Blezek, Daniel J.

Background/Objectives: Characterizing spinal cord multiple sclerosis (MS) lesions in MRI is critical for diagnosis, monitoring, and treatment evaluation. However, current automated approaches for lesion detection and segmentation are typically designed for specific MRI contrasts or acquisition sites, limiting their generalizability in real-world clinical settings where imaging protocols vary widely. This work proposes a robust multi-site, multi-contrast segmentation framework for spinal cord lesions. Methods: The segmentation model was trained and evaluated on a large-scale dataset comprising 4428 annotated images from 1849 persons with MS across 23 imaging centers, encompassing six MRI contrasts (T1w, T2w, T2*w, PSIR, STIR, and UNIT1) acquired at 1.5 tesla (T), 3 T, and 7 T. Results: Likert-type assessment performed by neuroradiologist ratings demonstrated superior generalization of the model compared to existing contrast-specific pipelines ( p < 0.01). Additional experiments evaluated robustness across spinal levels, acquisition resolutions, binarization thresholds, and quantitative evaluation on external labeled datasets. Conclusions: The proposed model can achieve accurate and reliable spinal cord MS lesion segmentation across heterogeneous MRI data, addressing a key barrier to clinical translation. The model is available in the Spinal Cord Toolbox v7.2 and higher. Code repository: https://github.com/ivadomed/seg-sc-ms-lesion-multicontrast

Abstract BACKGROUND Patients with primary brain tumors often have very complex histories, but an accurate understanding of their care timeline is critical to provide the best care possible. Extracting pertinent facts from the medical record can be time consuming and frustrating. We sought to apply generative AI large language models (LLM) to this problem in order to supply providers with concise, accurate, and focused data including graphs and tables. METHODS After evaluating multiple LLMs, GEMINI 2.5 Pro was selected. Prompt engineering was iteratively refined across a range of patients to generate concise outputs using tables, graphs, and prose to describe timing of chemoradiation, tumor markers, procedures, and trends in MRI results. The model inferred BT-RADS scores from the MRI report prose. Using a 5-point Likert scale, neuroradiologists assessed the AI-output both prospectively and retrospectively comparing it to the medical record. LLM’s compared AI summaries to source material. RESULTS Six neuroradiologists each ranked 25 prospective patient outputs as accurate (82%/16%/1%), useful (94%/6%/0%), trustworthy (86%/13%/1%), and satisfactory (86%/10%/1%) for Likert ratings of highly agree/agree/neutral, respectively. No responses were disagree or highly disagree. Interobserver agreement was high (Gwent AC2 &gt; 0.95). The LLM evaluation of referenced portions of the record determined &gt; 85% of summary statements were supported by source documents. While users estimated time savings of a 1-10 minutes per case, they felt the greater benefit was having more organized data presented efficiently. CONCLUSIONS AI summaries of the medical record of complex patients with primary brain tumors give providers an accurate, well-organized display of the pertinent facts necessary for high quality care and save time in chart review. These summaries, which are generated early each morning, have been found to be helpful and are now being incorporated into our clinical practice and multidisciplinary tumor board.

Abstract Background Neurofibromatosis type 2 (NF2)-related schwannomatosis is an autosomal dominant tumor-predisposition syndrome characterized by bilateral vestibular schwannomas (VS). In patients with VS associated with NF2, vascular endothelial growth factor A inhibitor, bevacizumab, is a systemic treatment option. The aim of this study is to retrospectively evaluate NF2 patient responses to bevacizumab on VS growth and symptom progression. Methods This is a retrospective analysis of patients seen at the Mayo Clinic Rochester Multidisciplinary NF2 Clinic. Results Out of 76 patients with NF2 evaluated between 2020 and 2022, we identified 19 that received treatment with bevacizumab. Thirteen of these patients discontinued bevacizumab after median treatment duration of 12.2 months. The remaining 6 patients are currently receiving bevacizumab treatment for a median duration of 9.4 months as of March, 2023. Fifteen patients had evaluable brain MRI data, which demonstrated partial responses in 5 patients, stable disease in 8, and progression in 2. Within 6 months of bevacizumab discontinuation, 5 patients had rebound growth of their VS greater than 20% from their previous tumor volume, while 3 did not. Three patients with rebound growth went on to have surgery or irradiation for VS management. Conclusions Our single-institution experience confirms prior studies that bevacizumab can control progression of VS and symptoms associated with VS growth. However, we note that there is the potential for rapid VS growth following bevacizumab discontinuation, for which we propose heightened surveillance imaging and symptom monitoring for at least 6 months upon stopping anti-VEGF therapy.

