Prediction of pseudoprogression and long-term outcome of vestibular schwannoma after Gamma Knife radiosurgery based on preradiosurgical MR radiomics

Federated Learning: A Cross-Institutional Feasibility Study of Deep Learning Based Intracranial Tumor Delineation Framework for Stereotactic Radiosurgery

BACKGROUND

Deep learning-based segmentation algorithms usually required large or multi-institute data sets to improve the performance and ability of generalization. However, protecting patient privacy is a key concern in the multi-institutional studies when conventional centralized learning (CL) is used.

PURPOSE

To explores the feasibility of a proposed lesion delineation for stereotactic radiosurgery (SRS) scheme for federated learning (FL), which can solve decentralization and privacy protection concerns.

STUDY TYPE

Retrospective.

SUBJECTS

506 and 118 vestibular schwannoma patients aged 15-88 and 22-85 from two institutes, respectively; 1069 and 256 meningioma patients aged 12-91 and 23-85, respectively; 574 and 705 brain metastasis patients aged 26-92 and 28-89, respectively.

FIELD STRENGTH/SEQUENCE

1.5T, spin-echo, and gradient-echo [Correction added after first online publication on 21 August 2023. Field Strength has been changed to "1.5T" from "5T" in this sentence.].

ASSESSMENT

The proposed lesion delineation method was integrated into an FL framework, and CL models were established as the baseline. The effect of image standardization strategies was also explored. The dice coefficient was used to evaluate the segmentation between the predicted delineation and the ground truth, which was manual delineated by neurosurgeons and a neuroradiologist.

STATISTICAL TESTS

The paired t-test was applied to compare the mean for the evaluated dice scores (p < 0.05).

RESULTS

FL performed the comparable mean dice coefficient to CL for the testing set of Taipei Veterans General Hospital regardless of standardization and parameter; for the Taichung Veterans General Hospital data, CL significantly (p < 0.05) outperformed FL while using bi-parameter, but comparable results while using single-parameter. For the non-SRS data, FL achieved the comparable applicability to CL with mean dice 0.78 versus 0.78 (without standardization), and outperformed to the baseline models of two institutes.

DATA CONCLUSION

The proposed lesion delineation successfully implemented into an FL framework. The FL models were applicable on SRS data of each participating institute, and the FL exhibited comparable mean dice coefficient to CL on non-SRS dataset. Standardization strategies would be recommended when FL is used.

LEVEL OF EVIDENCE

4 TECHNICAL EFFICACY: Stage 1.

Stereotactic radiosurgery for Koos grade IV vestibular schwannoma: a systematic review and meta-analysis

BACKGROUND

Stereotactic radiosurgery (SRS) is a well-established treatment option for Koos stage I-III vestibular schwannomas (VS), often used as the first line of treatment or after subtotal resection. However, the optimal treatment for Koos-IV VS remains unclear. Therefore, our study aimed to evaluate the effectiveness of SRS as a primary treatment for large VS classified as Koos-IV.

METHODS

A systematic search was performed on December 28th, 2022, based on PubMed, Web of Science, and Scopus according to the PRISMA statement. The review was updated on September 7th, 2023. The risk of bias was assessed using the NIH Quality Assessment Tool. The R software (ver. 4.3.2) was used for all quantitative analyses and preparation of the forest plots. Publication bias and sensitivity analysis were performed to evaluate the reliability of the obtained results.

RESULTS

Among 2941 screened records, ten studies (1398 patients) have been included in quantitative synthesis. The overall tumor control rate was 90.7% (95%CI 86.3-94.4). Kaplan-Meier estimates of tumor control at 2, 6, and 10 years were 96.0% (95% CI 92.9-97.6%), 88.8% (95% CI 86.9-89.8%), and 84.5% (95% CI, 81.2-85.8%), respectively. The overall hearing preservation rate was 56.5% (95%CI 37-75.1). Kaplan-Meier estimates of hearing preservation rate at 2, 6, and 10 years were 77.1% (95% CI 67.9-82.5%), 53.5% (95% CI 44.2-58.5%), and 38.1% (95% CI 23.4-40.7%), respectively. The overall facial nerve preservation rate was 100% (95%CI 99.9-100.0). The overall trigeminal neuropathy rate reached 5.7% (95%CI 2.9-9.2). The overall rate of new-onset hydrocephalus was 5.6% (95%CI 3-9). The overall rates of worsening or new-onset tinnitus and vertigo were 6.8% (95%CI 4.2-10.0) and 9.1% (95%CI 2.1-19.6) respectively. No publication bias was detected according to the used methods.

CONCLUSIONS

Our systematic review and meta-analysis demonstrated a high overall tumor control rate, excellent facial nerve preservation, and low incidence of new-onset or worsened tinnitus and vertigo. However, several drawbacks associated with SRS should be noted, such as the presence of post-SRS hydrocephalus risk, mediocre long-term hearing preservation, and the lack of immediate tumor decompression. Nevertheless, the use of SRS may be beneficial in appropriately selected cases of Koos-IV VS. Moreover, further prospective studies directly comparing SRS with surgery are necessary to determine the optimal treatment for large VS and verify our results on a higher level of evidence. Registration and protocol: CRD42023389856.

Magnetic resonance radiomics-derived sphericity correlates with seizure in brain arteriovenous malformations

OBJECTIVES

Angioarchitectural analysis of brain arteriovenous malformations (BAVMs) is qualitative and subject to interpretation. This study quantified the morphology of and signal changes in the nidal and perinidal areas by using MR radiomics and compared the performance of MR radiomics and angioarchitectural analysis in detecting epileptic BAVMs.

