Neuroimaging assessment of facility-bound severely-affected MS reveals the critical role of cortical gray matter pathology: results from the CASA-MS case-controlled study.

BACKGROUND: A subgroup of people with multiple sclerosis (pwMS) will develop severe disability. The pathophysiology underlying severe MS is unknown. The comprehensive assessment of severely affected MS (CASA-MS) was a case-controlled study that compared severely disabled in skilled nursing (SD/SN) (EDSS ≥ 7.0) to less-disabled (EDSS 3.0-6.5) community dwelling (CD) progressive pwMS, matched on age-, sex- and disease-duration (DDM). OBJECTIVES: To identify neuroimaging and molecular biomarker characteristics that distinguish SD/SN from DDM-CD progressive pwMS. METHODS: This study was carried at SN facility and at a tertiary MS center. The study collected clinical, molecular (serum neurofilament light chain, sNfL and glial acidic fibrillary protein, sGFAP) and MRI quantitative lesion-, brain volume-, and tissue integrity-derived measures. Statistical analyses were controlled for multiple comparisons. RESULTS: 42 SD/SN and 42 DDM-CD were enrolled. SD/SN pwMS showed significantly lower cortical volume (CV) (p < 0.001, d = 1.375) and thalamic volume (p < 0.001, d = 0.972) compared to DDM-CD pwMS. In a logistic stepwise regression model, the SD/SN pwMS were best differentiated from the DDM-CD pwMS by lower CV (p < 0.001) as the only significant predictor, with the accuracy of 82.3%. No significant differences between the two groups were observed for medulla oblongata volume, a proxy for spinal cord atrophy and white matter lesion burden, while there was a statistical trend for numerically higher sGFAP in SD/SN pwMS. CONCLUSIONS: The CASA-MS study showed significantly more gray matter atrophy in severe compared to less-severe progressive MS.

Dynamic disability measures decrease the clinico-radiological gap in people with severely affected multiple sclerosis.

BACKGROUND: Expanded Disability Status Scale (EDSS) is limited when utilized in highly disabled people with multiple sclerosis (pwMS). OBJETIVE: To explore the relationship between disability measures and MRI outcomes in severely-affected pwMS. METHODS: PwMS recruited from The Boston Home (TBH), a specialized residential facility for severly-affected pwMS and University at Buffalo (UB) MS Center were assessed using EDSS, MS Severity Scale, age-related MSS, Scripps Neurological Rating Scale (SNRS) and Combinatorial Weight-Adjusted Disability Score (CombiWISE). In all scores except SNRS, higher score indicates greater disability. MRI measures of T1, T2-lesion volume (LV), whole brain, gray matter, medulla oblongata and thalamic volumes (WBV, GMV, MOV, TV) and thalamic dysconnectivity were obtained. RESULTS: Greatest disability differences between the TBH and UB pwMS were in SNRS (24.4 vs 71.9, p < 0.001, Cohen's d = 4.05) and CombiWISE (82.3 vs. 38.9, p < 0.001, Cohen's d = 4.02). In combined analysis of all pwMS, worse SNRS scores were correlated with worse MRI pathology in 8 out of 9 outcomes. EDSS only with 3 measures (GMV, MOV and TV). In severely-affected pwMS, SNRS was associated with T1-LV, T2-LV and WBV (not surviving false discovery rate (FDR) correction for multiple comparisons) whereas EDSS did not. CONCLUSION: Granular and dynamic disability measures may bridge the clinico-radiologcal gap present in severely affected pwMS.

Object-oriented hand dexterity and grasping abilities, from the animal quarters to the neurosurgical OR: a systematic review of the underlying neural correlates in non-human, human primate and recent findings in awake brain surgery.

