 |
Located within Education City, HBKU seeks to provide unparalleled opportunities where inquiry and discovery are integral to teaching and learning at all levels utilising a multidisciplinary approach across all focus areas.
HBKU is committed to actively contribute to achieving the Qatar National Vision 2030 by building and cultivating human capacity through an enriching academic experience and an innovative research ecosystem.
In this context, HBKU has developed a study on the use of AI in detecting cardiovascular disease.
Tanvir Alam, an assistant professor at College of Science and Engineering at HBKU has explained in an interview the use of AI in detecting cardiovascular disease (CVD).
Q. Given that CVD is a primary cause of death among the Qatari population, what is the potential for AI — as well as its subcategories, like machine learning — in providing personalised treatment plans for CVD diseases in Qatar?
A. Complex diseases like cardiovascular disease (CVD) often involve the interplay among demographic profile (i.e., gender, age, etc.), genetic profile, lifestyle, and environmental factors. As a result, integration of such diverse factors requires attention to cover the heterogeneity of the population as well. AI-based approaches for CVD rely on multimodal datasets that can excel the discovery of personalised treatment plans by identifying complex relationships among these factors. It is also demonstrated in literature that combining such a wide variety of clinical and genetic data can improve the early detection of CVD.
As part of our study, we integrated a multimodal dataset from Qatar Biobank reflecting the health status of Qatari nationals and we developed a highly accurate AI model for CVD detection and risk factor prediction. We think there is huge potential for such AI models in clinical setup. It will not only leverage the early detection of CVD in an accurate manner by avoiding redundant diagnosis but also support the medical practitioners in their decision-making and reducing their workload.
What is the primary data you are using in this study and how sufficient is this data? Do you have any specific targets groups such as Qataris, expats from certain countries or regions to explore certain traits of the diseases?
We mainly relied upon the clinical data and demographic information from Qatar Biobank (QBB) to enable evidence-based research toward the development of sustainable novel healthcare interventions and treatment plans in Qatar, focusing on the field of precision medicine and personalised healthcare.
We considered adults (18 years or above) Qatari nationals only as part of our study. We did not consider any expat living in Qatar as part of our study. Our study was the first in Qatar considering the largest collection of biomedical measurements such as clinical biomarkers and behavioural factors (sedentary lifestyle) of the CVD cohort. Moreover, we developed AI models based on gender (male, female) as well age-stratified groups (young adults, middle agers, seniors). AI models based on stratified data also showed high accuracy in detecting CVD.
What are the most important outcomes of the study on (AI) algorithms in the detection of CVD across Qatar’s population? What do you think about the accuracy of AI in detecting these illnesses in Qatar as well as the reliability of the findings?
As part of our study, we confirmed the known risk factors for CVD in Qatar and proposed novel risk factors and comorbidities related to CVD in Qatar. The proposed machine learning (ML) (branch of AI) model was highly accurate in the early detection of CVD as well.
Our study confirmed known CVD risk factors (blood pressure, lipid profile, smoking, sedentary life, and diabetes) in Qatar and the Middle East, and identified potential novel risk factors linked to CVD-related comorbidities such as renal disorder (e.g., creatinine, uric acid, homocysteine, albumin), atherosclerosis, hypercoagulable state, and liver function. Our model achieved over 93% accuracy in identifying Qatari CVD from the control group.
The proposed model is reliable, considering age, gender, smoking habit, diabetes status, BP, cholesterol value to calculate the risk score for CVD. Machine Learning models considering traditional known risk factors achieved 82.6% accuracy. On the other hand, after the inclusion of the novel clinical measurements proposed in our study, AI model achieved 93% accuracy. This clearly indicates the superiority of the proposed AI model as well as the importance of integrating novel measurements in clinical setup for CVD diagnosis plan. As a next step, we are working on the validation of the proposed model on a larger cohort and further improvement of the model.
AI presents next-generation potential in the healthcare sector. What are some of the challenges associated with developing the AI model in Qatar?
It is quite conceivable that AI would be the state-of-the-art tool for the next generation healthcare system. But there are many challenges in the development of AI model for predicting early onset of disease; and detecting proper treatment plans and delivering cost-effective healthcare for all the nationals and residents in Qatar. And it requires an immense effort from all the stakeholders in the healthcare sector.
To develop AI models for healthcare, hospitals and all medical institutes need to gather and store data digitally and the data quality needs to be maintained. This requires a huge amount of effort from all personnel in the healthcare industry.
Interoperability of healthcare data within and across the healthcare provides is a key challenge in developing AI models in Qatar. Though Qatar has implemented a single electronic health record (EHR) to minimise the effort for interoperability, this may risk the monopoly for a vendor.
Moreover, interoperability of healthcare data with other government data and external data sources is still a challenge. AI models are data driven, there is always a need for the repeated reuse of data to develop and update AI models to provide better decision-making and patient care. But the participants/patients are requested to provide their consent to use their data in a specific project, which bars the usage of the data in other projects.