MATERIALS AND METHODS

From 2010 to 2020, a total of 111 patients with supratentorial BAVMs were retrospectively included and grouped in accordance with the initial presentation of seizure. Patients' angiograms and MR imaging results were analyzed to determine the corresponding angioarchitecture. The BAVM nidus was contoured on time-of-flight MR angiography images. The perinidal brain parenchyma was contoured on T2-weighted images, followed by radiomic analysis. Logistic regression analysis was performed to determine the independent risk factors for seizure. ROC curve analysis, decision curve analysis (DCA), and calibration curve were performed to compare the performance of angioarchitecture-based and radiomics-based models in diagnosing epileptic BAVMs.

RESULTS

In multivariate analyses, low sphericity (OR: 2012.07, p = .04) and angiogenesis (OR: 5.30, p = .01) were independently associated with a high risk of seizure after adjustment for age, sex, temporal location, and nidal volume. The AUC for the angioarchitecture-based, MR radiomics-based, and combined models was 0.672, 0.817, and 0.794, respectively. DCA confirmed the clinical utility of the MR radiomics-based and combined models.

CONCLUSIONS

Low nidal sphericity and angiogenesis were associated with high seizure risk in patients with BAVMs. MR radiomics-derived tools may be used for noninvasive and objective measurement for evaluating the risk of seizure due to BAVM.

CLINICAL RELEVANCE STATEMENT

Low nidal sphericity was associated with high seizure risk in patients with brain arteriovenous malformation and MR radiomics may be used as a noninvasive and objective measurement method for evaluating seizure risk in patients with brain arteriovenous malformation.

KEY POINTS

• Low nidal sphericity was associated with high seizure risk in patients with brain arteriovenous malformation. • The performance of MR radiomics in detecting epileptic brain arteriovenous malformations was more satisfactory than that of angioarchitectural analysis. • MR radiomics may be used as a noninvasive and objective measurement method for evaluating seizure risk in patients with brain arteriovenous malformation.

Association Between Pseudoprogression of Vestibular Schwannoma After Radiosurgery and Radiological Features of Solid and Cystic Components

BACKGROUND AND OBJECTIVES

The pathophysiology of vestibular schwannoma (VS) pseudoprogression after Gamma Knife radiosurgery (GKRS) remains unclear. Radiological features in pretreatment magnetic resonance images may help predict VS pseudoprogression. This study used VS radiological features quantified using an automated segmentation algorithm to predict pseudoprogression after GKRS treatment.

METHODS

This is a retrospective study comprising 330 patients with VS who received GKRS. After image preprocessing and T2W/contrast-enhanced T1-weighted image (CET1W) image generation, with fuzzy C-means clustering, VSs were segmented into solid and cystic components and classified as solid and cystic. Relevant radiological features were then extracted. The response to GKRS was classified into "nonpseudoprogression" and "pseudoprogression/fluctuation". The Z test for two proportions was used to compare solid and cystic VS for the likelihood of pseudoprogression/fluctuation. Logistic regression was used to assess the correlation between clinical variables and radiological features and response to GKRS.

RESULTS

The likelihood of pseudoprogression/fluctuation after GKRS was significantly higher for solid VS compared with cystic VS (55% vs 31%, P < .001). For the entire VS cohort, multivariable logistic regression revealed that a lower mean tumor signal intensity (SI) in T2W/CET1W images was associated with pseudoprogression/fluctuation after GKRS ( P = .001). For the solid VS subgroup, a lower mean tumor SI in T2W/CET1W images ( P = .035) was associated with pseudoprogression/fluctuation after GKRS. For the cystic VS subgroup, a lower mean SI of the cystic component in T2W/CET1W images ( P = .040) was associated with pseudoprogression/fluctuation after GKRS.

CONCLUSION

Pseudoprogression is more likely to occur in solid VS compared with cystic VS. Quantitative radiological features in pretreatment magnetic resonance images were associated with pseudoprogression after GKRS. In T2W/CET1W images, solid VS with a lower mean tumor SI and cystic VS with a lower mean SI of cystic component were more likely to have pseudoprogression after GKRS. These radiological features can help predict the likelihood of pseudoprogression after GKRS.

Joint Vestibular Schwannoma Enlargement Prediction and Segmentation Using a Deep Multi-task Model

OBJECTIVE

To develop a deep-learning-based multi-task (DMT) model for joint tumor enlargement prediction (TEP) and automatic tumor segmentation (TS) for vestibular schwannoma (VS) patients using their initial diagnostic contrast-enhanced T1-weighted (ceT1) magnetic resonance images (MRIs).

METHODS

Initial ceT1 MRIs for VS patients meeting the inclusion/exclusion criteria of this study were retrospectively collected. VSs on the initial MRIs and their first follow-up scans were manually contoured. Tumor volume and enlargement ratio were measured based on expert contours. A DMT model was constructed for jointly TS and TEP. The manually segmented VS volume on the initial scan and the tumor enlargement label (≥20% volumetric growth) were used as the ground truth for training and evaluating the TS and TEP modules, respectively.

RESULTS

We performed 5-fold cross-validation with the eligible patients (n = 103). Median segmentation dice coefficient, prediction sensitivity, specificity, accuracy, and area under the receiver operating characteristic curve (AUC) were measured and achieved the following values: 84.20%, 0.68, 0.78, 0.72, and 0.77, respectively. The segmentation result is significantly better than the separate TS network (dice coefficient of 83.13%, p = 0.03) and marginally lower than the state-of-the-art segmentation model nnU-Net (dice coefficient of 86.45%, p = 0.16). The TEP performance is significantly better than the single-task prediction model (AUC = 0.60, p = 0.01) and marginally better than a radiomics-based prediction model (AUC = 0.70, p = 0.17).

CONCLUSION

The proposed DMT model is of higher learning efficiency and achieves promising performance on TEP and TS. The proposed technology has the potential to improve VS patient management.

LEVEL OF EVIDENCE

NA Laryngoscope, 133:2754-2760, 2023.

