*2.4.2 AI-enhanced data analysis for outbreak detection, early warning and flow adjustment*

In order to enhance the timeliness and accuracy of outbreak detection and early warning approaches, public health researchers continuously analyze and explore sensor data and indicators to and from the physical world, including health, environmental, social, financial, and economic aspects, among others. Deep learning has been used to identify multiple infectious disease outbreaks. A dynamic neural network model was created to predict the probability of infectious disease outbreaks in the United States, such as Zika virus (ZIKV). Decisionmakings can easily modify the risk of an indicator, the risk classification system, and the forecast window for prediction based on their own unique needs [19]. Support vector machine (SVM), gradient boosting machine, and random forest (RF) were applied to simulate the global distribution of infectious diseases. To train the models, multidimensional and multidisciplinary datasets were qualified and quantified, such as social variables, incident medical records, high-risk areas, and cyberspace data. The suitability of the temperature has been stated to have the best discriminatory power among variables, and random forest (RF) is known to obtain the highest area under curve (AUC) value [20]. Each bootstrap sample was fitted with an unpruned decision tree. The risk maps were accurate in over 80% of the observed risk ranks falling within the 80% prediction interval, according to random bootstrap samples drawn from the results [21]. The use of data from

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

the cyberspace, such as keyword google searches, Key Opinion Leaders' blogs, and social media networking messages, has taken significant effort. Machine learning has been used for sentiment analysis and text classification from social media data for surveillance purposes. In India, a social media-based early warning system for mosquito-borne disease has been proposed [18].

## *2.4.3 What are the latest effects of intelligent early detection of infectious diseases, and what does this mean for the global battle against the epidemic in the future?*

There are a number of major effects: For instance, there was no intelligent big data research in the past. The network accounts of various hospitals could not be compared after a single patient was diagnosed with an infectious disease. Now, using AI's dynamic perception, the device may display an outbreak or cluster of infectious disease under uncommon conditions in real time via case reports. Second, the use of AI technology to evaluate the infectious case's time, space, and meteorological factors may have an effect on the local agricultural product market and economic conditions. Third, disease patterns can be forecast and early alerts for key ties can be issued using infectious disease data and local environmental monitoring. While AI's dynamic models of infectious diseases are consistent, the neural network model of experts must be introduced because infectious diseases have different epidemics in different regions.

### **2.5 Suggestions for accelerating AI-enabled public health emergency crisis response**

#### *2.5.1 Strengthen scientific research*

The AI industry should concentrate on core technology research and development in order to address technological challenges. Overall, the application of AI technology for disease prevention and control is still in its early stages of growth. Furthermore, AI also has an inexplicability that prevents it from being fully incorporated into the epidemiological system. The use of AI technology in disease prevention has been hampered by the lack of timely data collection and integration capabilities. As a result, play a bigger role in command. Epidemic modeling can be used to perform theoretical research on interpretability and improve the processing of large multi-dimensional data to this end.

#### *2.5.2 Expand AI application scenarios*

AI has been commonly used in the medical field as a result of continuous optimization of medical data and algorithm models. AI has achieved a great improvement in work efficiency in a subversive way, particularly during this special time of the new crown epidemic, and has spawned new demands. The use of AI has demonstrated quick landings, a wide range of effects, and major effects. However, AI in disease prevention and control is still in its early stages of research, and there are still flaws and issues in many areas. The use of AI in disease prevention and control should be thoroughly investigated, and a set of creative and reliable AI approaches should be used to aid in the detection and treatment of epidemics, as well as to minimize the risk of staff cross-infection. Improve disease management and control effectiveness, and provide strong scientific and technical support for winning the fight against epidemic prevention and control.