Introduction: The sensorimotor integrations subserving object-oriented manipulative actions have been extensively investigated in non-human primates via direct approaches, as intracortical micro-stimulation (ICMS), cytoarchitectonic analysis and anatomical tracers. However, the understanding of the mechanisms underlying complex motor behaviors is yet to be fully integrated in brain mapping paradigms and the consistency of these findings with intraoperative data obtained during awake neurosurgical procedures for brain tumor removal is still largely unexplored. Accordingly, there is a paucity of systematic studies reviewing the cross-species analogies in neural activities during object-oriented hand motor tasks in primates and investigating the concordance with intraoperative findings during brain mapping. The current systematic review was designed to summarize the cortical and subcortical neural correlates of object-oriented fine hand actions, as revealed by fMRI and PET studies, in non-human and human primates and how those were translated into neurosurgical studies testing dexterous hand-movements during intraoperative brain mapping. Methods: A systematic literature review was conducted following the PRISMA guidelines. PubMed, EMBASE and Web of Science databases were searched. Original articles were included if they: (1) investigated cortical activation sites on fMRI and/or PET during grasping task; (2) included humans or non-human primates. A second query was designed on the databases above to collect studies reporting motor, hand manipulation and dexterity tasks for intraoperative brain mapping in patients undergoing awake brain surgery for any condition. Due to the heterogeneity in neurosurgical applications, a qualitative synthesis was deemed more appropriate. Results: We provided an updated overview of the current state of the art in translational neuroscience about the extended frontoparietal grasping-praxis network with a specific focus on the comparative functioning in non-human primates, healthy humans and how the latter knowledge has been implemented in the neurosurgical operating room during brain tumor resection. Discussion: The anatomical and functional correlates we reviewed confirmed the evolutionary continuum from monkeys to humans, allowing a cautious but practical adoption of such evidence in intraoperative brain mapping protocols. Integrating the previous results in the surgical practice helps preserve complex motor abilities, prevent long-term disability and poor quality of life and allow the maximal safe resection of intrinsic brain tumors.

Surgical management and outcome of newly diagnosed glioblastoma without contrast enhancement (low-grade appearance): a report of the RANO resect group.

BACKGROUND: Resection of the contrast-enhancing (CE) tumor represents the standard of care in newly diagnosed glioblastoma. However, some tumors ultimately diagnosed as glioblastoma lack contrast enhancement and have a 'low-grade appearance' on imaging (non-CE glioblastoma). We aimed to (a) volumetrically define the value of non-CE tumor resection in the absence of contrast enhancement, and to (b) delineate outcome differences between glioblastoma patients with and without contrast enhancement. METHODS: The RANO resect group retrospectively compiled a global, eight-center cohort of patients with newly diagnosed glioblastoma per WHO 2021 classification. The associations between postoperative tumor volumes and outcome were analyzed. Propensity score-matched analyses were constructed to compare glioblastomas with and without contrast enhancement. RESULTS: Among 1323 newly diagnosed IDH-wildtype glioblastomas, we identified 98 patients (7.4%) without contrast enhancement. In such patients, smaller postoperative tumor volumes were associated with more favorable outcome. There was an exponential increase in risk for death with larger residual non-CE tumor. Accordingly, extensive resection was associated with improved survival compared to lesion biopsy. These findings were retained on a multivariable analysis adjusting for demographic and clinical markers. Compared to CE glioblastoma, patients with non-CE glioblastoma had a more favorable clinical profile and superior outcome as confirmed in propensity score analyses by matching the patients with non-CE glioblastoma to patients with CE glioblastoma using a large set of clinical variables. CONCLUSIONS: The absence of contrast enhancement characterizes a less aggressive clinical phenotype of IDH-wildtype glioblastomas. Maximal resection of non-CE tumors has prognostic implications and translates into favorable outcome.

Segmentation of glioblastomas in early post-operative multi-modal MRI with deep neural networks.

