Artificial Intelligence Model Trained with Sparse Data to Detect Facial and Cranial Bone Fractures from Head CT.

The presence of cranial and facial bone fractures is an important finding on non-enhanced head computed tomography (CT) scans from patients who have sustained head trauma. Some prior studies have proposed automatic cranial fracture detections, but studies on facial fractures are lacking. We propose a deep learning system to automatically detect both cranial and facial bone fractures. Our system incorporated models consisting of YOLOv4 for one-stage fracture detection and improved ResUNet (ResUNet++) for the segmentation of cranial and facial bones. The results from the two models mapped together provided the location of the fracture and the name of the fractured bone as the final output. The training data for the detection model were the soft tissue algorithm images from a total of 1,447 head CT studies (a total of 16,985 images), and the training data for the segmentation model included 1,538 selected head CT images. The trained models were tested on a test dataset consisting of 192 head CT studies (a total of 5,890 images). The overall performance achieved a sensitivity of 88.66%, a precision of 94.51%, and an F1 score of 0.9149. Specifically, the cranial and facial regions were evaluated and resulted in a sensitivity of 84.78% and 80.77%, a precision of 92.86% and 87.50%, and F1 scores of 0.8864 and 0.8400, respectively. The average accuracy for the segmentation labels concerning all predicted fracture bounding boxes was 80.90%. Our deep learning system could accurately detect cranial and facial bone fractures and identify the fractured bone region simultaneously.

Inferior Vena Cava Volume Is an Independent Predictor of Massive Transfusion in Patients With Trauma.

OBJECTIVES: Early adequate resuscitation of patients with trauma is crucial in preventing shock and early mortality. Thus, we aimed to determine the performance of the inferior vena cava (IVC) volume and other risk factors and scores in predicting massive transfusion and mortality. METHODS: We included all patients with trauma who underwent computed tomography (CT) scan of the torso, which included the abdominal area, in our emergency department (ED) from January 2014 to January 2017. We calculated the 3-dimensional IVC volume from the left renal vein to the IVC bifurcation. The primary outcome was the performance of IVC volume in predicting massive transfusion, and the secondary outcome was the performance of IVC volume in predicting 24-hour and 30-day in-hospital mortality. RESULTS: Among the 236 patients with trauma, 7.6% received massive transfusions. The IVC volume and revised trauma score (RTS) were independent predictors of massive transfusion (adjusted odds ratio [OR]: 0.79 vs 1.86, 95% confidence interval [CI], 0.71-0.89 vs 1.4-2.47, respectively). Both parameters showed the good area under the curve (AUC) for the prediction of massive transfusion (adjusted AUC: 0.83 and 0.82, 95% CI, 0.74-0.92 vs 0.72-0.93, respectively). Patients with a large IVC volume (fourth quartile) were less likely to receive massive transfusion than those with a small IVC volume (first quartile, ≥28.29 mL: 0% vs <15.08 mL: 20.3%, OR: 0.13, 95% CI, 0.03-0.66). CONCLUSIONS: The volume of IVC measured on CT scan and RTS are independent predictors of massive transfusion in patients with trauma in the ED.

Three-Dimensional Volumetric Changes and Clinical Outcomes after Decompression with DIAM™ Implantation in Patients with Degenerative Lumbar Spine Diseases.

Background and objectives: The prevalence of degenerative lumbar spine diseases has increased. In addition to standard lumbar decompression and/or fusion techniques, implantation of interspinous process devices (IPDs) can provide clinical benefits in highly selected patients. However, changes in spinal structures after IPD implantation using magnetic resonance imaging (MRI) have rarely been discussed. This volumetric study aimed to evaluate the effect of IPD implantation on the intervertebral disc and foramen using three-dimensional assessment. Materials and Methods: We retrospectively reviewed patients with lumbar degenerative disc diseases treated with IPD implantation and foraminotomy and/or discectomy between January 2016 and December 2019. The mean follow-up period was 13.6 months. The perioperative lumbar MRI data were processed for 3D-volumetric analysis. Clinical outcomes, including the Prolo scale and visual analog scale (VAS) scores, and radiographic outcomes, such as the disc height, foraminal area, and translation, were analyzed. Results: Fifty patients were included in our study. At the one-year follow-up, the VAS and Prolo scale scores significantly improved (both p < 0.001). The disc height and foraminal area on radiographs also increased significantly, but with limited effects up to three months postoperatively. MRI revealed an increased postoperative disc height with a mean difference of 0.5 ± 0.1 mm (p < 0.001). Although the mean disc volume difference did not significantly increase, the mean foraminal volume difference was 0.4 ± 0.16 mm3 (p < 0.05). Conclusions: In select patients with degenerative disc diseases or lumbar spinal stenosis, the intervertebral foramen was enlarged, and disc loading was reduced after IPD implantation with decompression surgery. The 3D findings were compatible with the clinical benefits.