To overcome this challenge, a new paradigm of consent called “broad consent” has emerged which allows the participants/patients to provide their consent for their data to be used in a wider range of projects. Interestingly, Qatar Genome Programme (QGP) and Qatar Biobank (QBB) are adopting this concept to leverage better usage of healthcare data.
From a technical point of view, there exist other challenges. Majority of the healthcare data is imbalanced which means there are few instances for the cases (disease) and more instances are normal (free from disease). This type of dataset makes the AI model development more challenging. Developing an explainable AI model is one of the major challenges in healthcare.
Recently deep learning-based AI techniques have performed well in multiple healthcare sectors, including radiological imaging, but the explanation of the model is still a challenging task and needs to be improved to be fully functional in clinical setup. Catastrophic forgetting is a phenomenon for AI based models that forget the previously learned knowledge upon learning new information from recent dataset. Training new AI models with recent datasets on top of the existing model are computationally exhaustive and may take time as well.
In Qatar, we have all the resources to overcome the challenges. We have world class universities and research centers working continuously and coherently to develop AI capabilities for the healthcare industry. Under the leadership of Prof. Mounir Hamdi (Dean, College of Science and Engineering, HBKU) the College of Science and Engineering (CSE) at HBKU is offering two master programs: (a) Master of Data Science & Engineering, and (b) Master of Data Analytics in Healthcare Management. The world-renowned faculty members are contributing to develop the required AI skill set for the students and nurturing the postdocs to generate next generation scientific resources for Qatar who will take Qatar into a top position in the landscape of AI in healthcare.
How can such an AI model be implemented in clinical decision-making i.e., what is the role of physicians and clinicians to ensure the incorporation of such model in diagnosing and treating CVD diseases? How can AI help health professionals?
This is crucial and challenging to transfer AI-based models from lab environment to clinical setup. To fulfil this step, we are developing an “Action Research Method” based framework to evaluate our AI based models in clinical setup involving both end-users (clinicians) and researchers (computational and clinical background) working together in identifying a proper plan (precise diagnosis plan for cardiology) to implement AI-based intervention and reflect upon the experiences for continuous improvement (“Lessons Learned”), which will support us to improve our work in future. Given the feedback from the clinicians and researchers collaborating on the project, will improve the impact of AI models in clinical setup and for the clinicians. And this is an ongoing process where the model needs to be updated at least once in a year to incorporate new knowledge coming from the recent clinical case studies.
When do you plan to complete the study and who are the major collaborators of the project?
We have completed the first phase of our study in February 2021. Now, we are working on the second phase of the study which is planned to be completed by the end of 2021. In the second phase, we are combining the genetic factors and other phenotypic factors to develop a more robust AI model. The final phase of our project is planned to be completed by the end of 2022. Our major collaborators are Hamad Medical Corporation (HMC), and Qatar Biobank (QBB), Qatar. We are planning to implement the proposed AI model in clinical setup with the support from our collaborators HMC. The feedback from the users (doctors and medical practitioners) will help us to understand the difficulties and overcome the challenges in implementing lab-based model into clinical setup.
Since the AI model is based on Qatar specific risk factors, how can the model be used to tackle CVD diseases in the region?
Risk factor identification for cardiovascular disease is an active research area and many studies exist across the globe in this direction. And most of the studies mentioned diabetes, hypertension, obesity, lipid profile, diet, alcohol consumption, smoking, physical inactivity, as potential risk factors that, collectively, attributed to 86% of the cardiovascular disease. But these studies mainly focused on populations from North America or Europe and large-scale comprehensive studies did not consider the Qatari cohort as part of their analysis. So, it was essential for us to emphasise on the Qatari nationals to identify Qatar specific risk factors for CVD.
As part of our study, we confirmed the known risk factors as well as proposed novel risk factors. As these known risk factors are also known to be prevalent in other Gulf Cooperation Council (GCC) countries, we believe that our proposed model would fit for other GCC countries. But this requires rigorous testing using the data from other GCC countries. Moreover, applicability of AI models developed based on a cohort, in different populations is an ongoing challenge for AI in precision medicine. Health data could be biased considering the background of the participants and the methods adopted to process the dataset. Any AI model trained on a particular group of participants/patients may not work well for other cohorts due to the diversity in race, ethnicity, and other demographics parameters. Moreover, real life clinical setup and workflow may impact the quality of the data which is the key component of any AI based model.
Recently Google Health built an AI based system for diabetic retinopathy screening with high accuracy (over 90% sensitivity and specificity) and tested in real life clinical setup in Thailand. While implementing the AI model in clinical setup, the Google Health team faced a variety of challenges in clinical setup leading to impaired quality of images to have high rejection rate. We are planning for a prospective study emphasising the AI model validation for different population in the clinical environment while considering the users’ (physicians and researchers) feedback to improve the AI model.
