PANAJI: Assistant professor, National Institute of Technology (NIT), Goa, Shivnaryan Patidar, is part of a three-member team that developed a software which enables a smartphone or similar hand-held device to detect the cardiac disorder, atrial fibrillation (AF). The condition affects around two per cent of world's population.
The software was recently tested by PhysioNet, a wing of the primary agency of the United States responsible for biomedical and public health research, and secured third level in terms of accuracy.
"Patients suffering from AF have high risk of stroke, heart failure and coronary artery disease. A timely, easy and cost effective diagnosis of this disease is the only solution to avoid the risk associated with it," said Patidar, who developed the software along with NIT Goa research scholar Ashish Sharma and interventional cardiologist at Nakshatra Heart and Multispeciality Hospital, Indore, Madhya Pradesh, Dr Niranjan Garg.
The software was developed after six months of extensive work as part of a 'AF Classification' challenge thrown open by PhysioNet and an annual international scientific conference, Computing in Cardiology (CinC).
When presented at the conference in late September this year, the software was tested and recognised at the event held at University of Rennes-I, France.
"In many cases of AF, patients are not symptomatic at the initial stage and remain undiagnosed. Sometimes, self-negligence and unavailability of the appropriate medical facility, especially in remote areas, also become a reason for the disease to remain undiagnosed," Patidar said.
"It will be easier if this disease can be diagnosed automatically using hand-held devices such as the smartphone, which does not require any specialized skill to operate," Patidar said.
The researcher said that previous methods used to detect AF can only identify AF and normal heart rhythms, whereas the new software can detect a range of rhythms. This is important in accurately detecting AF because many non-AF rhythms can exhibit similar irregular intervals, Patidar said. The new software can tell the difference between the AF rhythms and other such similar ones.
"It was very challenging to reliably detect AF from a single short lead of ECG, and the broad taxonomy of rhythms makes this particularly difficult," he said.
The software was recently tested by PhysioNet, a wing of the primary agency of the United States responsible for biomedical and public health research, and secured third level in terms of accuracy.
"Patients suffering from AF have high risk of stroke, heart failure and coronary artery disease. A timely, easy and cost effective diagnosis of this disease is the only solution to avoid the risk associated with it," said Patidar, who developed the software along with NIT Goa research scholar Ashish Sharma and interventional cardiologist at Nakshatra Heart and Multispeciality Hospital, Indore, Madhya Pradesh, Dr Niranjan Garg.
The software was developed after six months of extensive work as part of a 'AF Classification' challenge thrown open by PhysioNet and an annual international scientific conference, Computing in Cardiology (CinC).
When presented at the conference in late September this year, the software was tested and recognised at the event held at University of Rennes-I, France.
"In many cases of AF, patients are not symptomatic at the initial stage and remain undiagnosed. Sometimes, self-negligence and unavailability of the appropriate medical facility, especially in remote areas, also become a reason for the disease to remain undiagnosed," Patidar said.
"It will be easier if this disease can be diagnosed automatically using hand-held devices such as the smartphone, which does not require any specialized skill to operate," Patidar said.
The researcher said that previous methods used to detect AF can only identify AF and normal heart rhythms, whereas the new software can detect a range of rhythms. This is important in accurately detecting AF because many non-AF rhythms can exhibit similar irregular intervals, Patidar said. The new software can tell the difference between the AF rhythms and other such similar ones.
"It was very challenging to reliably detect AF from a single short lead of ECG, and the broad taxonomy of rhythms makes this particularly difficult," he said.
Get latest news & live updates on the go on your pc with News App. Download The Times of India news app for your device.