Applications of Artificial Intelligence in Temporal Bone Imaging: Advances and Future Challenges

The applications of artificial intelligence (AI) in temporal bone (TB) imaging have gained significant attention in recent years, revolutionizing the field of otolaryngology and radiology. Accurate interpretation of imaging features of TB conditions plays a crucial role in diagnosing and treating a range of ear-related pathologies, including middle and inner ear diseases, otosclerosis, and vestibular schwannomas. According to multiple clinical studies published in the literature, AI-powered algorithms have demonstrated exceptional proficiency in interpreting imaging findings, not only saving time for physicians but also enhancing diagnostic accuracy by reducing human error. Although several challenges remain in routinely relying on AI applications, the collaboration between AI and healthcare professionals holds the key to better patient outcomes and significantly improved patient care. This overview delivers a comprehensive update on the advances of AI in the field of TB imaging, summarizes recent evidence provided by clinical studies, and discusses future insights and challenges in the widespread integration of AI in clinical practice.

Correlation between initial tumor enlargement and magnetic resonance imaging characteristics following linear accelerator-based stereotactic radiosurgery for acoustic neuromas

BACKGROUND

Initial tumor enlargement (or pseudoprogression) instead of true tumor progression is a common phenomenon in patients with acoustic neuromas who are treated with stereotactic radiosurgery (SRS). This phenomenon can affect clinical decision-making and patient management. This study assessed the correlation between initial tumor enlargement and magnetic resonance imaging characteristics in patients with acoustic neuromas who were treated with linear accelerator (LINAC)-based SRS. The long-term tumor control outcomes were also analyzed.

MATERIALS AND METHODS

In total, 330 patients with sporadic acoustic neuromas who were treated with LINAC SRS between March 2006 and March 2020 were retrospectively evaluated to assess their initial tumor enlargement. The tumors were divided into homogeneously enhanced, heterogeneously enhanced, and cystic types based on the morphological characteristics noted on magnetic resonance images. Tumor control was assessed in 275 patients with a follow-up duration of more than 2 years.

RESULTS

Initial enlargement was observed in 137 of 330 (41.5%) tumors as early as 3 months after LINAC SRS. Data analysis revealed that postoperative tumors with a residual volume lower than 2.5 cm3 had a lower incidence of initial enlargement (p = 0.039). No correlation was noted between the initial enlargement and morphological characteristics of tumors. In patients with a mean follow-up duration of 82.8 ± 37.2 months, heterogeneously enhanced tumors exhibited a lower control rate than homogeneously enhanced and cystic tumors (p = 0.045). No correlation was noted between initial enlargement and tumor control.

CONCLUSION

Initial enlargement can occur as early as 3 months after SRS. Postoperative residual tumors with a volume lower than 2.5 cm3 exhibit a lower incidence of initial enlargement. Heterogeneously enhanced tumors have a lower local control rate.

Development of Predictive Models for the Response of Vestibular Schwannoma Treated with Cyberknife®: A Feasibility Study Based on Radiomics and Machine Learning

PURPOSE

to predict vestibular schwannoma (VS) response to radiosurgery by applying machine learning (ML) algorithms on radiomic features extracted from pre-treatment magnetic resonance (MR) images.

METHODS

patients with VS treated with radiosurgery in two Centers from 2004 to 2016 were retrospectively evaluated. Brain T1-weighted contrast-enhanced MR images were acquired before and at 24 and 36 months after treatment. Clinical and treatment data were collected contextually. Treatment responses were assessed considering the VS volume variation based on pre- and post-radiosurgery MR images at both time points. Tumors were semi-automatically segmented and radiomic features were extracted. Four ML algorithms (Random Forest, Support Vector Machine, Neural Network, and extreme Gradient Boosting) were trained and tested for treatment response (i.e., increased or non-increased tumor volume) using nested cross-validation. For training, feature selection was performed using the Least Absolute Shrinkage and Selection Operator, and the selected features were used as input to separately build the four ML classification algorithms. To overcome class imbalance during training, Synthetic Minority Oversampling Technique was used. Finally, trained models were tested on the corresponding held out set of patients to evaluate balanced accuracy, sensitivity, and specificity.

RESULTS

108 patients treated with Cyberknife^® were retrieved; an increased tumor volume was observed at 24 months in 12 patients, and at 36 months in another group of 12 patients. The Neural Network was the best predictive algorithm for response at 24 (balanced accuracy 73% ± 18%, specificity 85% ± 12%, sensitivity 60% ± 42%) and 36 months (balanced accuracy 65% ± 12%, specificity 83% ± 9%, sensitivity 47% ± 27%).

CONCLUSIONS

radiomics may predict VS response to radiosurgery avoiding long-term follow-up as well as unnecessary treatment.

Radiomics and machine learning for predicting the consistency of benign tumors of the central nervous system: A systematic review

PURPOSE

Predicting the consistency of benign central nervous system (CNS) tumors prior to surgery helps to improve surgical outcomes. This review summarizes and analyzes the literature on using radiomics and/or machine learning (ML) for consistency prediction.

METHOD

The Medical Literature Analysis and Retrieval System Online (MEDLINE) database was screened for studies published in English from January 1st 2000. Data was extracted according to the PRISMA guidelines and quality of the studies was assessed in compliance with the Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2).

RESULTS

Eight publications were included focusing on pituitary macroadenomas (n = 5), pituitary adenomas (n = 1), and meningiomas (n = 2) using a retrospective (n = 6), prospective (n = 1), and unknown (n = 1) study design with a total of 763 patients for the consistency prediction. The studies reported an area under the curve (AUC) of 0.71-0.99 for their respective best performing model regarding the consistency prediction. Of all studies, four articles validated their models internally whereas none validated their models externally. Two articles stated making data available on request with the remaining publications lacking information with regard to data availability.