Multi-parametric and multi-regional histogram analysis of MRI: modality integration reveals imaging phenotypes of glioblastoma.

OBJECTIVES: Integrating multiple imaging modalities is crucial for MRI data interpretation. The purpose of this study is to determine whether a previously proposed multi-view approach can effectively integrate the histogram features from multi-parametric MRI and whether the selected features can offer incremental prognostic values over clinical variables. METHODS: Eighty newly-diagnosed glioblastoma patients underwent surgery and chemoradiotherapy. Histogram features of diffusion and perfusion imaging were extracted from contrast-enhancing (CE) and non-enhancing (NE) regions independently. An unsupervised patient clustering was performed by the multi-view approach. Kaplan-Meier and Cox proportional hazards regression analyses were performed to evaluate the relevance of patient clustering to survival. The metabolic signatures of patient clusters were compared using multi-voxel spectroscopy analysis. The prognostic values of histogram features were evaluated by survival and ROC curve analyses. RESULTS: Two patient clusters were generated, consisting of 53 and 27 patients respectively. Cluster 2 demonstrated better overall survival (OS) (p = 0.007) and progression-free survival (PFS) (p < 0.001) than Cluster 1. Cluster 2 displayed lower N-acetylaspartate/creatine ratio in NE region (p = 0.040). A higher mean value of anisotropic diffusion in NE region was associated with worse OS (hazard ratio [HR] = 1.40, p = 0.020) and PFS (HR = 1.36, p = 0.031). The seven features selected by this approach showed significantly incremental value in predicting 12-month OS (p = 0.020) and PFS (p = 0.022). CONCLUSIONS: The multi-view clustering method can provide an effective integration of multi-parametric MRI. The histogram features selected may be used as potential prognostic markers. KEY POINTS: • Multi-parametric magnetic resonance imaging captures multi-faceted tumor physiology. • Contrast-enhancing and non-enhancing tumor regions represent different tumor components with distinct clinical relevance. • Multi-view data analysis offers a method which can effectively select and integrate multi-parametric and multi-regional imaging features.

Neuroimaging classification of progression patterns in glioblastoma: a systematic review.

BACKGROUND: Our primary objective was to report the current neuroimaging classification systems of spatial patterns of progression in glioblastoma. In addition, we aimed to report the terminology used to describe 'progression' and to assess the compliance with the Response Assessment in Neuro-Oncology (RANO) Criteria. METHODS: We conducted a systematic review to identify all neuroimaging studies of glioblastoma that have employed a categorical classification system of spatial progression patterns. Our review was registered with Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) registry. RESULTS: From the included 157 results, we identified 129 studies that used labels of spatial progression patterns that were not based on radiation volumes (Group 1) and 50 studies that used labels that were based on radiation volumes (Group 2). In Group 1, we found 113 individual labels and the most frequent were: local/localised (58%), distant/distal (51%), diffuse (20%), multifocal (15%) and subependymal/subventricular zone (15%). We identified 13 different labels used to refer to 'progression', of which the most frequent were 'recurrence' (99%) and 'progression' (92%). We identified that 37% (n = 33/90) of the studies published following the release of the RANO classification were adherent compliant with the RANO criteria. CONCLUSIONS: Our review reports significant heterogeneity in the published systems used to classify glioblastoma spatial progression patterns. Standardization of terminology and classification systems used in studying progression would increase the efficiency of our research in our attempts to more successfully treat glioblastoma.

