Coronavirus is proved to be a severe epidemic diseasethroughout the world. Despite of endeavoring lot of medicalfacilities for mitigating with this pandemic, still the number ofinfected cases increased rapidly, which leads to lack of healthcareresources (i.e. hospitals, doctors and other healthcare amenities). Early stage risk prediction by analyzing several clinical andbehavioral risk factors is considered to be a promising solutionfor prescribing appropriate triage to patients and to reduce themortality rate due to this fetal disease. To cope up with thisproblem, in our study we have proposed a deep learning basedapproach for the early stage prediction of risk of infection andrisk of mortality in individuals possessing certain risk factors. Wehave utilized a publically available covid-19 dataset incorporatingseveral risk factors that may cause this infection. For the selectionof most significant risk factors i.e. with respect to their level ofimportance in risk prediction, we have employed three featuresselection techniques (i.e. f_classif, PCA and Tree). The set ofextracted features are the utilized for the training of proposed ANN for the prediction of infection risk and mortality risk due tocovid-19. For the performance analysis of proposed method, fourdifferent evaluation metrics are being employed including: MSE,MAE, ME and EV. The proposed model has achieved a minimum loss (MSE) of 0.00137 for infection risk prediction and MSE of0.000012 for mortality risk prediction.

Optical character recognition has received significant research focus to digitize the text in images. Urdu OCR is a difficult task as compared to English and similar languages due to its complex nature where a character can have multiple inflections depending upon its position in the word. The proposed research work presents segmentation-free approach (i.e. holistic approach) for offline Urdu printed text detection. To extract text lines in an image, horizontal histogram projection is employed whereas for ligature segmentation in extracted image text line, proposed technique has used connected components labelling. In this model, set of 14 statistical features along with HOG features are extracted for each sub-word/ligature and used for the training of the proposed model. An open-source dataset UPTI [10] has been used to train and test the proposed algorithm. SVM with RBF kernel function is used for the classification of ligatures. The proposed algorithm has achieved 97.3% character recognition rate on given dataset.