CONCLUSIONS

The research on consistency prediction of CNS tumors is still at an early stage regarding the use of radiomics and different ML techniques. Best-practice procedures regarding radiomics and ML need to be followed more rigorously to facilitate the comparison between publications and, accordingly, the possible implementation into clinical practice in the future.

Development and Validation of a Deep Learning Predictive Model Combining Clinical and Radiomic Features for Short-Term Postoperative Facial Nerve Function in Acoustic Neuroma Patients

OBJECTIVE

This study aims to construct and validate a predictable deep learning model associated with clinical data and multi-sequence magnetic resonance imaging (MRI) for short-term postoperative facial nerve function in patients with acoustic neuroma.

METHODS

A total of 110 patients with acoustic neuroma who underwent surgery through the retrosigmoid sinus approach were included. Clinical data and raw features from four MRI sequences (T1-weighted, T2-weighted, T1-weighted contrast enhancement, and T2-weighted-Flair images) were analyzed. Spearman correlation analysis along with least absolute shrinkage and selection operator regression were used to screen combined clinical and radiomic features. Nomogram, machine learning, and convolutional neural network (CNN) models were constructed to predict the prognosis of facial nerve function on the seventh day after surgery. Receiver operating characteristic (ROC) curve and decision curve analysis (DCA) were used to evaluate model performance. A total of 1050 radiomic parameters were extracted, from which 13 radiomic and 3 clinical features were selected.

RESULTS

The CNN model performed best among all prediction models in the test set with an area under the curve (AUC) of 0.89 (95% CI, 0.84-0.91).

CONCLUSION

CNN modeling that combines clinical and multi-sequence MRI radiomic features provides excellent performance for predicting short-term facial nerve function after surgery in patients with acoustic neuroma. As such, CNN modeling may serve as a potential decision-making tool for neurosurgery.

The study of predictive factors for the evolution of vestibular schwannomas

PURPOSE

The primary objective was to determine whether the analysis of textural heterogeneity of vestibular schwannomas on MRI at diagnosis was predictive of their radiological evolutivity. The secondary objective was to determine whether some clinical or radiological factors could also be predictive of growth.

METHODS

We conducted a pilot, observational and retrospective study of patients with a vestibular schwannoma, initially monitored, between April 2001 and November 2019 within the Oto-Neurosurgical Institute of Champagne Ardenne, Texture analysis was performed on gadolinium injected T1 and CISS T2 MRI sequences and six parameters were extracted: mean greyscale intensity, standard deviation of the greyscale histogram distribution, entropy, mean positive pixels, skewness and kurtosis, which were analysed by the Lasso method, using statistically penalised Cox models. Extrameatal location, tumour necrosis, perceived hearing loss < 2 years with objectified tone audiometry asymmetry, tinnitus at diagnosis, were investigated by the Log-Rank test to obtain univariate survival analyses.

RESULTS

78 patients were included and divided into 2 groups: group A comprising 39 "stable patients", and B comprising the remaining 39 "progressive patients". Independent analysis of the texture factors did not predict the growth potential of vestibular schwannomas. Among the clinical or radiological signs of interest, hearing loss < 2 years was identified as a prognostic factor for tumour progression with a significant trend (p = 0.05).

CONCLUSIONS

This study did not identify an association between texture analysis and vestibular schwannomas growth. Decreased hearing in the 2 years prior to diagnosis appears to predict potential radiological progression.

Using the deformity index of vital structures to predict outcome of patients with large vestibular schwannomas after Gamma Knife radiosurgery

PURPOSE

Microsurgery is the mainstay of treatment for large vestibular schwannomas (VS), but the benefits of radiosurgery remain incompletely defined. Here, we aim to use automated volumetric analysis software to quantify the degree of brain stem deformity to predict long-term outcomes of patients with large VS following GKRS.

METHODS

Between 2003 and 2020, 39 patients with large VS (volume > 8 cc) undergoing GKRS with a margin dose of 10-12 Gy were analyzed. The reconstruction 3D MRI was used to evaluate the extent of deformity for predicting the long-term outcome of patients.

RESULTS

Their mean tumor volume was 13.7 ± 6.3 cc, and their mean follow-up after GKRS was 86.7 ± 65.3 months. Favorable clinical outcome was observed in 26 (66.7%) patients, while 13 (33.3%) patients had treatment failure. Patients with small tumor volumes, low vital structure deformity indice [(TV/(BSV + CerV) and (TV + EV)/(BSV + CerV)], and long distance of tumor to the central line were more likely to have favorable clinical outcome after GKRS. Significant prognostic value was with tumor shrinkage ratio (< 50%) were CV, CV/TV, TV/CerV, (TV + EV)/(BSV + CerV), and the distance of tumor to the central line. In cox regression, favorable clinical outcome was correlated with the Charlson comorbidity index and cochlear dosage (both p < 0.05). In multivariant analysis, tumor regression was highly correlated with the CV/TV ratio (p < 0.001).

CONCLUSIONS

The brainstem deformity ratio is likely a useful index to assess the clinical and tumor regression outcomes. Clinical outcomes are multifactorial and the tumor regression was highly correlated with the ratio of cystic components.

Subclassification of the Koos grade 2 vestibular schwannoma into 2a and 2b for individualized patient care: a validity and reliability study

OBJECTIVE

Vestibular schwannoma (VS) growth of ≥2 mm during serial MRI observation, irrespective of size, is the benchmark for treatment initiation in almost all centers. Although the probability of less optimal outcomes significantly increases in VS closer to the brainstem, early intervention does not improve long-term quality of life. Moving beyond the recommendation of definitive treatment for all VS after detected growth, we subclassified Koos 2 tumors based on extrameatal extension and relation to the brainstem. The aim of the current study was to evaluate the Koos 2 subclassification's validity and the inter-and intra-rater reliability of the entire Koos classification.

