Greco, Elena

Abstract Background Accurate synthesis of computed tomography (CT) images from magnetic resonance imaging (MRI) is clinically valuable for cranial applications such as attenuation correction, radiotherapy planning, and image‐guided interventions. However, the heterogeneity of MRI data across magnetic field strengths and pulse sequences limits the generalizability of existing methods, posing a barrier to clinical translation. Purpose This study aims to reformulate cranial CT synthesis as a modular, structurally coupled problem using a deep learning approach, enhancing adaptability across heterogeneous MRI conditions, including different field strengths and sequence protocols. Methods We implemented a cascaded multitask pipeline that jointly models skull segmentation and Hounsfield Unit (HU) regression in anatomically targeted regions. A 3D patch‐based training paradigm shifts the modeling focus from global image translation to localized bone feature extraction. The backbone leverages a residual Mamba‐based state space model within a 3D U‐Net structure to improve spatial representation, while a Transformer U‐Net serves as a widely recognized reference baseline comparison. The model was trained and evaluated in terms of multi‐modal (T1‐weighted and T2‐FLAIR) and cross‐domain transferability from a public 1.5 Tesla brain dataset ( n = 37) to an independent 7 Tesla clinical brain dataset ( n = 44). Performance was assessed using Dice and Jaccard indices for segmentation accuracy and mean absolute error (MAE) for HU regression. Results Quantitative analysis showed that our multitask pipeline, incorporating both morphological and HU map prediction stages, significantly outperformed conventional direct MRI‐to‐CT mapping across all metrics on the public 1.5T dataset (p < 0.05). Consistent performance on an external 7T clinical dataset (p < 0.001) further demonstrated the method's robustness and adaptability across field strengths. Conclusion These findings support that task‐structured training for modality transformation markedly improves both accuracy and generalizability of cranial CT synthesis across heterogeneous MRI conditions. The observed consistency across field strengths validates the robustness of the proposed methodology.

BACKGROUND AND OBJECTIVES: Rising US healthcare costs and the COVID-19 pandemic put increased pressure on hospitals to reduce patient length of stay (LOS) while maintaining high-quality care. We aimed to determine if implementing a pathway for same-day discharge after elective neuroendovascular procedures to a monitored hotel setting is safe, feasible, and cost-effective. METHODS: We evaluated patient outcomes after elective neuroendovascular procedures and same-day discharge to an on-campus Care Hotel. The study enrolled patients aged ≥18 years from October 1, 2020, to February 28, 2023. Exclusion criteria comprised major comorbidities, additional care necessitating hospitalization, absence of a caregiver, insurer-mandated hospitalization, or patient refusal of the Care Hotel model. Primary outcomes were readmission rate or prolonged LOS, new ischemic or hemorrhagic complications rate, and death rate from vascular causes within 6 months. Secondary outcomes included adverse events and deaths related to causes other than vascular complications within 6 months and changes in modified Rankin Scale score by 1 or more points at last follow-up. Healthcare cost savings associated with same-day discharge were analyzed using a control group. RESULTS: The study included 42 patients discharged to the Care Hotel and 84 controls. The average age was 61 years in the Care Hotel group and 64 years in the control group, and aneurysm treatment was the most frequent neuroendovascular procedure (83.3% and 79.8%, respectively). The readmission and extended LOS rates were 4.8% for the Care Hotel group. The rate of new ischemic or hemorrhagic complications was 4.8%. Cost analysis estimated savings of $7730.12 per procedure. CONCLUSION: This study suggests that the Care Hotel model is a safe, feasible, and cost-effective option for carefully selected patients undergoing elective neuroendovascular interventions. Adherence to discharge readiness criteria and meticulous patient selection are critical to minimizing complications and readmissions.

