DNA better represents the genomic alterations of brain tumours than plasma. Nat Commun. 2015;6(1):8839.
39. Gondi V, Deshmukh S, Brown PD, Wefel JS, Armstrong TS, Tome WA, et al. Sustained preservation of cognition and prevention of patient-reported symptoms with hippocampal avoidance during whole-brain radiation therapy for brain metastases: final results of NRG oncology CC001. Int J Radiat Oncol Biol Phys. 2023;117(3):571–80.
40. Lin CY, Chou LS, Wu YH, Kuo JS, Mehta MP, Shiau AC, et al. Developing an AI-assisted planning pipeline for hippocampal avoidance whole brain radiotherapy. Radiother Oncol. 2023;181:109528.
41. Lipková J, Angelikopoulos P, Wu S, Alberts E, Wiestler B, Diehl C, et al. Personalized Radiotherapy Planning for Glioma Using Multimodal Bayesian Model Calibration. CoRR [Internet]. 2018;abs/1807.0. Available from: http://arxiv.org/abs/1807.00499
42. Gronberg MP, Beadle BM, Garden AS, Skinner H, Gay S, Netherton T, et al. Deep Learning–Based Dose Prediction for Automated, Individualized Quality Assurance of Head and Neck Radiation Therapy Plans. Pract Radiat Oncol [Internet]. 2023 May;13(3):e282–e291. Available from: http://dx.doi.org/10.1016/j.prro.2022.12.003
43. Claessens M, Oria CS, Brouwer CL, Ziemer BP, Scholey JE, Lin H, et al. Quality Assurance for AI-Based Applications in Radiation Therapy. Semin Radiat Oncol. 2022 Oct;32(4):421–31.
44. Fionda B, Placidi E, de Ridder M, Strigari L, Patarnello S, Tanderup K, et al. Artificial intelligence in interventional radiotherapy (brachytherapy): Enhancing  patient-centered care and addressing patients’ needs. Clin Transl Radiat Oncol. 2024 Nov;49:100865.
45. Parikh RB, Obermeyer Z, Navathe AS. Regulation of predictive analytics in medicine. Science (80- ). 2019;363(6429):810–2.
46. Li R, Liu Z, Wei Z, Huang R, Pei Y, Yang J, et al. Machine Learning-based Prognostic Model for Brain Metastasis Patients: Insights from Blood Test Analysis. J Cancer. 16(2):689.
47. Pasquini L, Napolitano A, Lucignani M, Tagliente E, Dellepiane F, Rossi-Espagnet MC, et al. Comparison of machine learning classifiers to predict patient survival and genetics of GBM: Towards a standardized model for clinical implementation. arXiv Prepr arXiv210206526.
48. Mut M, Zhang M, Gupta I, Fletcher PT, Farzad F, Nwafor D. Augmented surgical decision-making for glioblastoma: integrating AI tools into education and practice. Front Neurol. 2024;15:1387958.
49. Zanier O, Da Mutten R, Vieli M, Regli L, Serra C, Staartjes VE. DeepEOR: automated perioperative volumetric assessment of variable grade gliomas using deep learning. Acta Neurochir (Wien). 2023;165(2):555–66.
50. Morell AA, Eichberg DG, Shah AH, Luther E, Lu VM, Kader M, et al. Using machine learning to evaluate large-scale brain networks in patients with brain tumors: traditional and non-traditional eloquent areas. Neuro-oncology Adv. 2022;4(1):vdac142.
51. Mitchell D, Shireman JM, Dey M. Surgical neuro-oncology: Management of glioma. Neurol Clin. 2022;40(2):437–53.
52. Shah AH, Semonche A, Eichberg D, Borowy V, Luther E, Sarkiss CA, et al. The Role of Laser Interstitial Thermal Therapy in Surgical Neuro-Oncology: Series of 100 Consecutive Patients. Neurosurgery. 2019;
53. Tangsrivimol JA, Schonfeld E, Zhang M, Veeravagu A, Smith TR, Härtl R, et al. Artificial intelligence in neurosurgery: a state-of-the-art review from past to future. Diagnostics. 2023;13(14):2429.
54. Parduzi Q, Wermelinger J, Koller SD, Sariyar M, Schneider U, Raabe A, et al. Explainable AI for Intraoperative Motor-Evoked Potential Muscle Classification in Neurosurgery: Bicentric Retrospective Study. J Med Internet Res [Internet]. 2025;27. Available from: https://www.sciencedirect.com/science/article/pii/S1438887125004248
55. Wu F, Bresolin N, Greco A, Sartori L, D’Onofrio V, Antonini A, et al. EP111 / #336 THE IMPLEMENTATION OF INNOVATIVE ALGORITHM AND AI PROCESS FOR SEMI-AUTOMATED LOCALIZATION OF BRAIN BASAL NUCLEI IN FUNCTIONAL NEUROSURGERY: EPOSTER VIEWING: AS15 - DIGITALISATION AND AI IN NEUROMODULATION. Neuromodulation Technol Neural Interface [Internet]. 2023;26(8, Supplement):S67. Available from: https://www.sciencedirect.com/science/article/pii/S109471592300870X
56. Shaikh ST, Dwarakanath TA, Moiyadi A V. Evolution of Robotics in Neurosurgery. Asian J Neurosurg. 2024;
57. Villanueva-Meyer JE, Bakas S, Tiwari P, Lupo JM, Calabrese E, Davatzikos C, et al. Artificial Intelligence for Response Assessment in Neuro Oncology (AI-RANO), part 1: review of current advancements. Lancet Oncol [Internet]. 2024;25(11):e581–8. Available from: https://www.sciencedirect.com/science/article/pii/S1470204524003164
58. Zhan Y, Hao Y, Wang X, Guo D. Advances of artificial intelligence in clinical application and scientific research of neuro-oncology: Current knowledge and future perspectives. Crit Rev Oncol Hematol. 2025;104682