Whole Tumor Radiomics Analysis for Risk Factors Associated With Rapid Growth of Vestibular Schwannoma in Contrast-Enhanced T1-Weighted Images

Emerging strategies for the prediction of behaviour, growth, and treatment response in vestibular schwannoma

Vestibular schwannoma (VS) can present several management challenges for the clinician. Their unpredictable potential for growth creates uncertainty regarding when active treatment should be initiated, and once growth is confirmed which treatment option should be adopted, notably surgery or radiotherapy, and in particular stereotactic radiosurgery (SRS). The obvious benefits of SRS would ideally come with the ability to reliably predict long-term radiosurgery response/failure. Differentiation from temporary post-treatment phenomena such as transient tumour expansion or reactive swelling remains an unmet need. More powerful again would be the pre-treatment identification of which tumours will respond to radiosurgery and which will not. Over the past decade, there has been emerging interest in the development of non-invasive biomarkers, including imaging, for predicting growth and treatment response in VS. Alongside clinical radiographic predictors for VS growth such as extracanalicular tumour location and growth in the first year, studies have shown potential promise for advanced MRI and blood-based biomarkers that capture pathophysiological mechanism behind VS growth. Emerging interest in radiomics-based analyses of routinely acquired MRI, and the use of physiological imaging techniques such as dynamic-contrast enhanced MRI for pre- and post-treatment evaluation of tumour microvasculature and microstructure holds promise for revolutionizing this area. This article explores the current state of identifying VS growth at initial presentation, predicting treatment response to SRS and detecting early treatment failure, and finally the potential for developing more personalized patient selection for drug therapies, including bevacizumab, as well as emerging novel therapeutics for these tumours.

Artificial Intelligence in Temporal Bone Imaging: A Systematic Review

OBJECTIVE

The human temporal bone comprises more than 30 identifiable anatomical components. With the demand for precise image interpretation in this complex region, the utilization of artificial intelligence (AI) applications is steadily increasing. This systematic review aims to highlight the current role of AI in temporal bone imaging.

DATA SOURCES

A Systematic Review of English Publications searching MEDLINE (PubMed), COCHRANE Library, and EMBASE.

REVIEW METHODS

The search algorithm employed consisted of key items such as 'artificial intelligence,' 'machine learning,' 'deep learning,' 'neural network,' 'temporal bone,' and 'vestibular schwannoma.' Additionally, manual retrieval was conducted to capture any studies potentially missed in our initial search. All abstracts and full texts were screened based on our inclusion and exclusion criteria.

RESULTS

A total of 72 studies were included. 95.8% were retrospective and 88.9% were based on internal databases. Approximately two-thirds involved an AI-to-human comparison. Computed tomography (CT) was the imaging modality in 54.2% of the studies, with vestibular schwannoma (VS) being the most frequent study item (37.5%). Fifty-eight out of 72 articles employed neural networks, with 72.2% using various types of convolutional neural network models. Quality assessment of the included publications yielded a mean score of 13.6 ± 2.5 on a 20-point scale based on the CONSORT-AI extension.

CONCLUSION

Current research data highlight AI's potential in enhancing diagnostic accuracy with faster results and decreased performance errors compared to those of clinicians, thus improving patient care. However, the shortcomings of the existing research, often marked by heterogeneity and variable quality, underscore the need for more standardized methodological approaches to ensure the consistency and reliability of future data.

LEVEL OF EVIDENCE

NA Laryngoscope, 135:973-981, 2025.

Untreated Vestibular Schwannoma: Analysis of the Determinants of Growth

The growth rate of sporadic VS varies considerably, posing challenges for consistent clinical management. This systematic review examines data on factors associated with VS growth, following a protocol registered in the PROSPERO database. The analysis reveals that key predictors of tumor growth include tumor location, initial size, and specific clinical symptoms such as hearing loss and imbalance. Additionally, several studies suggest that growth observed within the first year may serve as an indicator of subsequent progression, enabling the earlier identification of high-risk cases. Emerging factors such as the posture swing test and MRI signal intensity have also been identified as novel predictors that could further refine growth assessments. Our meta-analysis confirms that tumor location, initial size, cystic components, and vestibular symptoms are closely linked to the likelihood of VS growth. This review provides valuable guidance for clinicians in identifying patients who may require closer monitoring or early intervention. By integrating these predictive factors into clinical practice, this review supports more personalized treatment and contributes to the development of more accurate prognostic models for managing untreated sporadic VS.

Analysis of the association between vestibular schwannoma and hearing status using a newly developed radiomics technique

PURPOSE

Vestibular schwannoma is a benign tumor originating from Schwann cells surrounding the eighth cranial nerve and can cause hearing loss, tinnitus, balance problems, and facial nerve disorders. Because of the slow growth of the tumor, predicting the hearing function of patients with vestibular schwannoma's is important to obtain information that would be useful for deciding the treatment modality. This study aimed to analyze the association between magnetic resonance imaging features and hearing status using a new radiomics technique.

METHODS

We retrospectively analyzed 115 magnetic resonance images and hearing results from 73 patients with vestibular schwannoma. A total of 70 radiomics features from each tumor volume were calculated using T1-weighted magnetic resonance imaging. Radiomics features were classified as histogram-based, shape-based, texture-based, and filter-based. The least absolute shrinkage and selection operator method was used to select the radiomics features among the 70 features that best predicted the hearing test. To ensure the stability of the selected features, the least absolute shrinkage and selection operator method was repeated 10 times. Finally, features set five or more times were selected as radiomics signatures.

RESULTS

The radiomics signatures selected using the least absolute shrinkage and selection operator method were: minimum, variance, maximum 3D diameter, size zone variance, log skewness, skewness slope, and kurtosis slope. In random forest, the mean performance was 0.66 (0.63-0.77), and the most important feature was Log skewness.

CONCLUSIONS

Newly developed radiomics features are associated with hearing status in patients with vestibular schwannoma and could provide information when deciding the treatment modality.

Applications and Integration of Radiomics for Skull Base Oncology

Radiomics, a quantitative approach to extracting features from medical images, represents a new frontier in skull base oncology. Novel image analysis approaches have enabled us to capture patterns from images imperceptible by the human eye. This rich source of data can be combined with a range of clinical features, holding the potential to be a noninvasive source of biomarkers. Applications of radiomics in skull base pathologies have centered around three common tumor classes: meningioma, sellar/parasellar tumors, and vestibular schwannomas. Radiomic investigations can be categorized into five domains: tumor detection/segmentation, classification between tumor types, tumor grading, detection of tumor features, and prognostication. Various computational architectures have been employed across these domains, with deep-learning methods becoming more common versus machine learning. Across radiomic applications, contrast-enhanced T1-weighted MRI images remain the most utilized sequence for model development. Efforts to standardize and connect radiomic features to tumor biology have facilitated more clinically applicable radiomic models. Despite the advancement in model performance, several challenges continue to hinder translatability, including small sample sizes and model training on homogenous single institution data. To recognize the potential of radiomics for skull base oncology, prospective, multi-institutional collaboration will be the cornerstone for a validated radiomic technology.