Abstract BACKGROUND Neurofibromatosis type 2-related schwannomatosis (NF2) is an autosomal dominant tumor-predisposition syndrome characterized by bilateral vestibular schwannomas (VS) as well as other nervous system tumors. Management decisions for NF2-related VS are typically predicated on tumor size, associated symptoms, patient age, and comorbidities. Observation, surgical resection, and stereotactic radiosurgery can be viable options. In patients with VS associated with NF2, vascular endothelial growth factor A (VEGF-A) inhibitor, bevacizumab, is a systemic treatment option. In this single institution study, we review our clinical experience in patients with NF2 who received bevacizumab for VS and evaluate the effect of bevacizumab treatment on VS growth and symptom progression. MATERIAL AND METHODS This is a retrospective analysis of patients seen at the Mayo Clinic Rochester Multidisciplinary NF2 Clinic. Three-dimensional volumetric data for VS were measured using manual segmentation with Visage PACS software under the direct supervision of a board-certified neuroradiologist. RESULTS Out of 76 patients with NF2 evaluated in the past 3 years, we identified 19 that received treatment with bevacizumab. Median age of NF2 diagnosis was 19 years (range 0.5-61) with a median NF2 phenotype severity score of 2 (range 1-3). Thirteen of these patients have discontinued bevacizumab, accounting for a median duration of 12.2 months (range 3.3-88 months) of bevacizumab treatment. Bevacizumab was discontinued due to hypertension (n=5), proteinuria and renal dysfunction (n=3), tumor progression (n=2), fatigue or other intolerance (n=2), or non-compliance (n=1). The remaining 6 patients are currently receiving bevacizumab treatment for a median duration of 9.4 months as of March 2023 (range 3.7-48.3 months). Fifteen patients had brain MRI data available to evaluate VS change during bevacizumab treatment, with partial response observed in 5 patients, stable disease in 8, and progression in 2. Ten of the 13 patients who stopped bevacizumab had MRI data to evaluate VS change following treatment termination. Within 6 months of bevacizumab discontinuation, 6 patients had rebound growth of their VS between 21-95% from their previous tumor volume, while 4 did not. Three patients with rebound growth went on to have surgery or irradiation for VS management. CONCLUSION Our single institution experience confirms that bevacizumab can slow progression of VS and symptoms associated with VS growth. However, we note that there is the potential for rapid VS growth following bevacizumab discontinuation, for which we propose heightened surveillance imaging and symptom monitoring for at least 6 months upon stopping anti-VEGF therapy.

