**3. Using AI to solve the issue of public health emergencies**

AI models have been applied to detect outbreaks of infectious diseases. Researchers have a long history of successfully developing a global outbreak surveillance approach using Internet-based approaches. Internet-based disease tracking approaches have provided a real-time alternative to conventional indicator-based public health disease surveillance [22, 23]. Internet-based monitoring systems use a range of open-source Internet data, including online news and social media, as well as other Internet-based data sources, to detect early warning signals of threats to public health. AI techniques have played a significant role in a series of data processing and analysis activities. AI techniques have recently become popular for completing tasks in highly dynamic, complex, and data-rich environments. In the modern age of public health approaches, it is critical and important. Machine learning and deep learning as AI core technologies are among the most important, methodologically, for fundamental and increasing interests, intense research activities in the interdisciplinary field of AI. Despite the impressive list of achievements already achieved, AI technologies in the sense of public health and public health monitoring are still in their early stages of growth, with a lot of potentials yet to be realized. Outbreak identification, early warning, trend prediction, and public health evidence-based approaches effectively response modeling and assessment are among the core tasks of public health surveillance and response, particularly in light of the current COVID-19 pandemic.

#### **3.1 Using AI to deal with public health emergencies**

By pinpointing specific demographics or geographies where population health issues exist, AI and machine learning can help to target and precisely implement education and treatment programs and reduce spending waste. AI enables computers to mimic the cognitive function of human minds, and machine learning gives computers the ability to learn without being explicitly programmed. By using AI and machine learning to review vast sets of real-time data, health experts can identify at-risk populations for any number of diseases, from diabetes to heart disease. Throughout the coronavirus pandemic, the industry has witnessed the power of clinical surveillance. With a broad array of discrete tests that can identify a COVID-19 infection, health systems and public health authorities have needed a way to interpret and track the patients with infections.

#### *3.1.1 Connecting the data with surveillance*

Data is at the heart of clinical surveillance. When data is combined with evidence-based clinical decision support, a single source of reality can be created that connects the disease's related symptoms, allowing for the discovery of how quickly a disease is progressing and what lab tests reveal. Keeping up with the latest advances in medical terminology and the related diagnosis and procedure codes is critical for recognizing clinical patterns as well as securing support, funding and reimbursement. Many health systems have transitioned to finding out the patterns in COVID-19 and better predicting respiratory and organ failures associated with the virus, despite being reluctant to implement technology in the past.

When the pandemic struck, healthcare providers immediately shifted their focus to include COVID-19 updates in their clinical surveillance activities. Hospitals and healthcare systems have been able to proactively monitor patient status for earlier interventions and broaden data flow in significant ways with a centralized, global

*Artificial Intelligence (AI) in Evidence-Based Approaches to Effectively Respond to Public… DOI: http://dx.doi.org/10.5772/intechopen.97499*

view of COVID-19 cases coupled with real-time alerting. Age, where the disease was possibly contracted, if the patient was examined, and how long the patient was in the ICU are only a few of the important patient measurements that have been monitored. Patients' pre-existing conditions were taken into account during surveillance. This data trail assists providers in developing a constantly evolving coronavirus profile and provides key data points for reporting to state and local governments and public health agencies. Clinical monitoring now brings together information from various areas of the hospital and clinics into a centralized view of COVID care, such as lab results, patient data, co-morbidities, mortality, and drugs, since there are no other ways to put together seemingly fragmented information.

#### *3.1.2 COVID-19 accelerated AI advancements*

COVID-19 puts people at risk of sepsis, so they wanted to identify those who were most at risk. Many AI-powered fast-tracking techniques were put to the test. This health epidemic shows what can be done to anticipate and avoid a variety of chronic health concerns. This technology can then be used to save lives and money in cases where prevention has proven to be ineffective. To achieve those savings, it is necessary to refine the use of AI for clinical surveillance; 2) extend access to everything from electronic health records (EHR) to knowledge that exists outside of direct clinical settings, ranging from the omics to social determinants of health; and 3) differentiate AI hype from solutions that offer proven, actionable insights for specific clinical concerns.

#### *3.1.3 The future of AI's prospects in clinical emergencies*

Though COVID-19 appeared to be a test ground for machine learning and AI, the industry had been focusing on harnessing technology's power for healthcareassociated infections (HAI) for some time. According to publicly available reports, HAIs cost the US healthcare system up to \$45 billion a year [24]. On any given day, about one out of every 31 patients will be infected with at least one HAI [25]. One example is *C. difficile* infections (C. diff). C. diff raises the risk of inpatient death and duration of stay, putting hospitals at risk of financial penalties. Machine learning, on the other hand, will predict which patients are at risk for C. diff infection, allowing physicians to treat patients more effectively and avoid the spread of the infection in hospitals. Hundreds of thousands of variables that may lead to C. diff, as well as how those factors interact, are analyzed using machine learning. It is always learning and incorporating new data and information. Machine learning, when used in a clinical surveillance system, may identify at-risk patients before their infection progresses, adding variables that physicians frequently find difficult to detect when handling several patients, as well as conditions that are outside of their normal scope of practice.

Rules-based systems are less effective for these "edge" scenarios, as researchers know, since each new data feature necessitates a new rule. AI at warp speed will help hospitals and communities respond to complex cases like COVID-19, C. diff, and even sepsis until clusters, outbreaks, or critical medical emergencies worsen. Clinical surveillance based on AI can monitor when relevant factors arise in a specific way and understand how timing plays a role in interactions. Time is difficult to incorporate, but recognizing when the white blood cell count has increased or decreased, for example, is crucial to make reliable C. diff predictions.

These types of forecasts can make a huge difference in clinical emergencies like brain injury, heart arrest, and respiratory failure in healthcare organizations all over the world - cases where minutes can mean the difference between life and death. Clinical surveillance with AI has the ability to provide next-generation decision-support resources that incorporate powerful technology, public health's preventive emphasis, and clinicians' diagnosis and treatment expertise. As a result, surveillance has the potential to play a key role in achieving the quality and cost goals that our industry has long pursued.
