Indian Researchers developed four Covid 19 detection methods using A I.
Dhaliya Nazreen
SARS-CoV-2 (virus) COVID-19 (disease) | credits: Wikipedia
The corona virus out break is being beyond the control of human more than a year. Even though remarkable achievements are there in medical field, we couldn't have escaped completed from Covid 19. Medical field and infrastructures are really frustrated due to the increase in cases, so it is really crucial to provide innovative ideas to handle this situation carefully. Indian AI researches came forward with the possibility of usage of AI for detecting the corona virus.
Deep learning for screening Covid 19 using chest X-ray images
Since the Covid 19 cases are increasing drastically and the high false negative on commonly used RT-PCR test, we need a simple alternative technique for the safety of man kind. For example, screening mechanism of covid 19 cases using radiological images like chest X-rays, assumes its importance. In comparting Deep Learning (DL) and Machine Learning (ML) offers fast, accurate strategies to detect the abnormalities of lungs which are likely to be the symptoms of Covid 19. However the available dataset of Covid 19 are inadequate to train deep neutral networks. Thus researchers brought up a new technique called Domain Extension Transfer Learning (DETL). When researchers implemented DETL it is found that 4 classes and one a large cheat X-ray dataset. Here deep consolation neutral network is used.
Multi-Task Driven Explainable Diagnoses of Covid 19 using chest X-ray images
X-ray is most commonly used image detection technology across the world. Researchers found out the possibilities of X-ray technique to detect Covid 19 cases. The first step is to segment and align the input images using a canonical form followed by classification and prediction. The classification pipeline is of 2 stages, the first step is to classify healthy and unhealthy chest X-ray images. If the output result is unhealthy, next step is to identify whether chest images have symptoms of pneumonia.
AI based Intelligent Covid 19 detector Technology for Medical Assistance (ATMAN)
The centre of Artificial Intelligence and Robotic developed at DRDO have developed an AI based intelligent, secure, web based covid detection application software named as ATMAN. This software can classify normal, Covid 19 and pneumonia classes of chest X-rays within a hundred samples. The virus affected lungs, even before showing symptoms, it compared with the RT-PCR test that had taken some how before and an X-ray based diagnostic tool can detect the infection and early stages by processing the image for seconds. This software in early stages and accessible in internet and can be used in various devices like mobile, tablets, laptop or computer.
Rapid Identification of Covid 19 biomarkers in blood plasma using Raman spectroscopy and AI (by Indian Institute of Science, Bangalore)
Raman spectroscopy is non destructive chemical analysis that utilises the interaction of light with chemical bonds of material. Since AI as deep learning is incorporated, this technique is rapid and there is no requirement of a procedure or reagent, only spectrometer and patients sample are required. Spectra collected from infected Covid 19 patients and healthy controls will be used for training and building robust classification model.