Less Invasive Phenotype Found in Isocitrate Dehydrogenase-mutated Glioblastomas than in Isocitrate Dehydrogenase Wild-Type Glioblastomas: A Diffusion-Tensor Imaging Study.

Purpose To explore the diffusion-tensor (DT) imaging-defined invasive phenotypes of both isocitrate dehydrogenase (IDH-1)-mutated and IDH-1 wild-type glioblastomas. Materials and Methods Seventy patients with glioblastoma were prospectively recruited and imaged preoperatively. All patients provided signed consent, and the local research ethics committee approved the study. Patients underwent surgical resection, and tumor samples underwent immunohistochemistry for IDH-1 R132H mutations. DT imaging data were coregistered to the anatomic magnetic resonance study and reconstructed to provide the anisotropic and isotropic components of the DT. The invasive phenotype was determined by using previously published criteria and correlated with IDH-1 mutation status by using the Freeman-Halton extension of the Fisher exact probability test. Results Nine patients had an IDH-1 mutation and 61 had IDH-1 wild type. All of the patients with IDH-1 mutation had a minimally invasive DT imaging phenotype. Among the IDH-1 wild-type tumors, 42 of 61 (69%) were diffusively invasive glioblastomas, 14 of 61 (23%) were locally invasive, and five of 61 (8%) were minimally invasive (P < .001). Conclusion IDH-mutated glioblastomas have a less invasive phenotype compared with IDH wild type. This finding may have implications for individualizing the extent of surgical resection and radiation therapy volumes.

Multiparametric MR Imaging of Diffusion and Perfusion in Contrast-enhancing and Nonenhancing Components in Patients with Glioblastoma.

Purpose To determine whether regions of low apparent diffusion coefficient (ADC) with high relative cerebral blood volume (rCBV) represented elevated choline (Cho)-to-N-acetylaspartate (NAA) ratio (hereafter, Cho/NAA ratio) and whether their volumes correlated with progression-free survival (PFS) and overall survival (OS) in patients with glioblastoma (GBM). Materials and Methods This retrospective analysis was approved by the local research ethics committee. Volumetric analysis of imaging data from 43 patients with histologically confirmed GBM was performed. Patients underwent preoperative 3-T magnetic resonance imaging with conventional, diffusion-weighted, perfusion-weighted, and spectroscopic sequences. Patients underwent subsequent surgery with adjuvant chemotherapy and radiation therapy. Overlapping low-ADC and high-rCBV regions of interest (ROIs) (hereafter, ADC-rCBV ROIs) were generated in contrast-enhancing and nonenhancing regions. Cho/NAA ratio in ADC-rCBV ROIs was compared with that in control regions by using analysis of variance. All resulting ROI volumes were correlated with patient survival by using multivariate Cox regression. Results ADC-rCBV ROIs within contrast-enhancing and nonenhancing regions showed elevated Cho/NAA ratios, which were significantly higher than those in other abnormal tumor regions (P < .001 and P = .008 for contrast-enhancing and nonenhancing regions, respectively) and in normal-appearing white matter (P < .001 for both contrast-enhancing and nonenhancing regions). After Cox regression analysis controlling for age, tumor size, resection extent, O-6-methylguanine-DNA methyltransferase-methylation, and isocitrate dehydrogenase mutation status, the proportional volume of ADC-rCBV ROIs in nonenhancing regions significantly contributed to multivariate models of OS (hazard ratio, 1.132; P = .026) and PFS (hazard ratio, 1.454; P = .017). Conclusion Volumetric analysis of ADC-rCBV ROIs in nonenhancing regions of GBM can be used to identify patients with poor survival trends after accounting for known confounders of GBM patient outcome.

Validation of a semi-automatic co-registration of MRI scans in patients with brain tumors during treatment follow-up.