Extent of resection after surgery is one of the main prognostic factors for patients diagnosed with glioblastoma. To achieve this, accurate segmentation and classification of residual tumor from post-operative MR images is essential. The current standard method for estimating it is subject to high inter- and intra-rater variability, and an automated method for segmentation of residual tumor in early post-operative MRI could lead to a more accurate estimation of extent of resection. In this study, two state-of-the-art neural network architectures for pre-operative segmentation were trained for the task. The models were extensively validated on a multicenter dataset with nearly 1000 patients, from 12 hospitals in Europe and the United States. The best performance achieved was a 61% Dice score, and the best classification performance was about 80% balanced accuracy, with a demonstrated ability to generalize across hospitals. In addition, the segmentation performance of the best models was on par with human expert raters. The predicted segmentations can be used to accurately classify the patients into those with residual tumor, and those with gross total resection.

Multi-class glioma segmentation on real-world data with missing MRI sequences: comparison of three deep learning algorithms.

This study tests the generalisability of three Brain Tumor Segmentation (BraTS) challenge models using a multi-center dataset of varying image quality and incomplete MRI datasets. In this retrospective study, DeepMedic, no-new-Unet (nn-Unet), and NVIDIA-net (nv-Net) were trained and tested using manual segmentations from preoperative MRI of glioblastoma (GBM) and low-grade gliomas (LGG) from the BraTS 2021 dataset (1251 in total), in addition to 275 GBM and 205 LGG acquired clinically across 12 hospitals worldwide. Data was split into 80% training, 5% validation, and 15% internal test data. An additional external test-set of 158 GBM and 69 LGG was used to assess generalisability to other hospitals' data. All models' median Dice similarity coefficient (DSC) for both test sets were within, or higher than, previously reported human inter-rater agreement (range of 0.74-0.85). For both test sets, nn-Unet achieved the highest DSC (internal = 0.86, external = 0.93) and the lowest Hausdorff distances (10.07, 13.87 mm, respectively) for all tumor classes (p < 0.001). By applying Sparsified training, missing MRI sequences did not statistically affect the performance. nn-Unet achieves accurate segmentations in clinical settings even in the presence of incomplete MRI datasets. This facilitates future clinical adoption of automated glioma segmentation, which could help inform treatment planning and glioma monitoring.

Prognostic evaluation of re-resection for recurrent glioblastoma using the novel RANO classification for extent of resection: A report of the RANO resect group.

BACKGROUND: The value of re-resection in recurrent glioblastoma remains controversial as a randomized trial that specifies intentional incomplete resection cannot be justified ethically. Here, we aimed to (1) explore the prognostic role of extent of re-resection using the previously proposed Response Assessment in Neuro-Oncology (RANO) classification (based upon residual contrast-enhancing (CE) and non-CE tumor), and to (2) define factors consolidating the surgical effects on outcome. METHODS: The RANO resect group retrospectively compiled an 8-center cohort of patients with first recurrence from previously resected glioblastomas. The associations of re-resection and other clinical factors with outcome were analyzed. Propensity score-matched analyses were constructed to minimize confounding effects when comparing the different RANO classes. RESULTS: We studied 681 patients with first recurrence of Isocitrate Dehydrogenase (IDH) wild-type glioblastomas, including 310 patients who underwent re-resection. Re-resection was associated with prolonged survival even when stratifying for molecular and clinical confounders on multivariate analysis; ≤1 cm3 residual CE tumor was associated with longer survival than non-surgical management. Accordingly, "maximal resection" (class 2) had superior survival compared to "submaximal resection" (class 3). Administration of (radio-)chemotherapy in the absence of postoperative deficits augmented the survival associations of smaller residual CE tumors. Conversely, "supramaximal resection" of non-CE tumor (class 1) was not associated with prolonged survival but was frequently accompanied by postoperative deficits. The prognostic role of residual CE tumor was confirmed in propensity score analyses. CONCLUSIONS: The RANO resect classification serves to stratify patients with re-resection of glioblastoma. Complete resection according to RANO resect classes 1 and 2 is prognostic.

Prognostic validation of a new classification system for extent of resection in glioblastoma: A report of the RANO resect group.