METHODS

Six experts, including neurosurgeons, otorhinolaryngologists and radiologists from two tertiary referral centers, classified 43 VS MRI scans. Validity of the Koos 2 subclassification was evaluated by the percentage agreement against the multidisciplinary skull base tumor board management advice. Inter- and intra-rater reliability were calculated using the intraclass correlation coefficient (ICC).

RESULTS

Validity was almost perfect in Koos 2a VSs with a 100% agreement and 87.5% agreement for Koos 2b. Inter-rater reliability for all Koos grades was significantly excellent (ICC 0.91; 95%CI 0.866 to 0.944, p= <0.001). Five raters had an excellent intra-rater reliability (ICC > 0.90; p= <0.01) and one rater had a good intra-rater reliability (ICC 0.88; 95% CI 0.742 to 0.949).

CONCLUSIONS

Although multiple factors influence decision-making, the classification of Koos 2a and 2b with excellent inter- and intra-rater reliability, can aid in recommending treatment initiation, moving beyond detected tumor growth, aiming to optimize patient centered care.

Application of artificial intelligence to stereotactic radiosurgery for intracranial lesions: detection, segmentation, and outcome prediction

BACKGROUND

Rapid evolution of artificial intelligence (AI) prompted its wide application in healthcare systems. Stereotactic radiosurgery served as a good candidate for AI model development and achieved encouraging result in recent years. This article aimed at demonstrating current AI application in radiosurgery.

METHODS

Literatures published in PubMed during 2010-2022, discussing AI application in stereotactic radiosurgery were reviewed.

RESULTS

AI algorithms, especially machine learning/deep learning models, have been administered to different aspect of stereotactic radiosurgery. Spontaneous tumor detection and automated lesion delineation or segmentation were two of the promising application, which could be further extended to longitudinal treatment follow-up. Outcome prediction utilized machine learning algorithms with radiomic-based analysis was another well-established application.

CONCLUSIONS

Stereotactic radiosurgery has taken a lead role in AI development. Current achievement, limitation, and further investigation was summarized in this article.

Lesion delineation framework for vestibular schwannoma, meningioma and brain metastasis for gamma knife radiosurgery using stereotactic magnetic resonance images

BACKGROUND AND OBJECTIVE

GKRS is an effective treatment for smaller intracranial tumors with a high control rate and low risk of complications. Target delineation in medical MR images is essential in the planning of GKRS and follow-up. A deep learning-based algorithm can effectively segment the targets from medical images and has been widely explored. However, state-of-the-art deep learning-based target delineation uses fixed sizes, and the isotropic voxel size may not be suitable for stereotactic MR images which use different anisotropic voxel sizes and numbers of slices according to the lesion size and location for clinical GKRS planning. This study developed an automatic deep learning-based segmentation scheme for stereotactic MR images.

METHODS

We retrospectively collected stereotactic MR images from 506 patients with VS, 1,069 patients with meningioma and 574 patients with BM who had been treated using GKRS; the lesion contours and individual T1W+C and T2W MR images were extracted from the GammaPlan system. The three-dimensional patching-based training strategy and dual-pathway architecture were used to manage inconsistent FOVs and anisotropic voxel size. Furthermore, we used two-parametric MR image as training input to segment the regions with different image characteristics (e.g., cystic lesions) effectively.

RESULTS

Our results for VS and BM demonstrated that the model trained using two-parametric MR images significantly outperformed the model trained using single-parametric images with median Dice coefficients (0.91, 0.05 versus 0.90, 0.06, and 0.82, 0.23 versus 0.78, 0.34, respectively), whereas predicted delineations in meningiomas using the dual-pathway model were dominated by single-parametric images (median Dice coefficients 0.83, 0.17 versus 0.84, 0.22). Finally, we combined three data sets to train the models, achieving the comparable or even higher testing median Dice (VS: 0.91, 0.07; meningioma: 0.83, 0.22; BM: 0.84, 0.23) in three diseases while using two-parametric as input.

CONCLUSIONS

Our proposed deep learning-based tumor segmentation scheme was successfully applied to multiple types of intracranial tumor (VS, meningioma and BM) undergoing GKRS and for segmenting the tumor effectively from stereotactic MR image volumes for use in GKRS planning.

Treatment of Intracranial Tumors With Stereotactic Radiosurgery: Short-Term Results From Cuba

Background Although international publications on radiosurgery have increased exponentially, reports of heterogeneous series treated with linear accelerator (LINAC) are scarce. Since most intracranial tumors are irregular in size and not spherical, LINACs (Elekta Precise®, Elekta AB, Sweden), fitted with a multi-leaf collimator, allow for precise stereotactic radiosurgery for the entire tumor. Aim To evaluate the effects of LINAC on an outpatient basis with patients diagnosed with various intracranial malignancies. Methodology A retrospective observational study of a series of cases of patients with intracranial lesions treated at the Institute of Oncology and Radiobiology using LINAC was carried out from October 2019 to May 2021 to evaluate the therapeutic results of radiosurgery in patients with intracranial tumors. Results A total of 22 lesions in 20 patients were treated with LINAC. The average age of the patients was 49.7, and the male-female ratio was 1:2. The cases consisted were mostly vestibular schwannoma (7 lesions), metastases from breast cancer (3 lesions), and tuberculum sellae meningioma (2 lesions). The prescription dose covered 99% of the planning target volume in 16 lesions (72.7%) and 100% in six lesions (27.3%) (prescription volume). In meningiomas and schwannomas, doses between 12 and 14 Gy were used, in plasmacytoma 13 Gy, in pilocytic astrocytoma 14 Gy, in cavernoma 15 Gy, in breast cancer metastasis between 18 and 20 Gy, and in lung cancer metastasis 22 Gy. When evaluating local control, 11 patients exhibited stable findings at the six-month control while 10 had partial regression, and a single patient had total regression. Minor complications such as perilesional edema, facial paresthesia, facial paralysis, and transient alopecia were observed in eight of the patients. Conclusions Patients with extra-axial, low-grade malignancy, and posterior fossa lesions were predominant in the studied population. Radiosurgery treatment is associated with good local control of the treated lesions. Complications are infrequent, mild, and predominated by perilesional edema.