There is an expanding research interest in high-grade gliomas because of their significant population burden and poor survival despite the extensive standard multimodal treatment. One of the obstacles is the lack of individualized monitoring of tumor characteristics and treatment response before, during and after treatment. We have developed a two-stage semi-automatic method to co-register MRI scans at different time points before and after surgical and adjuvant treatment of high-grade gliomas. This two-stage co-registration includes a linear co-registration of the semi-automatically derived mask of the preoperative contrast-enhancing area or postoperative resection cavity, brain contour and ventricles between different time points. The resulting transformation matrix was then applied in a non-linear manner to co-register conventional contrast-enhanced T1 -weighted images. Targeted registration errors were calculated and compared with linear and non-linear co-registered images. Targeted registration errors were smaller for the semi-automatic non-linear co-registration compared with both the non-linear and linear co-registered images. This was further visualized using a three-dimensional structural similarity method. The semi-automatic non-linear co-registration allowed for optimal correction of the variable brain shift at different time points as evaluated by the minimal targeted registration error. This proposed method allows for the accurate evaluation of the treatment response, essential for the growing research area of brain tumor imaging and treatment response evaluation in large sets of patients. Copyright © 2016 John Wiley & Sons, Ltd.

AB009. Distinct clinical profiles and treatment outcomes in invasive pituitary adenoma and Crooke cell tumor: a comparative case report.

BACKGROUND: Pituitary adenomas, particularly invasive pituitary adenomas and Crooke cell tumors, represent significant neuroendocrine tumors with diverse clinical and pathological profiles. Invasive pituitary adenomas are known for their aggressive growth and potential to invade surrounding structures, leading to notable endocrine and neurological complications. Crooke cell tumors, a rare variant of corticotroph adenomas, are characterized by Crooke cells-basophilic cells with cytoplasmic changes due to prolonged glucocorticoid exposure. Although these tumors have aggressive histological features, their clinical behavior can vary widely, often determined by their proliferative index, such as Ki-67. This case report presents two patients with pituitary adenomas, highlighting the contrasting clinical courses, management strategies, and pathological findings. CASE DESCRIPTION: Case 1 features a 42-year-old Asian male, initially diagnosed with a 3-cm pituitary macroadenoma at 23 years, experienced symptoms of blurred vision. Elevated adrenocorticotropic hormone (ACTH) levels were noted. Despite transnasal transsphenoidal surgery and subsequent interventions, including radiotherapy, the tumor recurred multiple times, leading to significant visual and neurological complications. Pathological assessments consistently showed an ACTH-positive adenoma with increasing markers of aggressiveness, including a Ki-67 index up to 5%. The patient was eventually diagnosed with Cushing's disease. Case 2 discusses a 56-year-old male who was incidentally found to have a pituitary tumor during a routine health check, with no visual symptoms despite optic nerve compression. Post-surgery, pathology revealed a Crooke cell tumor with a Ki-67 index of 1%. The patient had an uneventful recovery and no recurrence, with normal hormonal levels throughout follow-up. CONCLUSIONS: These cases underscore the heterogeneity of pituitary adenomas, emphasizing the importance of the Ki-67 index as a prognostic marker. While invasive adenomas often require aggressive multimodal therapy, Crooke cell tumors with low proliferative indices can be effectively managed with surgery alone. A multidisciplinary approach is crucial for optimizing patient outcomes in the management of these tumors.

AB040. A complication of recurrent artery of Huebner infarction after resection of an anterior cerebral artery thrombotic giant intracranial aneurysm: case report and literature review.

BACKGROUND: Giant aneurysms, comprising 3-5% of all intracranial aneurysms, pose a considerable challenge due to their heterogeneity and complex vascular anatomy. Defined as aneurysms exceeding 2.5 cm in diameter, they often develop intraluminal thrombosis. Despite advancements in neurosurgical techniques, managing giant aneurysms remains complex and highly individualized. Thrombotic giant aneurysms are particularly problematic due to their size and thrombosis potential. This case report is unique as it presents the first documented instance of recurrent artery of Heubner (RAH) infarction following surgical resection of a giant thrombotic aneurysm. CASE DESCRIPTION: A 53-year-old man with no prior systemic presented to our emergency department due to progressive left-sided weakness and slurred speech. Magnetic resonance imaging (MRI) of brain revealed a thrombotic giant intracranial aneurysm on right anterior cerebral artery (ACA). Surgical resection was performed using a right pterional craniotomy. During surgery, the aneurysm was confirmed to be completely thrombosed and was excised. Postoperatively, the patient experienced a generalized seizure and was intubated. Brain MRI revealed a new infarction in the RAH territory. Despite initial complications, the patient showed significant recovery with rehabilitation, regaining most motor functions by the 6-month follow-up. CONCLUSIONS: This case emphasizes the critical importance of comprehensive preoperative evaluation, particularly in assessing small perforating branches and collateral circulation. It highlights the challenges in managing giant aneurysms and the necessity of anticipating potential postoperative complications. This report adds valuable insights into the clinical management and surgical planning for giant aneurysms, particularly those involving the ACA and RAH.

