Why the COVID-19 pandemic was a watershed moment for machine learning

By Taha Kass-Hout, Director of Machine Learning at Amazon Web Services (AWS)

Times of crisis spur innovation and creativity, as evidenced by how organizations have come together to innovate for the greater good during the COVID-19 pandemic. 3D printing companies made face shields and nose swabs to meet massive demands and automakers shifted gears to make ventilators.

Machine learning (ML) – computer systems that learn and adapt autonomously using algorithms and statistical models to analyze and draw conclusions from data patterns to inform and automate processes – also played an important role, supporting virtually every aspect of health care. Amazon Web Services has helped customers enable remote patient care, develop predictive surge planning to help manage inpatient/ICU bed capacity, and tackle the unprecedented feat of developing a messenger ribonucleic acid (mRNA) COVID-19 vaccine in less than a year.

We now have the opportunity to leverage the lessons of the past year to apply ML to help address several underlying issues plaguing the healthcare and life sciences communities.

Supporting healthy populations everywhere

Telehealth was on the rise before COVID-19, but it has revealed its true potential during the pandemic. Telehealth is often thought of simply as patients and providers interacting online via video platforms, but has proven capable of doing so much more. The application of ML to telehealth offers a unique opportunity to innovate, evolve, and deliver more personalized patient experiences and ensure they have access to the resources and care they need, wherever they are. that they find themselves.

ML-based telehealth tools such as patient service chatbots, call center interactions to better triage and direct patients to the information and care they need, and online self-service screenings help optimize patient experiences and streamline provider assessments and diagnostics.

For example, GovChat, South Africa’s largest citizen engagement platform, launched a COVID-19 chatbot in less than two weeks using an artificial intelligence (AI) service to create conversational interfaces in n’ any application using voice and text. The chatbot provides health advice and recommendations on whether to get tested for COVID-19, information on the nearest COVID-19 testing facility, the possibility of receiving test results and the possibility for citizens to report symptoms of COVID-19 for themselves, family members, or other household members.

Additionally, at the onset of the COVID-19 crisis, New York-based MetroPlusHealth identified approximately 85,000 individuals at risk (eg, comorbid heart or lung disease, or immunocompromised) who would require additional support services while sheltering in place. In order to engage and meet the needs of this high-risk population, MetroPlusHealth has developed ML-enabled solutions, including an SMS-based chatbot that guides people through self-screening and check-in processes, SMS notification campaigns to provide updated pandemic alerts and information and a community organization referral platform, called Now Pow, to connect each individual to the right resource to ensure their specific needs have been met. satisfied.

By providing patients with an easy way to access the care, referrals and support they need, ML has given providers the ability to innovate and adapt their telehealth platforms to meet diverse and ever-changing needs. from the community. Agile, scalable and accessible telehealth continues to be important as providers look for ways to reach and engage patients in hard-to-reach or rural areas and those with mobility challenges. Organizations and policy makers around the world must make telehealth and easy access to care a priority now and in the future to close critical care gaps.

The Shift to Precision Treatment and Prevention

Beyond unprecedented changes in the approach to patient engagement, support and treatment, COVID-19 has dictated a clear direction for the future of patient care: precision medicine.

Guidelines for planning patient care have moved from statistically significant results collected from a general population to results based on the individual. This gives clinicians the ability to understand what type of patient is most likely to have a disease, not just what type of disease a specific patient has. Being able to predict the probability of contracting a disease long before it appears is important for determining and initiating preventive, intervention and corrective measures that can be adapted to the characteristics of each individual.

One of the best examples of how ML is enabling precision medicine is biotech company Moderna’s ability to accelerate every step of the development process for an mRNA vaccine for COVID-19. Moderna began working on its vaccine the moment the genetic sequence for the novel coronavirus was released. Within days, the company had finalized the sequence for its mRNA vaccine in partnership with the National Institutes of Health.

Moderna was able to begin manufacturing the first batch of clinical-grade vaccine within two months of completing sequencing, a process that has historically taken up to 10 years.

Enable better educated and more engaged patients

Personalized health isn’t just about treating illness, it’s about providing access to resources and information specific to a patient’s needs. ML plays a key role in curating content that can help educate and support patients, caregivers and their families.

Breastcancer.org allows people with breast cancer to upload their pathology report to a private and secure personal account. The organization uses ML-based natural language processing to analyze and understand the report and create personalized information for the patient based on their specific pathology.

Making COVID-19 lessons count

Over the past decade, organizations have focused on digitizing healthcare. Today, making sense of captured data will provide the greatest opportunity to transform care. A successful transformation will depend on data being able to flow where it needs to be at the right time while ensuring the security of all data exchanges.

Interoperability is by far one of the most important topics in this discussion. Today, most health data is stored in disparate formats (eg, medical history, doctor’s notes, and medical imaging reports), making it difficult to extract information. ML models trained to support healthcare and life sciences organizations help solve this problem by automatically normalizing, indexing, structuring and analyzing data.

ML has the potential to bring data together in ways that create a more complete view of a patient’s medical history, making it easier for providers to understand data relationships and compare specific data to the rest of the population. . Better data management and analysis leads to better information, which leads to smarter decisions. The net result is increased operational efficiency for better care delivery and management, and most importantly, improved patient experience and health outcomes.

Looking ahead, imagine a time when our pernicious medical conditions like cancer and diabetes can be treated with bespoke medicines and care plans enabled by AI and ML. The pandemic has been a turning point in how ML can be applied to tackle some of the toughest challenges in healthcare, though we’ve only scratched the surface of what it can accomplish.

Sherry J. Basler