Abstract Background: Sarcopenia is one of the early pathological features of cancer cachexia, yet the effect of sarcopenia on clinical outcomes in patients with advanced HCC is unclear. Our objective was to determine the effect of Sarcopenia on response to systemic therapy with advanced HCC. Methods: Patients with unresectable and advanced liver cancer were retrospectively evaluated between 2010 to 2019. Skeletal muscle area (SMA) was computed with a previously validated computed tomography (CT) body composition AI algorithm. The skeletal muscle index (SMI) was derived from the skeletal muscle area (SMA) (cm2) divided by patient height (m2) calculated from images at the level of L3 on pretreatment CT. Sarcopenia cutoffs for females was less than 39 cm2/m2 SMI or less than 50 cm2/m2 SMI for males. Patients’ demographics, systemic treatment response, toxicity, progression-free survival (PFS) and overall survival (OS) were compared between first-line therapy groups. Kaplan-Meier and Cox proportional hazards regression were used for survival analysis comparisons. Results: A total of 146 patients with advanced HCC were assessed from Mayo Clinic Arizona and Emory University. Seventy-six patients received immunotherapy as first-line treatment and seventy patients received TKI as first-line therapy. Sarcopenia at baseline was assessed in a total of 91 patients at first-line systemic treatment: n=48 with immunotherapy (IO) and n=43 with tyrosine kinase inhibitor (TKI). Sarcopenia was more prevalent in male patients (p=0.009), and in patients with low BMI (p=0.02). When comparing those with sarcopenia at baseline, higher response rate was observed in patients treated with IO without sarcopenia (47% versus 15%, p=0.019), this was not observed in TKI group. Interestingly, median PFS and OS was not significantly changed in patients with or without sarcopenia treated with IO or TKI on first-line therapy. Conclusions: Sarcopenia was associated with reduced response rate (RR) in patients treated with IO, but not in TKI groups. Both didn’t translate into deleterious effect in PFS and or OS. Further research with sufficient adjustments for confounding factors is warranted to better elucidate the prognostic value of sarcopenia in these patients. Citation Format: Gehan Botrus, Pedro Luiz Serrano Uson Junior, Heidi Kosiorek, Mehmet Akce, Isabela Chang, Zhubene Mesbah, Eric Yancey, Daniel Blezek, Jason Klug, Lana Khalil, Puneet Raman, Mohamad Sonbol, Mitesh Borad, Daniel Ahn, Tanios Bekaii-Saab. Prognostic impact of Sarcopenia on clinical outcomes in advanced hepatocellular carcinoma (HCC) treated with systemic therapy [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 6274.

AbstractIntroductionSacral tumor resection is known for a high rate of complications. Sarcopenia has been found to be associated with wound complications; however, there is a paucity of data examining the impact of sarcopenia on the outcome of sacral tumor resection.MethodsForty‐eight patients (31 primary sarcomas, 17 locally recurrent carcinomas) undergoing sacrectomy were reviewed. Central sarcopenia was assessed by measuring the psoas:lumbar vertebra index (PLVI), with the 50th percentile (0.97) used to determine which patients were high (>0.97) versus low (<0.97).ResultsTwenty‐four (50%) patients had a high PLVI and 24 (50%) had a low PLVI (sarcopenic). There was no difference (p > 0.05) in the demographics of patients with or without sarcopenia. There was no difference in the incidence of postoperative wound complications (odds ratio [OR] = 1.0, p = 1.0) or deep infection (OR = 0.83, p = 1.0). Sarcopenia was not associated with death due to disease (hazard ratio [HR] = 2.04, p = 0.20) or metastatic disease (HR = 2.47, p = 0.17), but was associated with local recurrence (HR = 6.60, p = 0.01).ConclusionsCentral sarcopenia was not predictive of wound complications or infection following sacral tumor resection. Sarcopenia was, however, an independent risk factor for local tumor recurrence following sacrectomy and should be considered when counseling patients on the outcome of sacrectomy.

Abstract Ultrasound localization microscopy (ULM) has been proposed to image microvasculature beyond the ultrasound diffraction limit. Although ULM can attain microvascular images with a sub-diffraction resolution, long data acquisition time and processing time are the critical limitations. Deep learning-based ULM (deep-ULM) has been proposed to mitigate these limitations. However, microbubble (MB) localization used in deep-ULMs is currently based on spatial information without the use of temporal information. The highly spatiotemporally coherent MB signals provide a strong feature that can be used to differentiate MB signals from background artifacts. In this study, a deep neural network was employed and trained with spatiotemporal ultrasound datasets to better identify the MB signals by leveraging both the spatial and temporal information of the MB signals. Training, validation and testing datasets were acquired from MB suspension to mimic the realistic intensity-varying and moving MB signals. The performance of the proposed network was first demonstrated in the chicken embryo chorioallantoic membrane dataset with an optical microscopic image as the reference standard. Substantial improvement in spatial resolution was shown for the reconstructed super-resolved images compared with power Doppler images. The full-width-half-maximum (FWHM) of a microvessel was improved from 133 μ m to 35 μ m, which is smaller than the ultrasound wavelength (73 μ m). The proposed method was further tested in an in vivo human liver data. Results showed the reconstructed super-resolved images could resolve a microvessel of nearly 170 μ m (FWHM). Adjacent microvessels with a distance of 670 μ m, which cannot be resolved with power Doppler imaging, can be well-separated with the proposed method. Improved contrast ratios using the proposed method were shown compared with that of the conventional deep-ULM method. Additionally, the processing time to reconstruct a high-resolution ultrasound frame with an image size of 1024 × 512 pixels was around 16 ms, comparable to state-of-the-art deep-ULMs.