Machine Learning in the Management of Lateral Skull Base Tumors: A Systematic Review

The application of machine learning (ML) techniques to otolaryngology remains a topic of interest and prevalence in the literature, though no previous articles have summarized the current state of ML application to management and the diagnosis of lateral skull base (LSB) tumors. Subsequently, we present a systematic overview of previous applications of ML techniques to the management of LSB tumors. Independent searches were conducted on PubMed and Web of Science between August 2020 and February 2021 to identify the literature pertaining to the use of ML techniques in LSB tumor surgery written in the English language. All articles were assessed in regard to their application task, ML methodology, and their outcomes. A total of 32 articles were examined. The number of articles involving applications of ML techniques to LSB tumor surgeries has significantly increased since the first article relevant to this field was published in 1994. The most commonly employed ML category was tree-based algorithms. Most articles were included in the category of surgical management (13; 40.6%), followed by those in disease classification (8; 25%). Overall, the application of ML techniques to the management of LSB tumor has evolved rapidly over the past two decades, and the anticipated growth in the future could significantly augment the surgical outcomes and management of LSB tumors.

MRI radiomics in the prediction of the volumetric response in meningiomas after gamma knife radiosurgery

PURPOSE

This report presents the first investigation of the radiomics value in predicting the meningioma volumetric response to gamma knife radiosurgery (GKRS).

METHODS

The retrospective study included 93 meningioma patients imaged by three Tesla MRI. Tumor morphology was quantified by calculating 337 shape, first- and second-order radiomic features from MRI obtained before GKRS. Analysis was performed on original 3D MR images and after their laplacian of gaussian (LoG), logarithm and exponential filtering. The prediction performance was evaluated by Pearson correlation, linear regression and ROC analysis, with meningioma volume change per month as the outcome.

RESULTS

Sixty calculated features significantly correlated with the outcome. The feature selection based on LASSO and multivariate regression started from all available 337 radiomic and 12 non-radiomic features. It selected LoG-sigma-1-0-mm-3D_firstorder_InterquartileRange and logarithm_ngtdm_Busyness as the predictively most robust and non-redundant features. The radiomic score based on these two features produced an AUC = 0.81. Adding the non-radiomic karnofsky performance status (KPS) to the score has increased the AUC to 0.88. Low values of the radiomic score defined a homogeneous subgroup of 50 patients with consistent absence (0%) of tumor progression.

CONCLUSION

This is the first report of a strong association between MRI radiomic features and volumetric meningioma response to radiosurgery. The clinical importance of the early and reliable prediction of meningioma responsiveness to radiosurgery is based on its potential to aid individualized therapy decision making.

Machine Learning for the Detection and Segmentation of Benign Tumors of the Central Nervous System: A Systematic Review

Simple Summary

Machine learning in radiology of the central nervous system has seen many interesting publications in the past few years. Since the focus has largely been on malignant tumors such as brain metastases and high-grade gliomas, we conducted a systematic review on benign tumors to summarize what has been published and where there might be gaps in the research. We found several studies that report good results, but the descriptions of methodologies could be improved to enable better comparisons and assessment of biases.

Abstract

Objectives: To summarize the available literature on using machine learning (ML) for the detection and segmentation of benign tumors of the central nervous system (CNS) and to assess the adherence of published ML/diagnostic accuracy studies to best practice. Methods: The MEDLINE database was searched for the use of ML in patients with any benign tumor of the CNS, and the records were screened according to PRISMA guidelines. Results: Eleven retrospective studies focusing on meningioma (n = 4), vestibular schwannoma (n = 4), pituitary adenoma (n = 2) and spinal schwannoma (n = 1) were included. The majority of studies attempted segmentation. Links to repositories containing code were provided in two manuscripts, and no manuscripts shared imaging data. Only one study used an external test set, which raises the question as to whether some of the good performances that have been reported were caused by overfitting and may not generalize to data from other institutions. Conclusions: Using ML for detecting and segmenting benign brain tumors is still in its infancy. Stronger adherence to ML best practices could facilitate easier comparisons between studies and contribute to the development of models that are more likely to one day be used in clinical practice.

Beyond Glioma: The Utility of Radiomic Analysis for Non-Glial Intracranial Tumors

Simple Summary

Tumor qualities, such as growth rate, firmness, and intrusion into healthy tissue, can be very important for operation planning and further treatment. Radiomics is a promising new method that allows the determination of some of these qualities on images performed before surgery. In this article, we provide a review of the use of radiomics in various tumors of the central nervous system, such as metastases, lymphoma, meningioma, medulloblastoma, and pituitary tumors.

Abstract

The field of radiomics is rapidly expanding and gaining a valuable role in neuro-oncology. The possibilities related to the use of radiomic analysis, such as distinguishing types of malignancies, predicting tumor grade, determining the presence of particular molecular markers, consistency, therapy response, and prognosis, can considerably influence decision-making in medicine in the near future. Even though the main focus of radiomic analyses has been on glial CNS tumors, studies on other intracranial tumors have shown encouraging results. Therefore, as the main focus of this review, we performed an analysis of publications on PubMed and Web of Science databases, focusing on radiomics in CNS metastases, lymphoma, meningioma, medulloblastoma, and pituitary tumors.

