The pandemic saw the rapid emergence of artificial intelligence (AI) and machine learning (ML) in providing timely and quality healthcare for millions. Healthtech startup Qure.ai created a niche for itself with advanced technology that reads and interprets medical images like X-rays, CTs, and ultrasounds in less than a minute, making equitable and high-quality healthcare a reality across the globe. Its automated medical imaging tools can shorten the time to diagnosis while enabling physicians to triage medical cases more effectively, especially in time-sensitive situations.
Below are edited excerpts from the interview:
The Covid-19 pandemic has cemented the use of AI in public health, being used from radiology to preventive health checks in India. What distinguishes Qure.ai from other companies using AI/ML for radiology?
AI enabled healthcare has emerged as a sunrise sector across the globe, especially since COVID19. The use of AI in global health scenarios will become prevalent as technology becomes more nuanced and applied. This is where Qure.ai has the upper hand among its competitors.
The biggest difference is that we are present in more than 600 sites in over 60 countries across the UK, the US, SEA, LATAM, Australia and Africa, making us the most deployed AI in medical imaging solution globally. That is the difference.
During COVID, we worked with partners across the spectrum – including Oman’s Ministry of Health, for patient monitoring; NHS Bolton in the UK, one of the first NHS Trusts to adopt AI-aided technology for COVID detection, and the BMC in Mumbai, India for screening and monitoring COVID positive cases.
The sheer volume of scans we were able to process validates the scalability of our team and the ease with which we can deploy AI, irrespective of the partner’s unique settings or their caseload.
Our unique positioning stems from the fact that we are present at every touch point on the healthcare delivery pathway. From public health screening programs to primary care to tertiary care to private care to the Ministry of Health, our solutions have been playing crucial roles in positively impacting patient outcomes.
We work with large teleradiology firms like Medica — the UK’s largest teleradiology, which serves over 200 NHS Trusts along with vRad which is US’s largest teleradiology serving more than 2,000 hospitals in the US.
Similarly, Lumius Imaging from Australia uses the AI in over 130+ imaging centres. In the UK, we have the most deployed AI solutions — over 20 NHS Trust hospitals use the AI solution.
We work with the largest GP network in Malaysia while at the same time work with MOH in countries like the UAE, South Africa, Philippines, India etc. The diversity and the huge scale of the user-base is the real advantage we have.
What are the challenges faced by medical practitioners in adopting and leveraging AI in India? How are you working with relevant stakeholders in making this adoption seamless?
AI in healthcare is a fast-advancing field with multiple players innovating across the spectrum. The biggest challenge is physicians not being aware of the exact way AI can help them in delivering patient care better.
Educating radiology teams and resources to use AI optimally in their daily workflow will be crucial in bringing about the change. Qure.ai works very closely with partners to help their resources integrate the solution as seamlessly as possible and leverage it to the maximum for the best results. Educating and informing professionals about the advantages of AI is crucial in changing its perception.
How do you ensure that the quality of data used in your AI & ML models is optimal?
Qure’s solutions are trained on one of the largest data sets in the market. Our algorithm is trained to detect sub-optimal scans automatically and flag them if needed. We ensure that the AI is regularly trained with the help of radiologists and specialists to maintain clinical accuracy.
We do double reading along with a strong ground-truthing mechanism. We also use an ideal mix of optimal and sub-optimal scans to improve the sensitivity of our solutions. The conscious decision to include sub-optimal and ill-captured scans into the data set was to make it sensitive to these during real-world deployment.
Our AI can read and report on images that are captured using handheld devices like a mobile camera in cases of Analog X-Ray setups. This makes it extremely relevant in remote deployment sites in developing markets.
It is also able to read scans from different machines with similar accuracy, making it machine-agnostic. The variety and quantity of data coupled with proper annotation makes the AI extremely robust.
Tell us about your expansion plans in India
We are deeply integrated with the country’s public health infrastructure — our solutions are in use across the band — from government PHCs to NGOs and private healthcare providers. There approximately 24,855 rural PHCs and 5,190 urban PHCs are operational in India.
Our aim is to be present in every single one of them and bring in an equitable standard of care, irrespective of the location. With schemes like Ayushman Bharat, our aim is to help strengthen the comprehensive primary healthcare system for all of India.
Currently, Qure is working with NITI Aayog and Municipal Corporation for Greater Mumbai (MCGM) for the past few years and is continuing to be their AI partner across districts to actively screen and test for TB and other pulmonary diseases. PATH, an international NGO, has also been one of the early adopters of AI for Tuberculosis using Chest X-rays and other innovations.
Most recently, we had received funding from USAID under the Samridh programme for lung health screening across select locations in India and have an active IHF grant that enables us to cater to multiple rural and semi urban healthcare networks and augment their lung health care pathways.
How do you ensure the safety of data/patient information on the cloud?
As a responsible healthcare technology provider, we are committed to ensuring that our AI software is safe and effective. Qure’s solutions are GDPR and HIPPA compliant. Our solutions also meet world-class regulatory standards – FDA and CE.
We have rigorous cybersecurity controls in place to keep our information system up to date and secure. Moreover, we have protected and encrypted data at every level, both at source and in transit by ensuring that any data is de-identified before it leaves a client’s premises for cloud processing.
What has cloud technology enabled you to do better?
Qure.ai is tackling AI challenges in the healthcare industry and advancing digital healthcare through medical imaging AI solutions. We are deployed through AWS cloud solutions across our sites. On AWS, we are using EC2 for heavy processing and easy scalability with better performance. It provides us with 99.8 per cent SLA, which improves our performance while keeping the downtime to a minimum. We have also enabled automated backups and failovers in real time.
When it comes to data security and privacy, our data is stored in S3 for better security, scalability, performance, and data availability. Also, we have RDS for database reliability. We are using CloudTrail & CloudWatch which monitors and records servers and account activity throughout AWS infrastructure. Moreover, we have AWS WAF, which is a web application firewall at perimeter level to secure our web apps and APIs against malicious traffic, web exploits, botnets, etc.
Being on AWS has numerous benefits, cost effective scalability and compliance being the major ones. We can deploy our solutions to the remotest part of the world, powered by AWS, all the while keeping the cost incurred in check.
Being present globally also comes with the added responsibility of being compliant with the respective data privacy guidelines of the region, AWS plays a crucial role in this as well.