Social Media powered Healthcare

The day-by-day increase in the rate of health-related problems in Pakistan is staggering. The lack of adequate financial and economic resources causes these health issues to give rise to new diseases that are even more challenging to treat. According to the World Health Organization (WHO) report, Pakistan is currently ranked 122 out of 190 countries in terms of health. Moreover, Pakistan spends less than 3% of its GDP on the health sector, which is alarming for any developing country because, according to WHO, a country should spend at least 5% of its GDP on the health department.

Disease prevention and control, particularly at an early stage, are essential aspects of public health that safeguard the community by lowering the risk of illness and death for individuals. In the present era of the internet, almost everyone is a part of social media networks. Like in the rest of the world, social media users in Pakistan have grown over time. As of early 2022, Pakistan has an internet penetration rate of 36.5% of the total population, or 82.9 million internet users, with 71.7 million using social media. People share their stories, daily life activities, and problems on social media platforms, specifically on Facebook and Twitter. In the absence of an easy-to-reach healthcare system, people often post their health-related issues on social media. This data can be used to identify health-related problems at an early stage. Our main objective is to find out the community's diseases through a social media platform. 

The main hurdle in the identification of health-related problems in underdeveloped countries is the lack of data. Pakistan is one of the underdeveloped countries which lack robust mechanisms to assemble the data to make improvements and decisions in the healthcare departments. For collecting the data, we will use the Twitter platform. The main steps of the solution involve:

• Train a deep learning model which would be able to identify tweets related to healthcare issues, such as diseases or symptoms associated with a disease
• Use the deep learning model to identify tweets mentioned in part 1
• Identify the geographical location of the tweets
• Cluster the tweets with respect to a location
• Identify potential disease from the tweets
• Inform relevant healthcare departments about potential spread of the disease

In such a way, the healthcare departments will be able to oversee the disease at its early stages. Consequently, health departments will be able to make policies accordingly to prevent the disease in the community.

The solution will serve the people of Pakistan in first stage, but it can be scaled up to the world level, particularly the developing countries with internet access. The proposed solution will help in improving the healthcare situation in Pakistan and other developing countries. The solution will be able to detect disease spread at early stages and inform relevant healthcare departments. In such a way, actions can be taken to prevent the spread of the disease. This will greatly impact the local communities in terms of better health and low rates of disease spread, hence improving the life expectancy of people.

The solution will be a low-cost solution and would not require any additional human resources to be placed in different localities

The team has vast experience in working with data from social media. Dr. Rabeeh Abbasi has published and supervised many research works which use data from social media for solving various issues, for example, how blood donations are effectively managed through social media in a developing country. Ms. Izzah Salam’s MPhil research was about understanding the grievances of people through their posts on Twitter. Dr Akmal Khattak’s main area of research is information retrieval, which is of particular importance to this project, especially for understanding posts written in natural language.

Although we have implemented similar solutions, we have not tested these solutions on a larger scale. Through MIT Solve, we are hopeful to get support from WFP’s Innovation Accelerator to make our solution scalable and hence strengthen the healthcare systems in developing countries.

In developing nations, there are more industrial accidents, occupational illnesses, and pollution-related health problems than there are in developed nations.

Currently, in Pakistan, which is listed as a under developed country, has no mechanism to identify the community diseases on the social media platform. So, our approach is unique in identifying the community health related problems. After this we will also forward the extracted diseases to the respective health department. 

As, person put status on their platform, our model will categories. If that tweet will contain some symptoms or disease, notification will send to the respective health care department.

Our primary goal for the next year is to protect the community of Pakistan from health-related problems. As all know, in any community, the provision of medical services and the promotion of public health are vital for the development of communities and nations since they are directly tied to people's daily lives. In many developing nations, where many people continue to suffer as a result of high rates of infant, child, and maternal mortality; the spread of infectious diseases; a lack of access to a safe water supply; and inadequate sanitary facilities, improvements to public health and medical services are therefore becoming a top priority.

Our proposed method will continue to serve the community of Pakistan and also in other underdeveloped countries in the next five years. 

We can measure our progress by observing the posted tweets. Like, if the healthcare department takes proper action on the extracted disease, most probably there will be a minimum chance the community of that area will suffer from this type of disease again. 

Through this method, we can achieve the goal of good health and well-being in one of the underdeveloped countries.

Our proposed approach is based on machine learning, deep learning, artificial intelligence, and geospatial technology. We will also make software and applications for the health care department, who will respond to the extracted diseases in a timely manner and take action accordingly. 

Our team member, Ms. Izzah Salam, will collect the data from Twitter. 

Our model will have an impact on the community directly. If people's diseases were cured on time, the community would be healthy and they would perform their activities in a better way. It will automatically generate good revenue for the economy of Pakistan.