ABSTRACT
Recent research trends in the field image processing have focussed on challenges and few techniques for processing and classification tasks related to it. Image classification aims at classifying images based on several predefined categories. Several research works have been carried out to overcome shortcomings in image classification, nevertheless the output was restricted to the elementary low-level picture. Several deep neural network techniques are employed for image classification such as Convolutional Neural Network, Machine Learning Algorithms like Random Forest, SVM, etc. In this paper, we aim at designing a COVID-19 detection using the CNN model with support of Open-Source software such as Keras, Python, Google Colab, Google Drive, Kaggle, and Visual Studio for aggregate, design, create, train, visualize, and analyze bulk load of data on the cloud after programing a Deep neural network without a need for high-end processing hardware. We have made use of weights to test and analyse the accuracy, visualize and predict the condition of a lung using chest X-Rays at certain accuracy. This will help in identifying the problems of the patients at a faster rate, thus giving an appropriate treatment at an early stage itself to saving one life. © 2023 IEEE.
ABSTRACT
Introduction and Objective: Telemedicine is an established modality to deliver health care to patients remotely. Its role in pediatric urology followup among middle-class semi-urban families is unknown. We conducted a prospective observational questionnaire-based study to assess the patient and provider(urologist) satisfaction and feasibility of teleconsultation across different socioeconomic strata in follow-up of paediatric urology patients during the COVID 19 pandemic. Method(s): The guardians of children treated earlier and due for follow-up were explained and the appointment for teleconsultation was fixed using a video conferencing app. After consultation, consenting caregivers were explained about study and the provider survey was filled by consulting urologist, while the patient questionnaire was filled by principal investigator telephonically. Result(s): A total of 77 virtual visits were completed over 10 months. Median age was 8 years(IQR= 4 to 12) and 82% were boys. The clinical conditions were posterior urethral valves(22%), hypospadias(18%), PUJ obstruction(18%), vesicoureteric reflux(12%) and others(30%). Clinicians found that virtual visits were effective(78%) in deliverance of the care equivalent to the inpatient visit. Patients were equally satisfied(75%). Technical difficulties due to internet connectivity were faced in 24 visits(31%). Video clarity and inability to examine were additional limitations faced(23%). Majority(90%) belonged to the middle socioeconomic strata as per modified Kuppuswamy scale. Families were estimated to have saved a mean of 26,934 rupees(SD +- 7998.06) and a median of 7 days(Range 1-15) of travel time. Conclusion(s): Telemedicine has potential for successful follow-up with cost and time savings. With improving internet connectivity, its potential is likely to increase in future.