Fully Automated Skull Stripping from Brain Magnetic Resonance Images Using Mask RCNN-Based Deep Learning Neural Networks.

This research comprises experiments with a deep learning framework for fully automating the skull stripping from brain magnetic resonance (MR) images. Conventional techniques for segmentation have progressed to the extent of Convolutional Neural Networks (CNN). We proposed and experimented with a contemporary variant of the deep learning framework based on mask region convolutional neural network (Mask-RCNN) for all anatomical orientations of brain MR images. We trained the system from scratch to build a model for classification, detection, and segmentation. It is validated by images taken from three different datasets: BrainWeb; NAMIC, and a local hospital. We opted for purposive sampling to select 2000 images of T1 modality from data volumes followed by a multi-stage random sampling technique to segregate the dataset into three batches for training (75%), validation (15%), and testing (10%) respectively. We utilized a robust backbone architecture, namely ResNet-101 and Functional Pyramid Network (FPN), to achieve optimal performance with higher accuracy. We subjected the same data to two traditional methods, namely Brain Extraction Tools (BET) and Brain Surface Extraction (BSE), to compare their performance results. Our proposed method had higher mean average precision (mAP) = 93% and content validity index (CVI) = 0.95%, which were better than comparable methods. We contributed by training Mask-RCNN from scratch for generating reusable learning weights known as transfer learning. We contributed to methodological novelty by applying a pragmatic research lens, and used a mixed method triangulation technique to validate results on all anatomical modalities of brain MR images. Our proposed method improved the accuracy and precision of skull stripping by fully automating it and reducing its processing time and operational cost and reliance on technicians. This research study has also provided grounds for extending the work to the scale of explainable artificial intelligence (XAI).

Frequency of Hyponatremia and Its Impact on Prognosis in Ischemic Stroke.

Background Hyponatremia, often encountered in hospitalized patients, is associated with adverse outcomes in ischemic stroke patients. In this study, we investigated the frequency of hyponatremia and its impact on prognosis and clinical outcomes in ischemic stroke patients from a tertiary care hospital. Methodology A total of 289 patients admitted to the hospital with ischemic stroke from September 2022 to February 2023 were considered in this cross-sectional study. Serum sodium level was measured on admission, and hyponatremia was defined as sodium less than 135 mmol/L. The primary outcome of the study was assessed by the National Institutes of Health Stroke Scale (NIHSS) score on admission and discharge and inpatient mortality. Data were analyzed using SPSS version 20 (IBM Corp., Armonk, NY, USA), and multivariate logistic regressions were conducted using variables identified as having a relationship with hyponatremia. Results Our study shows that among 289 patients with ischemic stroke, the mean age was 61 ± 8.53 years. Hyponatremia was observed in 101 (35%) patients, and all baseline characteristics and risk factors for stroke were similar between patients with and without hyponatremia. The patients with hyponatremia had higher NIHSS scores on admission (p = 0.041) and at discharge (p = 0.039). In the resultant multivariate analysis, hyponatremia was an independent predictor of mortality rates during the hospital stay. The cumulative incidence rates of in-hospital mortality for hyponatremia and normal sodium level were 16.8% and 10.1%, respectively. Conclusions Hyponatremia is prevalent in ischemic stroke and is independently associated with in-hospital mortality and worse NIHSS scores at admission and discharge.

Evaluation of different formulations of IGRs against Aedes albopictus and Culex quinquefasciatus (Diptera:Culicidae)

Objective: To test the relative efficacy of pyriproxyfen and methoprene on mortality, deformity, inhibition and emergence to adult stages of Culex quinquefasciatus and Aedes albopictus. Methods: Serial dilutions (0.01–0.05 mg/L) of methoprene, pyriproxyfen 0.5 water dispersible granules (WDG) and pyriproxyfen 1.0 WDG were used to assess mortality and inhibition of 3rd instar larvae of Aedes albopictus and Culex quinquefasciatus. Each concentration and control was replicated four times in completely randomized design. Data on larval mortality, growth inhibition, deformities and adult's emergence was recorded weekly. On the basis of best comparative performance, the efficacy of pyr-iproxyfen 1.0 WDG at 0.1 g/m3 was also tested in the field by collecting treated water samples monthly for 1–6 months after field application. Twenty five 3rd instar larvae of Aedes and Culex spp. of the same cohorts were used for bioassays and compared with larvae in control cups containing 1 L of untreated tap water. Results: Results revealed variations in fatality of different insect growth regulators (IGRs) to the 3rd instar larvae of Culex and Aedes mosquitoes. Among the IGRs, pyr-iproxyfen 1.0 WDG was found best that exhibited significantly high emergence inhibition against Culex and Aedes spp. Based on the results, the IGRs were classified in terms of the tested parameters in order of pyriproxyfen 1.0 WDG > pyriproxyfen 0.5 WDG > methoprene. In case of field studies, pyriproxyfen 1.0 WDG, pool data of the entire target treated sites showed minimum adult emergence from water sampled of habitats treated with 0.1 g/m3 of pyriproxyfen 1.0 WDG. Conclusions: It is thus concluded that IGRs can be utilized as environment friendly control measures for Culex and Aedes spp. of mosquitoes on small and large scale. This will reduce the use of conventional insecticides by the public health authorities and help in reducing selection pressure of insecticides.