Response prediction of vestibular schwannoma after gamma-knife radiosurgery using pretreatment dynamic contrast-enhanced MRI: a prospective study

OBJECTIVES

There are few known predictive factors for response to gamma-knife radiosurgery (GKRS) in vestibular schwannoma (VS). We investigated the predictive role of pretreatment dynamic contrast-enhanced (DCE)-MRI parameters regarding the tumor response after GKRS in sporadic VS.

METHODS

This single-center prospective study enrolled participants between April 2017 and February 2019. We performed a volumetric measurement of DCE-MRI-derived parameters before GKRS. The tumor volume was measured in a follow-up MRI. The pharmacokinetic parameters were compared between responders and nonresponders according to 20% or more tumor volume reduction. Stepwise multivariable logistic regression analyses were performed, and the diagnostic performance of DCE-MRI parameters for the prediction of tumor response was evaluated by receiver operating characteristic curve analysis.

RESULTS

Ultimately, 35 participants (21 women, 52 ± 12 years) were included. There were 22 (62.9%) responders with a mean follow-up interval of 30.2 ± 5.7 months. K^trans (0.036 min^-1 vs. 0.057 min^-1, p = .008) and initial area under the time-concentration curve within 90 s (IAUC90) (84.4 vs. 143.6, p = .003) showed significant differences between responders and nonresponders. K^trans (OR = 0.96, p = .021) and IAUC90 (OR = 0.97, p = .004) were significant differentiating variables in each multivariable model with clinical variables for tumor response prediction. K^trans showed a sensitivity of 81.8% and a specificity of 69.2%, and IAUC90 showed a sensitivity of 100% and a specificity of 53.8% for tumor response prediction.

CONCLUSION

DCE-MRI (particularly K^trans and IAUC90) has the potential to be a predictive factor for tumor response in VS after GKRS.

KEY POINTS

•Pretreatment prediction of gamma-knife radiosurgery response in vestibular schwannoma is still challenging. •Dynamic contrast-enhanced MRI could have predictive value for the response of vestibular schwannoma after gamma-knife radiosurgery.

Comparison of Conventional and Radiomic Features between 18F-FBPA PET/CT and PET/MR

Boron-10-containing positron emission tomography (PET) radio-tracer, ¹⁸F-FBPA, has been used to evaluate the feasibility and treatment outcomes of Boron neutron capture therapy (BNCT). The clinical use of PET/MR is increasing and reveals its benefit in certain applications. However, the PET/CT is still the most widely used modality for daily PET practice due to its high quantitative accuracy and relatively low cost. Considering the different attenuation correction maps between PET/CT and PET/MR, comparison of derived image features from these two modalities is critical to identify quantitative imaging biomarkers for diagnosis and prognosis. This study aimed to investigate the comparability of image features extracted from ¹⁸F-FBPA PET/CT and PET/MR. A total of 15 patients with malignant brain tumor who underwent ¹⁸F-FBPA examinations using both PET/CT and PET/MR on the same day were retrospectively analyzed. Overall, four conventional imaging characteristics and 449 radiomic features were calculated from PET/CT and PET/MR, respectively. A linear regression model and intraclass correlation coefficient (ICC) were estimated to evaluate the comparability of derived features between two modalities. Features were classified into strong, moderate, and weak comparability based on coefficient of determination (r²) and ICC. All of the conventional features, 81.2% of histogram, 37.5% of geometry, 51.5% of texture, and 25% of wavelet-based features, showed strong comparability between PET/CT and PET/MR. With regard to the wavelet filtering, radiomic features without filtering (61.2%) or with low-pass filtering (59.2%) along three axes produced strong comparability between the two modalities. However, only 8.2% of the features with high-pass filtering showed strong comparability. The linear regression models were provided for the features with strong and moderate consensus to interchange the quantitative features between the PET/CT and the PET/MR. All of the conventional and 71% of the radiomic (mostly histogram and texture) features were sufficiently stable and could be interchanged between ¹⁸F-FBPA PET with different hybrid modalities using the proposed equations. Our findings suggested that the image features high interchangeability may facilitate future studies in comparing PET/CT and PET/MR.

Clinical applications of artificial intelligence and radiomics in neuro-oncology imaging

This article is a comprehensive review of the basic background, technique, and clinical applications of artificial intelligence (AI) and radiomics in the field of neuro-oncology. A variety of AI and radiomics utilized conventional and advanced techniques to differentiate brain tumors from non-neoplastic lesions such as inflammatory and demyelinating brain lesions. It is used in the diagnosis of gliomas and discrimination of gliomas from lymphomas and metastasis. Also, semiautomated and automated tumor segmentation has been developed for radiotherapy planning and follow-up. It has a role in the grading, prediction of treatment response, and prognosis of gliomas. Radiogenomics allowed the connection of the imaging phenotype of the tumor to its molecular environment. In addition, AI is applied for the assessment of extra-axial brain tumors and pediatric tumors with high performance in tumor detection, classification, and stratification of patient's prognoses.

[Macroscopic and microscopic changes of the vestibulocochlear nerve after Gamma Knife treatment]

Wir berichten über einen Fall, bei dem makroskopische und mikroskopische Veränderungen des Verstibularnervs nach radiochirurgischer Behandlung eines intrameatalen Vestibularisschwannoms beobachtet wurden. Der Fallbericht zeigt das erste Mal ein morphologisches Korrelat der unerwünschten Effekte der Gamma-Knife-Therapie von Vestibularisschwannomen und unterstreicht, dass trotz eines deutlichen Abstands zum bestehenden Tumor degenerative Veränderungen der neuralen Strukturen erwartet werden können.

