While COVID-19 is extracting a terrible price on human life and causing significant economic hardship, it is helping to catalyze a transformation in care delivery.  As noted in my prior post, AI is improving and accelerating the discovery of new therapies. Beyond this, AI and machine learning (ML) are also transforming clinical diagnosis, by improving accuracy and enabling diagnosis to take place outside the doctor’s office. 

Not too long ago, the traditional process of clinical diagnosis – compiling a history of symptoms, physical examination and testing – could only be performed with in-person visits. Now, the demand on healthcare facilities and healthcare workers, along with the increased risk associated with in-person appointments due to COVID-19, are leading clinicians and consumers to rely on the technology advances for initial and final clinical diagnosis. 

Supported by the pattern recognition and mapping capabilities of AI, clinicians can review and confirm diagnoses without placing themselves at risk, enabling them to see more patients while focusing on those in greatest need. Consumer and clinician willingness to use digital diagnostics to battle COVID-19, as well as an evolving and supportive regulatory and reimbursement environment, are accelerating the adoption of remote and digital diagnosis. 

The Changing Diagnostic Landscape

In this aspect of the fight against COVID-19, the disease is being met by a determined set of researchers, investors and entrepreneurs and corporate titans whose innovations in manufacturing, connectivity and embedded analytical capabilities will ensure that, post-pandemic, a trip to the doctor’s office will no longer require a set of car keys.

The breadth of FDA-approved solutions already in market is as impressive as the array of conditions they are designed to treat or manage. Diagnostic solutions that sustain market and clinical relevance are distinguished by the elegance of their digital design and consumer-oriented interfaces, plus a functionality that integrates and collaborates with consumer technology platforms. 

Basic biometric data is captured through consumer devices such as an iWatch and diagnostic platforms such as Tyto Care. Both are widely available and easy to use, plus they integrate with consumer technology platforms. Condition-specific diagnostic solutions that benefit clinicians and consumers include apps such as eMurmur and point-of-care-oriented imaging platforms with embedded AI capabilities, such as Kosmos. Beyond consumer and clinical devices, entrepreneurs are pushing boundaries by embedding sensors in fabric – examples include Myant and Nanoware – which demonstrates how healthy activities intersect with healthcare and health outcomes. 

Turning to AI for Speed, Scale and Complexity 

The battle against COVID-19 is accelerating many efforts to integrate AI into diagnostic tools and platforms, in the interest of speeding clinical decision making and initiation of treatment. For example, researchers from the University of Maryland School of Medicine and RadLogics developed an AI-based system to identify COVID-19 patients from CT images, which could lead to reduced case volumes for radiologists and expand the healthcare system’s diagnostic capacity.

Another innovative tool – developed at Ben-Gurion University and currently being validated – employs spectroscopy analysis on a cloud-connected system to detect virus carriers via a breath test (or nose/throat swabs) within one minute and with greater than 90% accuracy.

The battle against COVID-19 requires a holistic approach, integrating clinical data with consumer data to understand disease progression and transmission, as well as the variances that exist in both aspects for different population groups. The scale and complexity of the information available requires the development and deployment of AI.

Our healthcare data experts have also engaged in the skirmishes that preceded and will succeed COVID-19. For example, we have combined genetic, preclinical/clinical, safety and real-world data from internal and external sources into a federated data model from which we can gather AI-informed patient health insights. An example of applying these insights would be automating a breast cancer patient’s journey through diagnostic image recognition, composite risk estimation, disease onset/progression prediction and personalized treatment recommendations.

The Changing Regulatory & Reimbursement View

All of these solutions must fit within our regulatory and reimbursement environment. Fortunately, in response to COVID-19, the Centers for Medicare and Medicaid Services (CMS) has used regulatory flexibility to significantly broaden both the services that Medicare covers via telehealth, as well as the circumstances under which they are covered. Commercial payers and self-insured plans are also expanding access to and reimbursement for these services.

In addition to these immediate regulatory changes, the U.S. Department of Health & Human Services’ DRIVe Initiative is helping to prepare us for the next pandemic by accelerating the development and availability of transformative technologies, such as novel sensors, disease tracking and monitoring platforms. The program also seeks to develop a broader suite of analytical tools that improve the speed and accuracy of diagnosis to protect public health.   

COVID-19 is a catalyst in the transformation of U.S. healthcare. Powered by digital tools and platforms, entrepreneurial zeal and consumer demand, AI and ML will transform how and where disease is diagnosed. Through the virtuous cycle of data-driven innovation, the personalized and predictive nature of diagnosis will help us tame COVID-19 and the diseases we will face tomorrow. 

Visit our COVID-19 resources page for additional insights and updates.

Brian Williams

Brian Williams

Brian is Cognizant’s Chief Digital Officer for Life Sciences and is responsible for designing digitally enabled solutions to facilitate care access and... Read more