The impact of surgeon's academic leave on surgical outcomes for endoscopic transsphenoidal resection of pituitary tumors.

Background: Endoscopic surgery has become the mainstay of pituitary surgery, but requires comprehensive surgical training. We evaluate the impact of a surgeon's academic leave during endoscopic training on surgical outcomes of patients with pituitary tumors. Methods: This retrospective study reviewed the surgical outcomes of endoscopic transsphenoidal surgery for pituitary tumors performed by a single surgeon. The last 56 surgical cases were performed between July 2010 and August 2014 before academic leave (Phase 1 surgery group), while another 56 consecutive cases were performed between November 2017 and March 2020 immediately after the surgeon's academic leave (Phase 2 surgery group). Demographic and clinical characteristics were collected and compared between the two surgery groups. Results: Overall, most surgical outcomes of endoscopic transsphenoidal surgery were not affected adversely by the period of academic leave. The operative time and length of hospital stay was lower in the Phase 2 surgery group compared to the Phase 1 surgery group (P<0.05). Postoperative tumor residual, intraoperative cerebrospinal fluid (CSF) leaks and reoperation also decreased significantly in the Phase 2 group compared to the Phase 1 group (P<0.05). Similar results were observed in patients operated using a one-hand/mono-nostril and two-hand/one-and-half nostril technique. Conclusions: Academic leave had no negative impact on most surgical outcomes for endoscopic transsphenoidal resection of pituitary tumors. Moreover, a trend toward shorter operative times and length of hospital stays was noted for patients receiving surgery immediately after surgeon's return from leave.

AB092. Imaging the histopathology subtypes of the growth hormone-secreting pituitary neuroendocrine tumor.

BACKGROUND: Sparsely granulated (SG) growth hormone-secreting pituitary neuroendocrine tumors (GH-PitNETs) often present with a more aggressive clinical course compared to densely granulated (DG) tumors. These subtypes exhibit distinct biological and imaging characteristics. Thus, this study aims to differentiate between the histopathological subtypes of GH-PitNETs using pre-operative magnetic resonance imaging (MRI). METHODS: A retrospective analysis was conducted on 83 acromegalic patients treated at our institution between 2000 and 2010. Tumor volumes were segmented from preoperative MRIs, including T1-weighted, T2-weighted, T1 with contrast, and T2 fluid attenuated inversion recovery (FLAIR) sequences. Reference regions of interest (ROIs) were delineated using gray and white matter from the same sequences. Two pathologists reviewed pathology specimens for anti-cytokeratin (CAM 5.2) and Pit-1 expression. Clinical and radiological biomarkers were compared between SG and DG patients. RESULTS: A total of 83 patients with complete histopathology and 51 patients with complete MRIs were included in the analysis. SG PitNETs exhibited higher rates of supra-sellar invasion (61.5%, P<0.001), larger tumor sizes, lower pre-operative GH levels, and increased post-operative residual tumor (65.4%, P<0.001) compared to DG PitNETs. Additionally, SG PitNETs showed greater hyperintensity on T2-weighted images and enhanced contrast, whereas DG PitNETs exhibited less contrast enhancement. Utilization of these imaging biomarkers demonstrated an 94.1% accuracy in T2 FLAIR and overall of 78.7% predicting the histopathological subtypes of GH-PitNETs. CONCLUSIONS: Distinct histopathological subtypes of GH-PitNETs represent crucial prognostic factors. Utilizing multimodal pre-operative MRIs, clinicians can accurately identify sparsely granulated GH-PitNETs, facilitating improved treatment planning strategies.

