Deep learning for automatic segmentation of vestibular schwannoma: a retrospective study from multi-center routine MRI.

Automatic segmentation of vestibular schwannoma (VS) from routine clinical MRI has potential to improve clinical workflow, facilitate treatment decisions, and assist patient management. Previous work demonstrated reliable automatic segmentation performance on datasets of standardized MRI images acquired for stereotactic surgery planning. However, diagnostic clinical datasets are generally more diverse and pose a larger challenge to automatic segmentation algorithms, especially when post-operative images are included. In this work, we show for the first time that automatic segmentation of VS on routine MRI datasets is also possible with high accuracy. We acquired and publicly release a curated multi-center routine clinical (MC-RC) dataset of 160 patients with a single sporadic VS. For each patient up to three longitudinal MRI exams with contrast-enhanced T1-weighted (ceT1w) (n = 124) and T2-weighted (T2w) (n = 363) images were included and the VS manually annotated. Segmentations were produced and verified in an iterative process: (1) initial segmentations by a specialized company; (2) review by one of three trained radiologists; and (3) validation by an expert team. Inter- and intra-observer reliability experiments were performed on a subset of the dataset. A state-of-the-art deep learning framework was used to train segmentation models for VS. Model performance was evaluated on a MC-RC hold-out testing set, another public VS datasets, and a partially public dataset. The generalizability and robustness of the VS deep learning segmentation models increased significantly when trained on the MC-RC dataset. Dice similarity coefficients (DSC) achieved by our model are comparable to those achieved by trained radiologists in the inter-observer experiment. On the MC-RC testing set, median DSCs were 86.2(9.5) for ceT1w, 89.4(7.0) for T2w, and 86.4(8.6) for combined ceT1w+T2w input images. On another public dataset acquired for Gamma Knife stereotactic radiosurgery our model achieved median DSCs of 95.3(2.9), 92.8(3.8), and 95.5(3.3), respectively. In contrast, models trained on the Gamma Knife dataset did not generalize well as illustrated by significant underperformance on the MC-RC routine MRI dataset, highlighting the importance of data variability in the development of robust VS segmentation models. The MC-RC dataset and all trained deep learning models were made available online.

Liquid crystal-assisted optical biosensor for early-stage diagnosis of mammary glands using HER-2.

Breast cancer (BC) is one of the most commonly diagnosed cancers and the second leading cause of cancer mortality among women around the world. The purpose of this study is to present a non-labeled liquid crystal (LC) biosensor, based on the inherent feature of nematic LCs, for the evaluation of BC using the human epidermal growth factor receptor-2 (HER-2) biomarker. The mechanism of this sensing is supported by surface modification with dimethyloctadecyl [3-(trimethoxysilyl) propyl] ammonium chloride (DMOAP) encouraging the long alkyl chains that induce a homeotropic orientation of the LC molecules at the interface. To enhance the binding efficacy of more HER-2 antibody (Ab) on LC aligning agents, a simple ultraviolet radiation-assisted method was also used to increase functional groups on the DMOAP coated slides, thereby improving binding affinity and efficiency onto HER-2 Abs. The designed biosensor makes use of the specific binding of HER-2 protein to HER-2 Ab and disruption of the orientation of LCs. This orientation change leads to a transition of the optical appearance from dark to birefringent, enabling the detection of HER-2. This novel biosensor exhibits a linear optical response to HER-2 concentration in the wide dynamic range of 10-6-102 ng/mL, with an ultra-low detection limit of 1 fg/mL. As a proof of concept, the designed LC biosensor was successfully investigated for the quantification of HER-2 protein in patients suffering from BC. Owing to the sensitivity, selectivity, and label-free detection, this biosensor may amplify the application of LC-based biosensors for the detection of most types of cancers.

On Natural History and Management of Colloid Cysts: Time to Rethink?