BACKGROUND: Terminology to describe extent of resection in glioblastoma is inconsistent across clinical trials. A surgical classification system was previously proposed based upon residual contrast-enhancing (CE) tumor. We aimed to (1) explore the prognostic utility of the classification system and (2) define how much removed non-CE tumor translates into a survival benefit. METHODS: The international RANO resect group retrospectively searched previously compiled databases from 7 neuro-oncological centers in the USA and Europe for patients with newly diagnosed glioblastoma per WHO 2021 classification. Clinical and volumetric information from pre- and postoperative MRI were collected. RESULTS: We collected 1,008 patients with newly diagnosed IDHwt glioblastoma. 744 IDHwt glioblastomas were treated with radiochemotherapy per EORTC-26981/22981 (TMZ/RT→TMZ) following surgery. Among these homogenously treated patients, lower absolute residual tumor volumes (in cm3) were favorably associated with outcome: patients with "maximal CE resection" (class 2) had superior outcome compared to patients with "submaximal CE resection" (class 3) or "biopsy" (class 4). Extensive resection of non-CE tumor (≤5 cm3 residual non-CE tumor) was associated with better survival among patients with complete CE resection, thus defining class 1 ("supramaximal CE resection"). The prognostic value of the resection classes was retained on multivariate analysis when adjusting for molecular and clinical markers. CONCLUSIONS: The proposed "RANO categories for extent of resection in glioblastoma" are highly prognostic and may serve for stratification within clinical trials. Removal of non-CE tumor beyond the CE tumor borders may translate into additional survival benefit, providing a rationale to explicitly denominate such "supramaximal CE resection."

Transcranial versus direct electrical stimulation for intraoperative motor-evoked potential monitoring: Prognostic value comparison in asleep brain tumor surgery.

Objective: Safe resection of gliomas involving motor pathways in asleep-anesthesia requires the combination of brain mapping, to identify and spare essential motor sites, and continuous monitoring of motor-evoked potentials (MEPs), to detect possible vascular damage to the corticospinal tract (CST). MEP monitoring, according to intraoperative neurophysiology societies, is generally recommended by transcranial electrodes (TES), and no clear indications of direct cortical stimulation (DCS) or the preferential use of one of the two techniques based on the clinical context is available. The main aim of the study was to identify the best technique(s) based on different clinical conditions, evaluating the efficacy and prognostic value of both methodologies. Methods: A retrospective series of patients with tumors involving the motor pathways who underwent surgical resection with the aid of brain mapping and combined MEP monitoring via TES and DCS was evaluated. Irreversible MEP amplitude reduction (>50% compared to baseline) was used as an intraoperative warning and correlated to the postoperative motor outcome. Selectivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were computed for both techniques. Results: Four hundred sixty-two patients were retrospectively analyzed, and only 1.9% showed a long-term motor impairment. Both TES and DCS obtained high specificity and NPV for the acute and 1-month motor deficit. Sensitivity was rather low for the acute deficit but excellent considering the 1-month follow-up for both techniques. DCS was extremely reliable in predicting a postoperative motor decline (PPV of 100% and 90% for acute and long-term deficit, respectively). Conversely, TES produced a high number of false-positive results, especially for long-term deficits (65, 87.8% of all warnings) therefore obtaining poor PPV values (18% and 12% for acute and 1-month deficits, respectively). TES false-positive results were significantly associated with parietal tumors and lateral patient positioning. Conclusions: Data support the use of mapping and combined monitoring via TES and DCS. The sole TES monitoring is reliable in most procedures but not in parietal tumors or those requiring lateral positioning. Although no indications are available in international guidelines, DCS should be recommended, particularly for cases approached by a lateral position.

Preoperative Brain Tumor Imaging: Models and Software for Segmentation and Standardized Reporting.