OBJECTIVE Pseudoaneurysms (PSAs) are complex vascular lesions. Flow diversion has been proposed as an alternative treatment to parent artery occlusion that preserves laminar flow. The authors of the present study investigated the safety and short-term (< 1 year) and long-term (≥ 1 year) aneurysm occlusion rates following the treatment of intracranial and extracranial PSAs using the Pipeline embolization device (PED). METHODS An electronic database search for full-text English-language articles in Ovid MEDLINE and Epub Ahead of Print, Ovid Embase, Ovid Cochrane Central Register of Controlled Trials, Ovid Cochrane Database of Systematic Reviews, and Scopus was conducted following the PRISMA guidelines. Studies of any design including at least 4 patients with intracranial or extracranial PSAs treated using a PED were included in this analysis. The primary outcome of interest was the rate of peri- and postprocedural complications. Secondarily, the authors analyzed the incidence of complete aneurysm occlusion. RESULTS A total of 90 patients with 96 PSAs across 9 studies were included. The mean age was 38.2 (SD 15.14) years, and 37.8% of the patients were women. The mean PSA size was 4.9 mm. Most PSAs were unruptured, and the most common etiology was trauma (n = 32, 35.5%), followed by spontaneous formation (n = 21, 23.3%) and iatrogenic injury (n = 19, 21.1%). Among the 51 (53.1%) intracranial and 45 (46.9%) extracranial PSAs were 19 (19.8%) dissecting PSAs. Sixty-six (77.6%) PSAs were in the internal carotid artery and 10 (11.8%) in the vertebral artery. Thirty-three (34.4%) PSAs were treated with ≥ 2 devices, and 8 (8.3%) underwent adjunctive coiling. The mean clinical and angiographic follow-up durations were 10.7 and 12.9 months, respectively. The short-term (< 1 year) and long-term (≥ 1 year) complete occlusion rates were 79% (95% CI 66%–88%, p = 0.82) and 84% (95% CI 70%–92%, p = 0.95), respectively. Complication rates were 8% for iatrogenic dissection (95% CI 3%–16%, p = 0.94), 10% for silent thromboembolism (95% CI 5%–21%, p = 0.77), and 12% for symptomatic thromboembolism (95% CI 6%–23%, p = 0.48). No treatment-related hemorrhage was observed. The overall mortality rate at the last follow-up was 14%. CONCLUSIONS The complete occlusion rate for PSAs treated with the PED was high and increased over time. Although postprocedural complications and mortality were not insignificant, flow diversion represents a reasonably safe option for managing these complex lesions.

The craniovertebral junction (CVJ) is a complex transition area between the skull and cervical spine. Pathologies such as chordoma, chondrosarcoma and aneurysmal bone cysts may be encountered in this anatomical area and may predispose individuals to joint instability. An adequate clinical and radiological assessment is mandatory to predict any postoperative instability and the need for fixation. There is no common consensus on the need for, timing and setting of craniovertebral fixation techniques after a craniovertebral oncological surgery. The aim of the present review is to summarize the anatomy, biomechanics and pathology of the craniovertebral junction and to describe the available surgical approaches to and considerations of joint instability after craniovertebral tumor resections. Although a one-size-fits-all approach cannot encompass the extremely challenging pathologies encountered in the CVJ area, including the possible mechanical instability that is a consequence of oncological resections, the optimal surgical strategy (anterior vs posterior vs posterolateral) tailored to the patient’s needs can be assessed preoperatively in many instances. Preserving the intrinsic and extrinsic ligaments, principally the transverse ligament, and the bony structures, namely the C1 anterior arch and occipital condyle, ensures spinal stability in most of the cases. Conversely, in situations that require the removal of those structures, or in cases where they are disrupted by the tumor, a thorough clinical and radiological assessment is needed to timely detect any instability and to plan a surgical stabilization procedure. We hope that this review will help shed light on the current evidence and pave the way for future studies on this topic.

OBJECTIVE The authors hypothesized that the proximity of deep brain stimulator contacts to the anterior thalamic nucleus–mammillothalamic tract (ANT-MMT) junction determines responsiveness to treatment with ANT deep brain stimulation (DBS) in drug-resistant epilepsy and conducted this study to test that hypothesis. METHODS This retrospective study evaluated patients who had undergone ANT DBS electrode implantation and whose devices were programmed to stimulate nearest the ANT-MMT junction based on direct MRI visualization. The proximity of the active electrode to the ANT and the ANT-MMT junction was compared between responders (≥ 50% reduction in seizure frequency) and nonresponders. Linear regression was performed to assess the percentage of seizure reduction and distance to both the ANT and the ANT-MMT junction. RESULTS Four (57.1%) of 7 patients had ≥ 50% reduction in seizures. All 4 responders had at least one contact within 1 mm of the ANT-MMT junction, whereas the 3 patients with < 50% seizure improvement did not have a contact within 1 mm of the ANT-MMT junction. Additionally, the 4 responders demonstrated contact positioning closer to the ANT-MMT junction than the 3 nonresponders (mean distance from MMT: 0.7 mm on the left and 0.6 mm on the right in responders vs 3.0 mm on the left and 2.3 mm on the right in nonresponders). However, proximity of the electrode contact to any point in the ANT nucleus did not correlate with seizure reduction. Greater seizure improvement was correlated with a contact position closer to the ANT-MMT junction (R2 = 0.62, p = 0.04). Seizure improvement was not significantly correlated with proximity of the contact to any ANT border (R2 = 0.24, p = 0.26). CONCLUSIONS Obtained using a combination of direct visualization and targeted programming of the ANT-MMT junction, data in this study support the hypothesis that proximity to the ANT alone does not correlate with seizure reduction in ANT DBS, whereas proximity to the ANT-MMT junction does. These findings support the importance of direct targeting in ANT DBS, as well as imaging-informed programming. Additionally, the authors provide supportive evidence for future prospective trials using ANT-MMT junction for direct surgical targeting.