Interview with Dr Carolyn Lam co-founder of eko.ai.
Echocardiography is a common diagnostic tool used for identification of heart disease. Ultrasound beams are used to produce images of the heart allowing the cardiologist to see the beating of the heart and pumping of blood.
Expanding from the current technology of echocardiography, Singapore-based start-up, eko.ai, is looking to amalgamate it with artificial intelligence tools for easier, more efficient, and scalable, automated detection of heart disease.
Founded by a team of experts in the fields of medicine and entrepreneurship, eko.ai aims to use machine learning tools to automate the fight against heart disease. With initial funding from Singapore’s Agency for Science, Technology and Research’s (A*STAR’s) commercial arm, A*ccelerate, the start-up was able to connect with three of the informatics institutions at A*STAR to kick start the development of the technology. These institutions include the Institute for Infocomm Research, Institute of High Performance Computing, and Bioinformatics Institute.
Since its inception in early 2018, eko.ai has been recognised as a promising and innovative game-changer for the fight against heart disease. Notably, it was awarded the Startup SG grand prize of S$250,000 Startup SG grant at the SLINGSHOT 2019 competition organised by Enterprise Singapore. The competition was held in Singapore in November 2019 and brought together start-ups from over 2,400 application from more than 120 countries.
Eko.ai: The Game Changer
“Our mission is to automate the fight against heart disease.” Shared Dr Carolyn Lam. “Echocardiography, or ultrasound of the heart, is the first tool that cardiologists will reach for to examine the heart – it’s very safe and highly acceptable to the patient, and provides important information not available in an electrocardiogram.”
Dr. Lam elucidated that echocardiography is now limited to specialists because cardiologists need to be specially trained to interpret the images. “With AI, we aim to broaden access to echocardiography by improving the efficiency, reproducibility and accuracy of image interpretation.”
Current waiting times and turnover of reports for an echocardiograph weigh down on efficiency. The current process is tedious and time-consuming for cardiologists who would have to analyse and manually outline the images of the heart from the echocardiograph. The eko.ai software automates this process; thus “turning a process that takes 30 minutes with 250 clicks and 20 percent variability, taken down to two minutes, a single click and with zero variability.”
“Beyond efficiency and reproducibility, eko.ai will measure every single frame [from the echocardiograph], thus using much more information than the human mind can process and giving you much more information from the same study and potentially allowing disease detection to happen earlier, even before symptoms develop.” Said Dr Lam explaining how the eko.ai software will match up and even do better compared with the current process.
Democratizing the Echocardiograph
“Our ultimate moonshot dream is for everyone to have access to echocardiography. The hardware is now getting simpler with echo probes available in the market which can be connected to one’s smart phone. But the software needed to interpret the images – to help you tell what is normal versus not – is lacking. This is the urgent need that eKo.ai is addressing.” Said Dr Lam.
By providing the needed decision support for fully automated echocardiographic analysis and interpretation, Eko.ai hopes to one day democratize the echocardiogram, placing heart health in everyone’s hands.
“[This tool] would be applicable all the way from preventive to those who are already diagnosed with heart disease and even for clinical trials.” Said Dr Lam.
Collaborations
At present, the eko.ai technology is being used in research settings in its partner laboratories. These include both commercial and academic research collaborations with multiple partners, such as Brigham and Women’s Hospital, Samsung Medical Center’s Heart, Vascular and Stroke Institute, and the University of Alberta.
To further its efforts in driving better disease management and efficient heart disease detection, eko.ai, has established a global partnership with AstraZeneca with particular focus on the early detection, prevention and treatment of heart failure.
Earlier in January 2020, eko.ai raised US$4 million in a funding round co-led by Sequoia India and Singapore government-linked strategic investor EDBI together with Partech ventures, SGInnovate and Startup Health. The funding will boost its efforts in the development machine learning platform for the automations of echocardiography. [APBN]
About the Interviewee
Dr Carolyn Lam, Senior Consultant Cardiologist, National Heart Centre of Singapore. Professor of Cardiology at Duke-NUS Graduate Medical School. Leads multiple global clinical trials. Received the National Medical Research Council Clinician Scientist Award, L’Oreal Women in Science Award and several other awards. Serves as Associate Editor of Circulation, and the European Journal of Heart Failure, as well as on the editorial board of multiple other journals. Has published over 300 peer-reviewed articles and books. Host of the weekly medical podcast Circulation on the Run, and co-host of the television show Body & Soul.