SD-GAN: A Style Distribution Transfer Generative Adversarial Network for Covid-19 Detection through X-ray Images

The Covid-19 pandemic is a prevalent health concern around the world in recent times. Therefore, it is essential to screen the infected patients at the primary stage to prevent secondary infections from person to person. The reverse transcription polymerase chain reaction (RT-PCR) test is commonly performed for Covid-19 diagnosis, while it requires significant effort from health professionals. Automated Covid-19 diagnosis using chest X-ray images is one of the promising directions to screen infected patients quickly and effectively. Automatic diagnostic approaches are used with the assumption that data originating from different sources have the same feature distributions. However, the X-ray images generated in different laboratories using different devices experience style variations e.g., intensity and contrast which contradict the above assumption. The prediction performance of deep models trained on such heterogeneous images of different distributions with different noises is affected. To address this issue, we have designed an automatic end-to-end adaptive normalization-based model called style distribution transfer generative adversarial network (SD-GAN). The designed model is equipped with the generative adversarial network (GAN) and task-specific classifier to transform the style distribution of images between different datasets belonging to different race people and carried out Covid-19 detection effectively. Evaluated results on four different X-ray datasets show the superiority of the proposed model to state-of-the-art methods in terms of the visual quality of style transferred images and the accuracy of Covid-19 infected patient detection. SD-GAN is publicly available at: https://github.com/tasleem-hello/SD-GAN/tree/SD-GAN. Author

A review on biochemical constituents of pumpkin and their role as pharma foods, a key strategy to improve health in post COVID 19 period

Progression of today's world has been given setback due to the adversity of a novel, viral, deadly outbreak COVID 19, which raised the concerns of the scientists, researchers and health related officials about the inherent and adaptive immune system of the living body and its relation with healthy diet balanced with pharma foods. Now world is coming out of the destructive pandemic era, the choice of right food can help to build and boost adaptive immunity and pumpkin due to excellent profile of functional and nutraceutical constituents could be the part of both infected and non-infected person's daily diet. Vitamins like A, C and E, minerals like zinc, iron and selenium, essential oils, peptides, carotenoids and polysaccharides present in pumpkin could accommodate the prevailing deficiencies in the body to fought against the viral pathogens. In current post COVID 19 scenario adequate supply of healthy diet, balanced with pharma foods could play a basic role in boosting immune system of the populations. This review covers the pharmacological activities of pumpkin functional constituents in relation with COVID 19 pandemic. Pumpkins are well equipped with nutraceuticals and functional bioactives like tocopherols, polyphenols, terpenoids and lutein therefore, consumption and processing of this remarkable vegetable could be encouraged as pharma food due to its antihyperlipidemic, antiviral, anti-inflammatory, antihyperglycemic, immunomodulatory, antihypertensive, antimicrobial and antioxidant potential. Need of healthy eating in current post COVID 19 period is very crucial for healthy population, and medicinal foods like pumpkin could play a vital role in developing a healthy community around the globe. Graphical : [Figure not available: see fulltext.]. © 2023, The Author(s).

Optical Chaos Based Secure Optical Body Area Network Employing Polarization Multiplexing and Free Space Optics for e-Health Applications

We propose and demonstrate an optical chaos secured optical body area network (OBAN) employing polarization multiplexing and free space optics links. The physiological data of patient coded in non-return to zero on-off keying (NRZ-OOK) format from two on-body nodes modulated on orthogonal polarization states of a continuous wave (CW) laser is secured by using additive chaos masking (ACM) technique with chaotic waveforms generated through direct modulation of semiconductor chaotic lasers (CLs). After polarization multiplexing, the secure NRZ- OOK modulated optical signals are transmitted over indoor and outdoor free space optics (FSO) links based on GammaGamma channel model towards remote healthcare center. After chaos subtraction, the NRZ-OOK modulated optical signals are photodetected and passed on to bit error rate (BER) estimator for performance analysis. The electronic health (e-health) system based on the proposed OBAN provides adequate privacy for classified patient related information with added advantages of acceptable BER results, cost efficiency, speedy installation and suitable for use in current pandemic situation. © 2022 IEEE.