Psoriatic arthritis increases the risk of venous thromboembolism following degenerative lumbar spine surgery: An analysis of U.S. Nationwide Inpatient Sample 2005-2018.

Background: This retrospective study aimed to determine the risk of venous thromboembolism (VTE) in patients with PsA after surgery for lumbar degenerative disease (LDD). Methods: The study data of adults aged ≥20 years admitted to U.S. hospitals with diagnoses of LDD and undergoing spinal decompression or fusion between 2005 and 2018 were extracted from the National Inpatient Sample (NIS) database. Patients were further divided into two groups based on a diagnosis of PsA or not via codes ICD-9: 696.0 and ICD-10: L40.50. Patients with missing information were excluded. Propensity score matching (PSM) was employed to enhance comparability between groups. Logistic regression was used to determine associations between PsA and various outcomes, including complications, unfavorable discharge, and prolonged length of stay (LOS). Results: Data on 471,283 patients with LDD was extracted from the NIS database.from 2005 to 2018. Before propensity score matching, patients with PsA had higher proportions of overall morbidity (8.8 % vs. 6.9 %), VTE (1.4 % vs. 0.7 %), and unfavorable discharge (20.8 % vs. 16.9 %). After matching, patients with PsA still had higher VTE incidence and unfavorable discharge proportions. After adjustments, multivariable regression analysis indicated that patients with PsA had a higher risk of unfavorable discharge (aOR: 1.26, 95 % CI: 1.03-1.55) and VTE (aOR: 1.99, 95 % CI: 1.05-3.75). Conclusions: Among patients undergoing surgery for LDD, pre-existing PsA may be associated with increased risks of unfavorable discharge and VTE occurrence. The findings may benefit preoperative risk stratifications before LDD surgeries.

Impact of metabolic syndrome on postoperative outcomes of transsphenoidal pituitary surgery: analysis of U.S. nationwide inpatient sample data 2005-2018.

Introduction: Transsphenoidal surgery (TSS) is the preferred surgical method for most pituitary adenomas owing to high efficacy and low mortality. This study aimed to evaluate the influence of metabolic syndrome (MetS) on postoperative outcomes of TSS for pituitary adenoma. Methods: This population-based, retrospective observational study extracted data of adults 20-79 y receiving TSS for pituitary adenoma from the US Nationwide Inpatient Sample (NIS) between 2005-2018. Primary outcomes were pituitary-related complications, poor outcomes (i.e., in-hospital mortality or unfavorable discharge), prolonged length of stay (LOS), and patient safety indicators (PSIs). Univariate and multivariate regressions were performed to determine the associations between study variables and outcomes. Results: 19,076 patients (representing a 93,185 US in-patient population) were included, among which 2,109 (11.1%) patients had MetS. After adjustment, pre-existing MetS was not significantly associated with presence of pituitary-related complications and poor outcomes. In contrast, MetS was significantly associated with an increased risk for prolonged LOS (adjusted OR (aOR) = 1.19; 95% CI: 1.05-1.34), PSIs (aOR = 1.31; 95% CI: 1.07-1.59) and greater hospital costs (adjusted ß = 8.63 thousand USD; 95% CI: 4.98-12.29). Among pituitary-related complications, MetS was independently associated with increased risk of cerebrospinal fluid (CSF) rhinorrhea (aOR = 1.22, 95% CI: 1.01, 1.47) but lowered diabetes insipidus (aOR = 0.83, 95% CI: 0.71, 0.97). Discussion: MetS does not pose excessive risk of in-hospital mortality or unfavorable discharge. However, MetS independently predicted having PSIs, prolonged LOS, greater hospital costs, and CSF rhinorrhea. Study findings may help clinicians achieve better risk stratification before TSS.

Long-term follow-up of patients with surgical intractable acromegaly after linear accelerator radiosurgery.

