An integrated feature frame work for automated segmentation of COVID‐19 infection from lung CT images

The novel coronavirus disease (SARS‐CoV‐2 or COVID‐19) is spreading across the world and is affecting public health and the world economy. Artificial Intelligence (AI) can play a key role in enhancing COVID‐19 detection. However, lung infection by COVID‐19 is not quantifiable due to a lack of studies and the difficulty involved in the collection of large datasets. Segmentation is a preferred technique to quantify and contour the COVID‐19 region on the lungs using computed tomography (CT) scan images. To address the dataset problem, we propose a deep neural network (DNN) model trained on a limited dataset where features are selected using a region‐specific approach. Specifically, we apply the Zernike moment (ZM) and gray level co‐occurrence matrix (GLCM) to extract the unique shape and texture features. The feature vectors computed from these techniques enable segmentation that illustrates the severity of the COVID‐19 infection. The proposed algorithm was compared with other existing state‐of‐the‐art deep neural networks using the Radiopedia and COVID‐19 CT Segmentation datasets presented specificity, sensitivity, sensitivity, mean absolute error (MAE), enhance‐alignment measure (EM_φ), and structure measure (S_m) of 0.942, 0.701, 0.082, 0.867, and 0.783, respectively. The metrics demonstrate the performance of the model in quantifying the COVID‐19 infection with limited datasets.     

Feasibility Evaluation of Pressurized Irrigation in Canal Commands

India has one of the largest and most ambitious irrigation programme in the world with net irrigated area exceeding 47 million hectares. However, the overall project efficiency from the headwork to the farmer’s field has been quite low which leads to not only poor utilization of irrigation potential created at huge cost, but also aggravates the degradation of soil and water resources and thereby endangers the sustainability of agricultural production system. As the cost of creating additional irrigation potential in terms of financial, human and environmental aspects has increased tremendously, need of the hour is to increase the irrigation efficiency of existing projects and use saved water for irrigating new areas or meeting the demand of non-agricultural sector. The contribution of application efficiency to poor irrigation efficiency is quite high and therefore increasing application efficiency by a shift in application method from surface to pressurized system has potential of vastly improving irrigation efficiency. To evaluate feasibility of this concept, a pilot study was initiated at Water Technology Centre for Eastern Region, Bhubaneswar, on one outlet of a minor irrigation command. The system has been designed in such a way that it provides pipe conveyance and surface irrigation for rice cultivation during monsoon season and pressurized irrigation during post monsoon period through a hybrid system of sprinkler and drip with four outlets for sprinkler irrigating 2.8 ha area and two outlets for drip irrigating 1.9 ha area. The system is also capable of providing irrigation through drip to part of a command during summer for third crop using water stored in service reservoir after the canal is closed in first week of April. To take care of sediment in the canal water, there are three stages of filtration: first by hydrocyclone filter which filters heavy suspended materials viz. sand, silt, etc., then by the sand filter and finally by the screen filter. The filtration at three stages reduces the turbidity to the desired level. It has been found that three-stage filtration reduced the turbidity to two NTU which is within permissible limit. Considering the cost of water saved, a benefit-cost ratio of the system was found out to be 1.126. This B: C ratio can be further increased by increasing the productivity of the fish and papaya in service reservoir area and better crop management during summer season.