Background: Lung cancer is one of the most fatal cancers worldwide, and malignant tumors are characterized by the growth of abnormal cells in the tissues of lungs. Usually, symptoms of lung cancer do not appear until it is already at an advanced stage. The proper segmentation of cancerous lesions in CT images is the primary method of detection towards achieving a completely automated diagnostic system. Method: In this work, we developed an improved hybrid neural network via the fusion of two architectures, MobileNetV2 and UNET, for the semantic segmentation of malignant lung tumors from CT images. The transfer learning technique was employed and the pre-trained MobileNetV2 was utilized as an encoder of a conventional UNET model for feature extraction. The proposed network is an efficient segmentation approach that performs lightweight filtering to reduce computation and pointwise convolution for building more features. Skip connections were established with the Relu activation function for improving model convergence to connect the encoder layers of MobileNetv2 to decoder layers in UNET that allow the concatenation of feature maps with different resolutions from the encoder to decoder. Furthermore, the model was trained and fine-tuned on the training dataset acquired from the Medical Segmentation Decathlon (MSD) 2018 Challenge. Results: The proposed network was tested and evaluated on 25% of the dataset obtained from the MSD, and it achieved a dice score of 0.8793, recall of 0.8602 and precision of 0.93. It is pertinent to mention that our technique outperforms the current available networks, which have several phases of training and testing.

The purpose of this article is to propose a framework of Social Impact Project Finance (SIPF) for financing infrastructure projects. In the framework, public sector commits to pay performance-based yield for improved services, which in turn motivate private investor to better construct and manage infrastructure assets. By embedding social impact performance criteria, SIPF has an endogenous incentive mechanism that can align perspectives of public and private sectors, while traditional project finance lacks. SIPF can also play a critical role in achieving sustainable infrastructure, because all three components of sustainability (social, environmental and economic) can be easily embedded as impact factors in the framework. In addition, it could be well customized for governments globally to address their own problems, such as carbon emission or congestion reduction. Private investor can also benefit from higher yield, liquidity, and tax break. With this innovative interdisciplinary initiation of impact finance and project finance, there will be significant opportunities to not only create a new format of impact-based project delivery methods, but also show us a prospective way to securitize and deliver public services under clear supervision in the future

Myocardial repolarization instability measured by electrocardiography (ECG) as the QT variability index (QTVI) predicts arrhythmic risk. This proof-of-concept study aims to validate a non-skin-contact wearable radar-based technique for monitoring myocardial repolarization instability. QTVI by ECG (normalized QT variability to normalized heart rate variability) was compared to the analogous J-point-to-T-end variability index (JTVI) by ECG. Simultaneous 2-minute ECG and wearable radar recordings acquired under controlled, seated conditions were analysed using an intelligent algorithm to extract mechanical correlates of the J-point and T-end from radar-derived cardiac motion signals, and compared to JTVI by ECG. Based on a selection of publicly available datasets of 9 patients with atrial fibrillation, 11 patients with sinus rhythm, and 20 healthy people (n=40, 23% with atrial fibrillation, mean ± SD heart rate 85 ± 12 beats/min, QTVI -0.36±0.75, JTVI -0.10±0.75), the SD of JT and QT intervals strongly agreed (R²>0.99, bias -0.4±1.5 ms, 1.1±3.9%), and JTVI and QTVI correlated closely (R²>0.98, QTVI =1.0× JTVI -0.26, bias -0.26±0.11 ms). Among a separate group of healthy volunteers (n=20, age 43±11 years, 25% female, heart rate 66±14 beats/min, JTVI -0.52±0.51), there was excellent agreement for ECG-derived and radar-estimated SD of JT (R²=0.88, bias 0.1±4.9 ms, 6.4±2.7%), SD of beat-to-beat interval (SDNN) (R²=0.93, bias 16.0±22.1 ms, 9.5±4.9%), and JTVI (R²>0.99, bias 0.00±0.05, 0.0±6.5%). QTVI and JTVI demonstrate strong agreement with a small systematic bias in quantifying myocardial repolarization instability, and JTVI can be accurately and precisely estimated using radar sensors. This study provides an initial proof-of-concept demonstration of the feasibility of non-skin-contact wearable radar-based monitoring of arrhythmic risk, with particular applicability to culturally sensitive populations.