For patients suffering from brain tumor, prognosis estimation and treatment decisions are made by a multidisciplinary team based on a set of preoperative MR scans. Currently, the lack of standardized and automatic methods for tumor detection and generation of clinical reports, incorporating a wide range of tumor characteristics, represents a major hurdle. In this study, we investigate the most occurring brain tumor types: glioblastomas, lower grade gliomas, meningiomas, and metastases, through four cohorts of up to 4,000 patients. Tumor segmentation models were trained using the AGU-Net architecture with different preprocessing steps and protocols. Segmentation performances were assessed in-depth using a wide-range of voxel and patient-wise metrics covering volume, distance, and probabilistic aspects. Finally, two software solutions have been developed, enabling an easy use of the trained models and standardized generation of clinical reports: Raidionics and Raidionics-Slicer. Segmentation performances were quite homogeneous across the four different brain tumor types, with an average true positive Dice ranging between 80 and 90%, patient-wise recall between 88 and 98%, and patient-wise precision around 95%. In conjunction to Dice, the identified most relevant other metrics were the relative absolute volume difference, the variation of information, and the Hausdorff, Mahalanobis, and object average symmetric surface distances. With our Raidionics software, running on a desktop computer with CPU support, tumor segmentation can be performed in 16-54 s depending on the dimensions of the MRI volume. For the generation of a standardized clinical report, including the tumor segmentation and features computation, 5-15 min are necessary. All trained models have been made open-access together with the source code for both software solutions and validation metrics computation. In the future, a method to convert results from a set of metrics into a final single score would be highly desirable for easier ranking across trained models. In addition, an automatic classification of the brain tumor type would be necessary to replace manual user input. Finally, the inclusion of post-operative segmentation in both software solutions will be key for generating complete post-operative standardized clinical reports.

Intraoperative AIRO mobile computer tomography in frameless stereotactic procedures.

BACKGROUND: Multiple factors can affect the accuracy of neuronavigation, that is a relevant issue, particularly for frameless stereotactic procedures, where precision and optimal image-guidance is crucial for the surgical performance, workflow, and outcome. OBJECTIVE: To investigate the impact of AIRO Mobile Computer Tomography in frameless stereotactic approaches. METHODS: A retrospective study on 12 patients was performed. All the procedures were deployed using a frameless stereotactic technique, both for the collection of biopsy pathological specimens for diagnosis and insertion of drainage in the treatment of intracranial cystic lesions. RESULTS: Twelve patients (eight males, four females) underwent the frameless stereotactic procedure. Mean age at surgery was 55 (±5 SE). The mean volume of the lesion was 23.85 cm3 (±3.13). Six diagnostic biopsies and six cyst drainages were performed. The mean trajectory length was 75.9 ± 11.8 mm. Three posterior fossa lesions (27%) were approached through a retro-sigmoidal burr-hole. A craniotomy for draining a haematoma was performed after detection with AIRO-CT. No permanent neurological dysfunction, in-hospital or 30-day mortality were recorded. CONCLUSION: The AIRO-CT resulted feasible with a potential utility for stereotactic procedures. We showed how it could grant the efficacy of the stereotactic procedures reducing some technical and physical sources of inaccuracy, also enhancing safety and allowing prompt detection and management of intraoperative complications.

Stimulation of frontal pathways disrupts hand muscle control during object manipulation.