BACKGROUND: Colloid cysts, although benign, may occasionally cause obstructive hydrocephalus and sudden death. Reliable prognostic factors for symptomatic progression have been sought, with heterogenous results. METHODS: We conducted a retrospective review of all cases of colloid cysts of the third ventricle managed at our center between 2009 and 2019. Clinical and neuroimaging characteristics were analyzed using logistic regression in relation to symptomatic status and hydrocephalus. The cutoff values for outcome prediction were calculated using the receiver operating characteristic curve analysis. RESULTS: There were 82 patients with colloid cysts, of whom 60 were asymptomatic and 22 symptomatic. None of the asymptomatic patients experienced acute neurologic decline or hydrocephalus during follow-up, whereas half (n = 11) of the symptomatic patients presented with hydrocephalus, 8 of whom had acute hydrocephalus. We found 3 putative candidate risk factors for symptomatic colloid cysts: T1-weighted magnetic resonance imaging hyperintense/mixed signal appearance (P = 0.004), location in risk zone I (P = 0.007), and a volume >236.49 mm3 (P = 0.007). Cyst diameter and volume/foramen of Monro diameter ratios had a decreasing trend over time among asymptomatic patients, providing new insights into the natural history of the disease. CONCLUSIONS: Only a few asymptomatic colloid cysts showed progression requiring surgery, with no acute deterioration or fatal events, whereas the rest remained stable over time, thus supporting a more conservative approach for this group of patients. Higher risk for developing symptomatic colloid cyst was defined by a risk score that included T1-weighted magnetic resonance imaging appearance, risk zone, and colloid cyst volume, aiding the detection of patients at risk of clinical deterioration.

Cognitive impairment in people with physical frailty using the phenotype model: A systematic review and meta analysis.

OBJECTIVE: We performed a systematic review and meta-analysis to study the relationship between cognitive functioning and phenotypic frailty status. METHODS: We searched Pubmed, Cochrane Library and Epistemonikos from 2000 until March 2022, and used the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. Samples included both sexes, age ≥55 years, assessed with standardized measures of the different cognitive domains and the frailty phenotype model and analyzing the relationship between the frailty subtypes pre-frail, frail and robust and specific cognitive function. RESULTS: Eleven studies published from 2008 until March 2022 fulfilled the inclusion criteria, and 10 were included in our meta-analyses. Sample sizes varied from 104 to 4649 individuals. Mean Mini-Mental State Examination (MMSE) scores ranged from 17.0 to 27.6, with mean difference (MD) of -2.55 (95% confidence interval [CI] -3.32, -1.78) in frail compared to robust, MD -1.64 (95% CI -2.21, -1.06) in frail compared to prefrail and MD -0.68 (95% CI -0.94, -0.43) in prefrail compared to robust. In subgroup analyses, frail persons had lower scores in the memory domain with standardized mean difference (SMD) -1.01 (95% CI -1.42, -0.59). CONCLUSION: MMSE scores were significantly lower in frail compared to robust and prefrail persons and in prefrail compared to robust persons. Subgroup analysis of memory revealed significantly poorer scores in frail compared to robust. The results indicate a strong relationship between physical frailty and cognitive impairment suggesting incorporation of cognitive function in frailty assessments.

COVID-19 leukoencephalopathy with subacute magnetic resonance imaging findings of vasculitis and demyelination.

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) commonly results in a respiratory illness in symptomatic patients; however, those critically ill can develop a leukoencephalopathy. We describe two patients who had novel subacute MRI findings in the context of coronavirus disease 2019 (COVID-19) leukoencephalopathy, which we hypothesize could implicate a potent small-vessel vasculitis, ischemic demyelination and the presence of prolonged ischemia. Recent evidence of the direct neuroinvasiness of SARS-CoV-2 leading to ischemia and vascular damage supports this hypothesis.

Investigation of quadratic electro-optic effects and electro-absorption process in GaN/AlGaN spherical quantum dot.

Quadratic electro-optic effects (QEOEs) and electro-absorption (EA) process in a GaN/AlGaN spherical quantum dot are theoretically investigated. It is found that the magnitude and resonant position of third-order nonlinear optical susceptibility depend on the nanostructure size and aluminum mole fraction. With increase of the well width and barrier potential, quadratic electro-optic effect and electro-absorption process nonlinear susceptibilities are decreased and blueshifted. The results show that the DC Kerr effect in this case is much larger than that in the bulk case. Finally, it is observed that QEOEs and EA susceptibilities decrease and broaden with the decrease of relaxation time.