BACKGROUND/PURPOSE: Radiotherapy is a crucial treatment for acromegalic patients with growth hormone (GH)-secreting pituitary tumors. However, its effect takes time. We retrospectively reviewed the long-term outcome of linear accelerator stereotactic radiosurgery (LINAC SRS) for patients with acromegaly from the perspective of biochemical remission and associated factors. METHODS: Twenty-two patients presenting with residual or recurrent (GH)-secreting functional pituitary tumor between 1994 and 2004 who received LINAC SRS were enrolled and followed up for at least 3 years. Residual or recurrent tumor was defined as persistent elevated GH or insulin-like growth factor-1 (IGF-1) level and image-confirmed tumor after previous surgical treatment. Biochemical remission was defined as fasting GH less than 2.5 ng/mL with normal sex-and-age adjusted IGF-1. RESULTS: The mean follow-up period was 94.7 months (range 36-161 months). Overall mean biochemical remission time was 53 months (median 30 months). Biochemical control was achieved in 15 patients (68.2%) over the follow up period. One patient experienced recurrence after SRS and underwent another operation. Initial GH at diagnosis and pre-SRS GH correlated with biochemical control (p = 0.005 and p < 0.0001, respectively). Further evaluation demonstrated that biochemical control stabilized after 7.5 years. Overall post-SRS hormone deficit persisted in five patients (22.7%). CONCLUSION: In comparison to other radiosurgery modalities, LINAC radiosurgery also provides a satisfactory outcome. SRS has maximum effect over the first 2 years and stabilizes after 7.5 years. Moreover, SRS elicits long-term biochemical effects and requires longer follow-up for better biochemical remission.

Outcomes of hemorrhagic stroke in patients with autoimmune rheumatic diseases: An analysis of the US Nationwide Inpatient Sample.

AIM: To determine whether and how rheumatoid arthritis (RA), systemic lupus erythematosus (SLE), and systemic sclerosis (SSc) affect outcomes in patients admitted for hemorrhagic stroke. METHODS: This study screened the Nationwide Inpatient Sample database for adults aged ≥20 years admitted to US hospitals with a principal diagnosis of intracerebral hemorrhage (ICH) between 2005 and 2018. Diagnoses were determined using the International Classification of Diseases, 9th and 10th revisions (ICD-9 and ICD-10) diagnostic codes for ICH (ICD-9: 431, 432; ICD-10: I61, I62). Study outcomes were: (1) in-hospital mortality; (2) unfavorable discharge, defined as transfer to nursing homes or long-term care facilities; and (3) prolonged length of stay (LOS), defined as LOS >75th centile. RESULTS: Associations between comorbid RA, SLE, and SSc and clinical outcomes show a significantly lower risk of in-hospital mortality and prolonged LOS in RA patients. After admissions for ICH, the risk for in-hospital mortality and prolonged LOS was decreased in RA patients, and the risk for unfavorable discharge (long-term care) was reduced in SLE patients. CONCLUSIONS: Among patients admitted to US hospitals for hemorrhagic stroke, patients with RA had decreased risk for in-hospital mortality and prolonged LOS.

Augmented Reality Surgical Navigation System Integrated with Deep Learning.

Most current surgical navigation methods rely on optical navigators with images displayed on an external screen. However, minimizing distractions during surgery is critical and the spatial information displayed in this arrangement is non-intuitive. Previous studies have proposed combining optical navigation systems with augmented reality (AR) to provide surgeons with intuitive imaging during surgery, through the use of planar and three-dimensional imagery. However, these studies have mainly focused on visual aids and have paid relatively little attention to real surgical guidance aids. Moreover, the use of augmented reality reduces system stability and accuracy, and optical navigation systems are costly. Therefore, this paper proposed an augmented reality surgical navigation system based on image positioning that achieves the desired system advantages with low cost, high stability, and high accuracy. This system also provides intuitive guidance for the surgical target point, entry point, and trajectory. Once the surgeon uses the navigation stick to indicate the position of the surgical entry point, the connection between the surgical target and the surgical entry point is immediately displayed on the AR device (tablet or HoloLens glasses), and a dynamic auxiliary line is shown to assist with incision angle and depth. Clinical trials were conducted for EVD (extra-ventricular drainage) surgery, and surgeons confirmed the system's overall benefit. A "virtual object automatic scanning" method is proposed to achieve a high accuracy of 1 ± 0.1 mm for the AR-based system. Furthermore, a deep learning-based U-Net segmentation network is incorporated to enable automatic identification of the hydrocephalus location by the system. The system achieves improved recognition accuracy, sensitivity, and specificity of 99.93%, 93.85%, and 95.73%, respectively, representing a significant improvement from previous studies.