The thalamic ventral intermediate nucleus (VIM) can be targeted for treatment of tremor by several procedures, including deep brain stimulation (DBS) and, more recently, MR-guided focused ultrasound (MRgFUS). To date, such targeting has relied predominantly on coordinate-based or atlas-based techniques rather than directly targeting the VIM based on imaging features. While general regional differences of features within the thalamus and some related white matter tracts can be distinguished with conventional imaging techniques, internal nuclei such as the VIM are not discretely visualized. Advanced imaging methods such as quantitative susceptibility mapping (QSM) and fast gray matter acquisition T1 inversion recovery (FGATIR) MRI and high-field MRI pulse sequences that improve the ability to image the VIM region are emerging but have not yet been shown to have reliability and accuracy to serve as the primary method of VIM targeting. Currently, the most promising imaging approach to directly identify the VIM region for clinical purposes is MR diffusion tractography. In this review and update, the capabilities and limitations of conventional and emerging advanced methods for evaluation of internal thalamic anatomy are briefly reviewed. The basic principles of tractography most relevant to VIM targeting are provided for familiarization. Next, the key literature to date addressing applications of DTI and tractography for DBS and MRgFUS is summarized, emphasizing use of direct targeting. This literature includes 1-tract (dentatorubrothalamic tract [DRT]), 2-tract (pyramidal and somatosensory), and 3-tract (DRT, pyramidal, and somatosensory) approaches to VIM region localization through tractography. The authors introduce a 3-tract technique used at their institution, illustrating the oblique curved course of the DRT within the inferior thalamus as well as the orientation and relationship of the white matter tracts in the axial plane. The utility of this 3-tract tractography approach to facilitate VIM localization is illustrated with case examples of variable VIM location, targeting superior to the anterior commissure–posterior commissure plane, and treatment in the setting of pathologic derangement of thalamic anatomy. Finally, concepts demonstrated with these case examples and from the prior literature are synthesized to highlight several potential advantages of tractography for VIM region targeting.

Abstract Magnetic resonance guided high intensity focused ultrasound is a novel, non-invasive, image-guided procedure that is able to ablate intracranial tissue with submillimetre precision. It is currently FDA approved for essential tremor and tremor dominant Parkinson’s disease. The aim of this update is to review the limitations of current landmark-based targeting techniques of the ventral intermediate nucleus and demonstrate the role of emerging imaging techniques that are relevant for both magnetic resonance guided high intensity focused ultrasound and deep brain stimulation. A significant limitation of standard MRI sequences is that the ventral intermediate nucleus, dentatorubrothalamic tract, and other deep brain nuclei cannot be clearly identified. This paper provides original, annotated images demarcating the ventral intermediate nucleus, dentatorubrothalamic tract, and other deep brain nuclei on advanced MRI sequences such as fast grey matter acquisition T1 inversion recovery, quantitative susceptibility mapping, susceptibility weighted imaging, and diffusion tensor imaging tractography. Additionally, the paper reviews clinical efficacy of targeting with these novel MRI techniques when compared to current established landmark-based targeting techniques. The paper has widespread applicability to both deep brain stimulation and magnetic resonance guided high intensity focused ultrasound.