The activity of frontal motor areas during hand-object interaction is coordinated by dense communication along specific white matter pathways. This architecture allows the continuous shaping of voluntary motor output but, despite extensive investigation in non-human primate studies, remains poorly understood in humans. Disclosure of this system is crucial for predicting and treatment of motor deficits after brain lesions. For this purpose, we investigated the effect of direct electrical stimulation on white matter pathways within the frontal lobe on hand-object manipulation. This was tested in 34 patients (15 left hemisphere, mean age 42 years, 17 male, 15 with tractography) undergoing awake neurosurgery for frontal lobe tumour removal with the aid of the brain mapping technique. The stimulation outcome was quantified based on hand-muscle activity required by task execution. The white matter pathways responsive to stimulation with an interference on muscles were identified by means of probabilistic density estimation of stimulated sites, tract-based lesion-symptom (disconnectome) analysis and diffusion tractography on the single patient level. Finally, we assessed the effect of permanent tract disconnection on motor outcome in the immediate postoperative period using a multivariate lesion-symptom mapping approach. The analysis showed that stimulation disrupted hand-muscle activity during task execution at 66 sites within the white matter below dorsal and ventral premotor regions. Two different EMG interference patterns associated with different structural architectures emerged: (i) an 'arrest' pattern, characterized by complete impairment of muscle activity associated with an abrupt task interruption, occurred when stimulating a white matter area below the dorsal premotor region. Local middle U-shaped fibres, superior fronto-striatal, corticospinal and dorsal fronto-parietal fibres intersected with this region. (ii) a 'clumsy' pattern, characterized by partial disruption of muscle activity associated with movement slowdown and/or uncoordinated finger movements, occurred when stimulating a white matter area below the ventral premotor region. Ventral fronto-parietal and inferior fronto-striatal tracts intersected with this region. Finally, only resections partially including the dorsal white matter region surrounding the supplementary motor area were associated with transient upper-limb deficit (P = 0.05; 5000 permutations). Overall, the results identify two distinct frontal white matter regions possibly mediating different aspects of hand-object interaction via distinct sets of structural connectivity. We suggest the dorsal region, associated with arrest pattern and postoperative immediate motor deficits, to be functionally proximal to motor output implementation, while the ventral region may be involved in sensorimotor integration required for task execution.

Asleep or awake motor mapping for resection of perirolandic glioma in the nondominant hemisphere? Development and validation of a multimodal score to tailor the surgical strategy.

OBJECTIVE: Resection of glioma in the nondominant hemisphere involving the motor areas and pathways requires the use of brain-mapping techniques to spare essential sites subserving motor control. No clear indications are available for performing motor mapping under either awake or asleep conditions or for the best mapping paradigm (e.g., resting or active, high-frequency [HF] or low-frequency [LF] stimulation) that provides the best oncological and functional outcomes when tailored to the clinical context. This work aimed to identify clinical and imaging factors that influence surgical strategy (asleep motor mapping vs awake motor mapping) and that are associated with the best functional and oncological outcomes and to design a "motor mapping score" for guiding tumor resection in this area. METHODS: The authors evaluated a retrospective series of patients with nondominant-hemisphere glioma-located or infiltrating within 2 cm anteriorly or posteriorly to the central sulcus and affecting the primary motor cortex, its fibers, and/or the praxis network-who underwent operations with asleep (HF monopolar probe) or awake (LF and HF probes) motor mapping. Clinical and imaging variables were used to design a motor mapping score. A prospective series of patients was used to validate this motor mapping score. RESULTS: One hundred thirty-five patients were retrospectively analyzed: 69 underwent operations with asleep (HF stimulation) motor mapping, and 66 underwent awake (LF and HF stimulation and praxis task evaluation) motor mapping. Previous motor (strength) deficit, previous treatment (surgery/radiotherapy), tumor volume > 30 cm3, and tumor involvement of the praxis network (on MRI) were identified and used to design the mapping score. Motor deficit, previous treatment, and location within or close to the central sulcus favor use of asleep motor mapping; large tumor volume and involvement of the praxis network favor use of awake motor mapping. The motor mapping score was validated in a prospective series of 52 patients-35 underwent operations with awake motor mapping and 17 with asleep motor mapping on the basis of the score indications-who had a low rate of postoperative motor-praxis deficit (3%) and a high extent of resection (median 97%; complete resection in > 70% of patients). CONCLUSIONS: Extensive resection of tumor involving the eloquent areas for motor control is feasible, and when an appropriate mapping strategy is applied, the incidence of postoperative motor-praxis deficit is low. Asleep (HF stimulation) motor mapping is preferable for lesions close to or involving the central sulcus and/or in patients with preoperative strength deficit and/or history of previous treatment. When a patient has no motor deficit or previous treatment and has a lesion (> 30 cm3) involving the praxis network, awake mapping is preferable.