Prognostic value of interim CT-based peritumoral and intratumoral radiomics in laryngeal and hypopharyngeal cancer patients undergoing definitive radiotherapy.

BACKGROUND AND PURPOSE: We aimed to investigate the prognostic value of peritumoral and intratumoral computed tomography (CT)-based radiomics during the course of radiotherapy (RT) in patients with laryngeal and hypopharyngeal cancer (LHC). MATERIALS AND METHODS: A total of 92 eligible patients were 1:1 randomly assigned into training and validation cohorts. Pre-RT and mid-RT radiomic features were extracted from pre-treatment and interim CT. LASSO-Cox regression was used for feature selection and model construction. Time-dependent area under the receiver operating curve (AUC) analysis was applied to evaluate the models' prognostic performances. Risk stratification ability on overall survival (OS) and progression-free survival (PFS) were assessed using the Kaplan-Meier method and Cox regression. The associations between radiomics and clinical parameters as well as circulating lymphocyte counts were also evaluated. RESULTS: The mid-RT peritumoral (AUC: 0.77) and intratumoral (AUC: 0.79) radiomic models yielded better performance for predicting OS than the pre-RT intratumoral model (AUC: 0.62) in validation cohort. This was confirmed by Kaplan-Meier analysis, in which risk stratification depended on the mid-RT peritumoral (p = 0.009) and intratumoral (p = 0.003) radiomics could be improved for OS, in comparison to the pre-RT intratumoral radiomics (p = 0.199). Multivariate analysis identified mid-RT peritumoral and intratumoral radiomic models as independent prognostic factors for both OS and PFS. Mid-RT peritumoral and intratumoral radiomics were correlated with treatment-related lymphopenia. CONCLUSION: Mid-RT peritumoral and intratumoral radiomic models are promising image biomarkers that could have clinical utility for predicting OS and PFS in patients with LHC treated with RT.

Frequency-Domain Analysis for Accurate and Robust Gait Cycle Time Detection with Clinical Data.

Gait tasks have become a topic of increasing inter-est in biological engineering research in recent years. One way to obtain the gait cycle time (GCT) is to analyze a subject's gait acceleration signal as recorded by an inertial measurement unit (IMU) [1]. An accurate peak detection of the IMU acceleration has thus become a requirement for GCT analysis. This study proposes a detection procedure for accurately detecting the peaks in a noisy IMU acceleration signal based on a frequency-domain analysis of the acceleration.

Surgical Outcome and Evaluation of Strategies in the Management of Growth Hormone-Secreting Pituitary Adenomas After Initial Transsphenoidal Pituitary Adenectomy Failure.

Acromegaly is a systemic disease that requires multidisciplinary treatment to achieve the best clinical outcome. This study aimed to evaluate the outcomes of the endoscopic transsphenoidal approach (TSA) as the primary treatment for somatotroph adenomas and further investigate patients who had suboptimal surgical results. This retrospective study included 83 patients with somatotroph adenomas treated by TSA at our institution from 1999 to 2010. Biochemical remission was defined as hGH <1 and <2.5 ng/ml. Factors associated with failure of TSA and strategy of secondary treatments for refractory and recurrent disease were analyzed. The mean age of patients was 41.1 ± 11.3 years, and the mean follow-up time was 54.2 ± 44.3 months. Approximately 44.5% of patients had residual tumors after TSA. Larger tumor size, higher GH level before the operation, and the existence of residual tumors were associated with TSA failure. Forty-one patients had an inadequate response to TSA or a recurrent lesion, and of these patients, 37 had residual tumor after TSA. Octreotide results in good outcomes in the treatment of DGSA patients, and SRS/EXRT generates good results in treating patients who receive second treatments when remission cannot be reached 6 months after TSA operation.