Enhancement of Radiosurgical Treatment Outcome Prediction Using MRI Radiomics in Patients with Non-Small Cell Lung Cancer Brain Metastases

Simple Summary

Non-small cell lung cancer (NSCLC) is the most common cause of brain metastasis (BM). Approximately 50% of patients with metastatic NSCLC harbor BMs. Within the past decade, Gamma Knife radiosurgery (GKRS) has become one of the first-line treatments for BMs. Ability to predict treatment response after GKRS can therefore guide treatment strategy. This study aimed to determine whether pre-radiosurgical neuroimaging radiomics can predict survival and local tumor control after GKRS. Based on the collected magnetic resonance images and clinical characteristics of the 237 NSCLC BM patients with BMs (for survival prediction) and 256 NSCLC patients with 976 BMs (for prediction of local tumor control), we concluded that the identified radiomic features could provide valuable additional information to enhance the prediction of BM responses after GKRS. The proposed approach provided physicians with an intuitive way to predict the patient outcome based on pre-radiosurgical magnetic resonance images.

Abstract

The diagnosis of brain metastasis (BM) is commonly observed in non-small cell lung cancer (NSCLC) with poor outcomes. Accordingly, developing an approach to early predict BM response to Gamma Knife radiosurgery (GKRS) may benefit the patient treatment and monitoring. A total of 237 NSCLC patients with BMs (for survival prediction) and 256 patients with 976 BMs (for prediction of local tumor control) treated with GKRS were retrospectively analyzed. All the survival data were recorded without censoring, and the status of local tumor control was determined by comparing the last MRI follow-up in patients’ lives with the pre-GKRS MRI. Overall 1763 radiomic features were extracted from pre-radiosurgical magnetic resonance images. Three prediction models were constructed, using (1) clinical data, (2) radiomic features, and (3) clinical and radiomic features. Support vector machines with a 30% hold-out validation approach were constructed. For treatment outcome predictions, the models derived from both the clinical and radiomics data achieved the best results. For local tumor control, the combined model achieved an area under the curve (AUC) of 0.95, an accuracy of 90%, a sensitivity of 91%, and a specificity of 89%. For patient survival, the combined model achieved an AUC of 0.81, an accuracy of 77%, a sensitivity of 78%, and a specificity of 80%. The pre-radiosurgical radiomics data enhanced the performance of local tumor control and survival prediction models in NSCLC patients with BMs treated with GRKS. An outcome prediction model based on radiomics combined with clinical features may guide therapy in these patients.

Repeat stereotactic radiosurgery for progressive vestibular schwannomas after previous radiosurgery: a systematic review and meta-analysis

Vestibular schwannomas (VS) are slow-growing intracranial extraaxial benign tumors, developing from the vestibular part of the eight cranial nerves. Stereotactic radiosurgery (SRS) has now a long-term scientific track record as first intention treatment for small- to medium-sized VS. Though its success rate is very high, SRS for VS might fail to control tumor growth in some cases. However, the literature on repeat SRS after previously failed SRS remains scarce and reported in a low number of series with a limited number of cases. Here, we aimed at performing a systematic review and meta-analysis of the literature on repeat SRS for VS. Using PRISMA guidelines, we reviewed manuscripts published between January 1990 and October 2020 and referenced in PubMed. Tumor control and cranial nerve outcomes were evaluated with separate meta-analyses. Eight studies comprising 194 patients were included. The overall rate of patients treated in repeat SRS series as per overall series with first SRS was 2.2% (range 1.2–3.2%, p < 0.001). The mean time between first and second SRS was 50.7 months (median 51, range 44–64). The median marginal dose prescribed at first SRS was 12 Gy (range 8–24) and at second SRS was 12 Gy (range 9.8–19). After repeat SRS, tumor stability was reported in 61/194 patients, i.e., a rate of 29.6% (range 20.2–39%, I² = 49.1%, p < 0.001). Tumor decrease was reported in 83/194 patients, i.e., a rate of 54.4% (range 33.7–75.1%, I² = 89.1%, p < 0.001). Tumor progression was reported in 50/188 patients, i.e., a rate of 16.1% (range 2.5–29.7%, I² = 87.1%, p = 0.02), rarely managed surgically. New trigeminal numbness was reported in 27/170 patients, i.e., a rate of 9.9% (range 1.4–18.3%, p < 0.02). New facial nerve palsy of worsened of previous was reported in 8/183 patients, i.e., a rate of 4.3% (range 1.4–7.2%, p = 0.004). Hearing loss was reported in 12/22 patients, i.e., a rate of 54.3% (range 24.8–83.8%, I² = 70.7%, p < 0.001). Repeat SRS after previously failed SRS for VS is associated with high tumor control rates. Cranial nerve outcomes remain favorable, particularly for facial nerve. The rate of hearing loss appears similar to the one related to first SRS.

Applying artificial intelligence to longitudinal imaging analysis of vestibular schwannoma following radiosurgery

Artificial intelligence (AI) has been applied with considerable success in the fields of radiology, pathology, and neurosurgery. It is expected that AI will soon be used to optimize strategies for the clinical management of patients based on intensive imaging follow-up. Our objective in this study was to establish an algorithm by which to automate the volumetric measurement of vestibular schwannoma (VS) using a series of parametric MR images following radiosurgery. Based on a sample of 861 consecutive patients who underwent Gamma Knife radiosurgery (GKRS) between 1993 and 2008, the proposed end-to-end deep-learning scheme with automated pre-processing pipeline was applied to a series of 1290 MR examinations (T1W+C, and T2W parametric MR images). All of which were performed under consistent imaging acquisition protocols. The relative volume difference (RVD) between AI-based volumetric measurements and clinical measurements performed by expert radiologists were + 1.74%, - 0.31%, - 0.44%, - 0.19%, - 0.01%, and + 0.26% at each follow-up time point, regardless of the state of the tumor (progressed, pseudo-progressed, or regressed). This study outlines an approach to the evaluation of treatment responses via novel volumetric measurement algorithm, and can be used longitudinally following GKRS for VS. The proposed deep learning AI scheme is applicable to longitudinal follow-up assessments following a variety of therapeutic interventions.