On the cutting edge of glioblastoma surgery: where neurosurgeons agree and disagree on surgical decisions.

OBJECTIVE: The aim of glioblastoma surgery is to maximize the extent of resection while preserving functional integrity. Standards are lacking for surgical decision-making, and previous studies indicate treatment variations. These shortcomings reflect the need to evaluate larger populations from different care teams. In this study, the authors used probability maps to quantify and compare surgical decision-making throughout the brain by 12 neurosurgical teams for patients with glioblastoma. METHODS: The study included all adult patients who underwent first-time glioblastoma surgery in 2012-2013 and were treated by 1 of the 12 participating neurosurgical teams. Voxel-wise probability maps of tumor location, biopsy, and resection were constructed for each team to identify and compare patient treatment variations. Brain regions with different biopsy and resection results between teams were identified and analyzed for patient functional outcome and survival. RESULTS: The study cohort consisted of 1087 patients, of whom 363 underwent a biopsy and 724 a resection. Biopsy and resection decisions were generally comparable between teams, providing benchmarks for probability maps of resections and biopsies for glioblastoma. Differences in biopsy rates were identified for the right superior frontal gyrus and indicated variation in biopsy decisions. Differences in resection rates were identified for the left superior parietal lobule, indicating variations in resection decisions. CONCLUSIONS: Probability maps of glioblastoma surgery enabled capture of clinical practice decisions and indicated that teams generally agreed on which region to biopsy or to resect. However, treatment variations reflecting clinical dilemmas were observed and pinpointed by using the probability maps, which could therefore be useful for quality-of-care discussions between surgical teams for patients with glioblastoma.

Glioblastoma Surgery Imaging-Reporting and Data System: Validation and Performance of the Automated Segmentation Task.

For patients with presumed glioblastoma, essential tumor characteristics are determined from preoperative MR images to optimize the treatment strategy. This procedure is time-consuming and subjective, if performed by crude eyeballing or manually. The standardized GSI-RADS aims to provide neurosurgeons with automatic tumor segmentations to extract tumor features rapidly and objectively. In this study, we improved automatic tumor segmentation and compared the agreement with manual raters, describe the technical details of the different components of GSI-RADS, and determined their speed. Two recent neural network architectures were considered for the segmentation task: nnU-Net and AGU-Net. Two preprocessing schemes were introduced to investigate the tradeoff between performance and processing speed. A summarized description of the tumor feature extraction and standardized reporting process is included. The trained architectures for automatic segmentation and the code for computing the standardized report are distributed as open-source and as open-access software. Validation studies were performed on a dataset of 1594 gadolinium-enhanced T1-weighted MRI volumes from 13 hospitals and 293 T1-weighted MRI volumes from the BraTS challenge. The glioblastoma tumor core segmentation reached a Dice score slightly below 90%, a patientwise F1-score close to 99%, and a 95th percentile Hausdorff distance slightly below 4.0 mm on average with either architecture and the heavy preprocessing scheme. A patient MRI volume can be segmented in less than one minute, and a standardized report can be generated in up to five minutes. The proposed GSI-RADS software showed robust performance on a large collection of MRI volumes from various hospitals and generated results within a reasonable runtime.

Targeting Primary Motor Cortex (M1) Functional Components in M1 Gliomas Enhances Safe Resection and Reveals M1 Plasticity Potentials.

