A Long Way to Go: How AI Powered Tech Can Close the Gap on Long COVID

It is estimated more than 23 million Americans have long COVID, with as many as a quarter of worldwide COVID patients suffering from long COVID symptoms. And long COVID does not discriminate based on age or health, either – about 53% of long COVID sufferers are between the ages of 18 and 49, and many were previously healthy with no pre-existing conditions.
 
Although we do not know everything about long COVID yet, we do know it can cause a range of medical issues, including inflammation, damage to vital organs, and neurological problems. As the research progresses, a few things have become clear: firstly, many more people will be affected as the pandemic continues; secondly, this is a whole body, multiorgan problem; and thirdly, we need a comprehensive plan to diagnose and treat long COVID resulting from future variants.
 
As is often the case in medical research, the best place to start is by collecting data that helps us understand the disease: if we do not understand it, we cannot treat it. Data collection and research around long COVID has been difficult because symptoms vary so widely, and it’s new.
 
One of the significant shortcomings in the U.S. response has been the lack of centralized data collection. The UK has fared a bit better because of its centralized healthcare system, with its existing infrastructure called Biobank that allows for compiling large-scale patient data regularly and gives researchers access to long COVID data for clinical research and trials. In the UK, we have also opened 80+ clinics around the country and allocated more than £27m in funding to begin collecting data earlier in the pandemic.
 
As a result of our existing health care infrastructures and additional government support, we have deployed a cross-disciplinary study called Stimulate-ICP, which is currently recruiting 4,000+ people with long COVID. The Stimulate study team members range from clinicians, including primary care and specialist service health care providers, epidemiologists, mental health experts, and health economists. One of the tools used in this study is Perspectum’s artificial intelligence enhanced CoverScan multiorgan assessment tool, applied in earlier UK COVID studies, to inform new integrated care pathways for people with long COVID.
 
Advanced precision imaging validated and enabled by AI technology like ours can be a vital tool in the effort to clinically understand long COVID. As noted, long COVID is typically a multiorgan, whole body problem: primary research data shows that patients can suffer inflammation of the heart, scarring to the lungs, and damage to the kidneys. But every patient’s experience with long COVID is different and as researchers and clinicians we fall short if we only evaluate the impact on one part of the body versus another if the effects vary patient-to-patient.
 
This is where I believe precision medicine powered by AI excels by leaps on bounds over standard, non-AI enabled, imaging modalities and can eliminate the need for invasive diagnostic procedures and lessen our reliance on often less than accurate test results. 
 
Products like Perspectum’s FDA cleared CoverScan – also with the UK’s CA-marking for specific use to help diagnose long COVID – can provide a broader picture of health than a traditional scan. By offering a 49-metric assessment of multiorgan tissue at once, physicians can see the effect over time on a patients’ heart, kidneys, liver, lungs, pancreas, and spleen, allowing for accurate data visualization of long COVID’s impact. This is especially critical when looking at the results of a recent study conducted by Dennis et al., which showed that even in patients with normal blood test results, CoverScan uncovered organ damage that otherwise would have gone unnoticed.
 
The real value of this technology – and the reason why I was originally so keen to pivot its application to long COVID in 2020 – is to provide physicians with more information so that patients who are struggling can make more informed decisions with their health care provider.
 
This is the true value proposition of AI-powered technology: to provide physicians and patients with a more efficient, more accurate and interpretable picture of health. By leveraging the benefits of artificial intelligence – namely, the efficiency it provides and what it can detect and when – clinicians can have more information about their patients’ health, quicker. 
 
To help patients who are grappling with the lonely and confusing journey of long COVID, we first must ensure the global healthcare community is seeing the condition for what it is. Tackling this challenge head on requires solutions that transcend geography, specialties, and models of care. We must recognize the value to future generations of aggregating disparate patient data on long COVID along with other comorbidities, share best practices and approaches, and leverage all the tools at our disposal. In the U.S., medical professionals must push for research and centralized data collection that will allow technology to understand better the patterns which will help them diagnose and rehabilitate long COVID patients.
 
As clinicians turned product developers, we bear a collective responsibility to develop the tools that will help us reach a point where we are accustomed to using the best available tools — often wrongly labeled as disruptive — for diagnosing, treating, and rehabilitating the millions of people worldwide whose lives have been changed by long COVID. Over the next several years more tools will become available, but I believe that we have what we need to begin helping patients immediately – we need only start taking advantage of them.
 
Bibliography

1. https://www.alumni.ox.ac.uk/article/long-COVID-experts-pespectum-share-startling-research
2. https://urldefense.com/v3/__https:/www.youtube.com/watch?v=MkmBgZ_nDiw__;!!Gajz09w!CozaZkiEnHxQvRVqlxqUIg-ZKeWgMVTeliQ8U2IgiQ1NLQ6xFUPLOGhn1-GF7Ohpi3njGPuRsvEOUbvGbvyUzJ53BFIQTi6h$
3. How US, UK Are Responding to Long COVID-19 Symptoms – Bloomberg
4. New Clues To Long COVID: Prolonged Inflammatory Response (forbes.com)
5. Long COVID: Rehab centers set up across Europe to treat long-term effects of coronavirus – CNN
6. Long COVID: Long-Term Effects of COVID-19 | Johns Hopkins Medicine
7. Multiorgan impairment in low-risk individuals with post-COVID-19 syndrome: a prospective, community-based study | BMJ Open
8. https://www.ukbiobank.ac.uk/
9. https://www.stimulate-icp.org/about

 

---

Rajarshi Banerjee, Ph.D. has served as chief executive officer and a member of the board of directors since November 2012. Prior to co-founding Perspectum in 2012, Dr. Banerjee was a research fellow at the University of Oxford and developed the magnetic resonance imaging techniques for rapid non-invasive liver assessment and commercialized the method as LiverMultiScan, which has received 510(k) clearance from the FDA and is used in over 400 sites. Dr. Banerjee continues to work as a consultant physician with the Oxford University Hospitals NHS Foundation Trust, with research into the phenotyping of liver disease at an individual and population level in adults and children. Dr. Banerjee has many years of experience in running clinical trials and working in multidisciplinary scientific teams to develop applications for cutting-edge imaging technology and is a keen advocate of smart trial design. Dr. Banerjee received a BM. BCh. in Medicine and a Ph.D. in Cardiovascular Health from the University of Oxford, as well as a MSc. in Public Health from the London School of Hygiene and Tropical Medicine.