AJ Abdallat is CEO of Beyond Limits, the leader in artificial intelligence and cognitive computing.
This year ushered in a period of unpredictability; individuals, businesses, industries and economies were suddenly battling Covid-19. Few were exempt from the challenges presented by this pandemic, and many are seeking solutions that can help.
This moment in time has uncovered just how crucial AI solutions can be for the future of healthcare. Rapid changes have made it difficult to manage the pandemic’s spread and determine what the industry will look like after coming through the other side. Regardless of complications, it’s still the responsibility of leadership teams to use every tool at their disposal to manage the pandemic and be better prepared for the future. It matters that legitimate attempts are made by all — for the good of all — to pursue pioneering solutions in the face of this global challenge.
As the CEO of an AI software company moving into the healthcare vertical, I quickly realized this move needed to be expedited as Covid-19 presented itself. In collaboration with the CEO of a medical facility, my company produced a Covid-19 predictive model in conjunction with their superlative medical knowledge. I learned during this experience how quickly this pandemic has shortened the transition from ideation to utilization, and how well positioned AI is to aid advancements in healthcare.
Artificial Intelligence: A Driving Force For Healthcare
Critical industries like healthcare require sophisticated methods for increasing visibility across operations. From cutting-edge wearables providing more insight into individual health to expert support via reliable prognostic exploration, the medical community is realizing the benefits of numerous AI solutions helping drive dynamic advancements for healthcare.
Examples of solutions taking a stand in the face of this pandemic include predictive models such as Penn Medicine’s CHIME, a Covid-19 capacity planning tool, and Washington State’s predictive dashboard, another pandemic risk assessment tool.
The predictive model my team built in partnership with renowned medical professionals focuses on the domain of expert knowledge. It works with multiple layers of heterogeneous and highly dimensional data categories to evaluate the consequences of social mobility on infection and hospitalization rates while estimating resource allocation supply and demand. Integrating multiple sources of time-series and spatial data into a model with learning capabilities was essential for more accurate decision-making.
This process has shown that when it comes to these predictive tools, it’s vital to understand that in order to utilize all of this data effectively, most conventional data processing, classification, pattern recognition, modeling and forecasting tasks must be automated and streamlined. This is key to delivering near real-time models with learning capabilities.
Such advancements in data science and AI can offer significant value in terms of creating data-driven, evidence-based tools that improve predictive performance over time. Codifying medical domain expertise in epidemiology with advanced data science techniques into models has the potential to protect millions of people from the risks of future pandemics.
Additional examples of AI integration into healthcare working in the fight against Covid-19, include chatbots and telehealth initiatives that utilize conversational AI, as well as AI-based drug and vaccine discovery and medical insurance fulfillment/billing optimization solutions.
The Secrets To Successful AI Adoption And Implementation
AI adoption and implementation are the two biggest challenges in this technology becoming more commonly used in healthcare. A dichotomy emerges when discussing the realities of the industry’s AI adoption efforts, but a critical principle also lives at the center of the divided sentiment. This is where we find a key consideration: prioritizing expert input to ensure humans are always in the loop through explainable recommendations.
Explainable AI is an important component to trusted AI that I believe could help forge the way forward to implementation — AI that’s auditable and accountable, and that amplifies the capabilities of the human experts it seeks to support rather than replace.
Other tactical, actionable considerations for how industry leaders and professionals facilitate successful AI adoption and implementation include:
• Working directly with and learning from creators of the AI solution in question. AI engineers should be training (and working on the ground in tandem with) healthcare leaders, doctors and professionals who will be using the solutions. No matter how powerful a solution may be, it will never be adopted if the UI/UX is confusing or impractical for users’ day-to-day purposes. User-friendly UI understandable by everyone in the organization is vital.
• A shift in industry perspective regarding wide-spread AI adoption. Leadership hesitation to adopt is often triggered by patient reluctance toward AI, a technology they may fear out of misunderstanding. To safeguard patient trust in AI, transparency regarding the inherent value of AI and its functionality is key. Easing patients into a better comprehension of AI’s benefits with clarity on how it can help aid in their care creates a more comfortable culture overall.
• Implementation of best practices to support this shift in perspective. In addition, nurturing a digital outlook by exploring new tech and involving staff at every level, while building on their capabilities and outlining an AI approach out the gate, are fundamental factors. Engaging doctors, nurses, technicians, administrators and other professionals with a well-defined outlook on success will encourage deployment participation.
Despite the difficulties created by Covid-19, leadership teams should stay the course for the good of their own business while making an effort to guide their customers, and the global community at large, by pursuing solutions that may help the world push past this pandemic.