Primary-Motor-Cortex (M1) hosts two functional components, at its posterior and anterior borders, being the first faster and more excitable. We developed a mapping-technique for M1 components identification and determined their functional cortical-subcortical architecture in M1 gliomas and the impact of their identification on tumor resection and motor performance. A novel advanced mapping technique was used in 102 tumors within M1 or CorticoSpinal-Tract to identify M1-two components. High-Frequency-stimulation (2-5 pulses) with an on-line qualitative and quantitative analysis of motor responses was used; the two components' cortical/subcortical spatial distribution correlated to clinical, tumor-related factor and patients' motor outcome; a cohort treated with standard-mapping was used for comparison. The two functional components were always identified on-line; in tumors not affecting M1, its functional segregation was preserved. In M1 tumors, two architectures, both preserving the two components, were disclosed: in 50%, a normal cortical/subcortical architecture emerged, while 50% revealed a distorted architecture with loss of anatomical reference and somatotopy, not associated with tumor histo-molecular features or volume, but with a previous treatment. Motor performance was maintained, suggesting functional compensation. By preserving the highest and resecting the lowest excitability component, the complete-resection increased with low morbidity. The real-time identification of two M1 functional components and the preservation of the highest excitability one increases safe resection, revealing M1 plasticity potentials.

Raman Spectroscopy and Machine Learning for IDH Genotyping of Unprocessed Glioma Biopsies.

Isocitrate dehydrogenase (IDH) mutational status is pivotal in the management of gliomas. Patients with IDH-mutated (IDH-MUT) tumors have a better prognosis and benefit more from extended surgical resection than IDH wild-type (IDH-WT). Raman spectroscopy (RS) is a minimally invasive optical technique with great potential for intraoperative diagnosis. We evaluated the RS's ability to characterize the IDH mutational status onto unprocessed glioma biopsies. We extracted 2073 Raman spectra from thirty-eight unprocessed samples. The classification performance was assessed using the eXtreme Gradient Boosted trees (XGB) and Support Vector Machine with Radial Basis Function kernel (RBF-SVM). Measured Raman spectra displayed differences between IDH-MUT and IDH-WT tumor tissue. From the 103 Raman shifts screened as input features, the cross-validation loop identified 52 shifts with the highest performance in the distinction of the two groups. Raman analysis showed differences in spectral features of lipids, collagen, DNA and cholesterol/phospholipids. We were able to distinguish between IDH-MUT and IDH-WT tumors with an accuracy and precision of 87%. RS is a valuable and accurate tool for characterizing the mutational status of IDH mutation in unprocessed glioma samples. This study improves RS knowledge for future personalized surgical strategy or in situ target therapies for glioma tumors.

Glioblastoma Surgery Imaging-Reporting and Data System: Standardized Reporting of Tumor Volume, Location, and Resectability Based on Automated Segmentations.

Treatment decisions for patients with presumed glioblastoma are based on tumor characteristics available from a preoperative MR scan. Tumor characteristics, including volume, location, and resectability, are often estimated or manually delineated. This process is time consuming and subjective. Hence, comparison across cohorts, trials, or registries are subject to assessment bias. In this study, we propose a standardized Glioblastoma Surgery Imaging Reporting and Data System (GSI-RADS) based on an automated method of tumor segmentation that provides standard reports on tumor features that are potentially relevant for glioblastoma surgery. As clinical validation, we determine the agreement in extracted tumor features between the automated method and the current standard of manual segmentations from routine clinical MR scans before treatment. In an observational consecutive cohort of 1596 adult patients with a first time surgery of a glioblastoma from 13 institutions, we segmented gadolinium-enhanced tumor parts both by a human rater and by an automated algorithm. Tumor features were extracted from segmentations of both methods and compared to assess differences, concordance, and equivalence. The laterality, contralateral infiltration, and the laterality indices were in excellent agreement. The native and normalized tumor volumes had excellent agreement, consistency, and equivalence. Multifocality, but not the number of foci, had good agreement and equivalence. The location profiles of cortical and subcortical structures were in excellent agreement. The expected residual tumor volumes and resectability indices had excellent agreement, consistency, and equivalence. Tumor probability maps were in good agreement. In conclusion, automated segmentations are in excellent agreement with manual segmentations and practically equivalent regarding tumor features that are potentially relevant for neurosurgical purposes. Standard GSI-RADS reports can be generated by open access software.