What is the role of College of Science & Engineering at HBKU in offering world-class education on the latest advances in big data analytics for healthcare applications?
In 2016, the College of Science and Engineering (CSE) at HBKU introduced the Master of Data Science and Engineering (DSE) programme for adapting to the technological development and changes that have led to the availability of huge amounts of digital data. DSE is an interdisciplinary program where mathematics, computer science, statistics and other scientific fields merge together to extract knowledge for generating insights out of data.
DSE is designed to train students, professionals, researchers, entrepreneurs and others who are interested in leveraging contemporary DSE methods, tools and technologies in multiple domains including healthcare. In 2018, the College of Science and Engineering (CSE), HBKU also introduced the Master of Data Analytics Programme for Health Management (MDA-HM) which was the first of its kind within Qatar and the Gulf region.
The programme has grown over the past few years successfully by a) recruiting and training local and international students, especially healthcare professionals; b) obtaining local, regional and international recognition through research work and grants; c) engaging the local healthcare community through collaborative research efforts, educational training, and community events. The curriculum of these MSc program aims to expose students to the latest advances in the field with a focus on big data analytics in healthcare that includes data collection from traditional and emerging data streams, data aggregation methods, data mining algorithms, predictive computational modeling, visualisation techniques, health care policy and social and ethical implication of data analytics.
Being a global leader for healthcare, how is Qatar’s investment in AI in healthcare and precision medicine compared to other countries in the region?
Qatar is investing heavily in AI. The National Artificial Intelligence Strategy for Qatar stated the importance to “Leverage AI expertise available within the country in strategically important domains for Qatar like oil & gas, transportation, health, and cybersecurity to build competitive advantage in specific use cases that could also generate export revenues in the future”.
Qatar is aiming to leverage AI-based technologies to access the knowledge-based economy. The impact of AI in the Middle East and North Africa (Mena) region would be $320bn by 2030 and 19% of it would be in the public sector including health and education.
In the Mena region, Saudi Arabia, the UAE and Qatar have shown strong commitment towards national development leveraging AI based techniques. By 2030, Saudi Arabia, the UAE and Qatar are expected to observe the impact of AI close to 12%, 14%, 8% of 2030 GDP, respectively. The estimated AI market size in Qatar was $1.5mn in 2018, and it is expected to reach $5.7mn in 2022”.
Qatar is also focusing on “Bench-to-Bedside Research” and this type of translation research work can only be realisable with the support of AI. The synergy between omics and clinical data is the key to this initiative and world class institutes like Qatar Biobank (QBB), Qatar Genome Program (QGP), Sidra Hospital, HBKU have been established to fulfil this goal. Qatar is now ranked 39th in Global Medical Tourism Index ranking. Incorporating the AI technologies in healthcare would uplift the ranking of Qatar and will generate revenue from this sector as well.
In your opinion, how is QF embracing the power of AI to offer solutions that simplify complex tasks through data analysis and machine learning?
QF believes that we are living in an era where advancements in any area – from social to economic progress – must embrace the power of AI to offer solutions that simplify complex tasks through data analysis and machine learning. Through research, development and innovation, QF is supporting the economic growth of Qatar through the application of AI in many fields and industries. For example, QF is enhancing health care through earlier detection, smarter diagnosis, and more tailored treatment decisions. QF is also supporting social progress through efficient governance, and management of resources. Additionally, QF is also emphasising on environmental sustainability “through the proper understanding of natural systems”, the development and implementation of green technology and preservation of cultural heritage through Arabic language technologies.
To implement and integrate AI-based initiatives into the social and economic development of Qatar, QF is focusing on multiple research areas including personalised healthcare, personalised education, Arabic language technologies, social good, cybersecurity. QF is delivering on this by, conducting research that advances AI development, incubating and supporting new startups based on AI technology, supporting a governance and policy framework to effectively deploy AI, training the workforce for an AI future, developing AI technologies in areas such as healthcare and food security, and engaging the global community through AI for social good.
HBKU is committed to actively contribute to achieving the Qatar National Vision 2030 by building and cultivating human capacity through an enriching academic experience and an innovative research ecosystem.
In this context, HBKU has developed a study on the use of AI in detecting cardiovascular disease.
Tanvir Alam, an assistant professor at College of Science and Engineering at HBKU has explained in an interview the use of AI in detecting cardiovascular disease (CVD).
Q. Given that CVD is a primary cause of death among the Qatari population, what is the potential for AI — as well as its subcategories, like machine learning — in providing personalised treatment plans for CVD diseases in Qatar?
A. Complex diseases like cardiovascular disease (CVD) often involve the interplay among demographic profile (i.e., gender, age, etc.), genetic profile, lifestyle, and environmental factors. As a result, integration of such diverse factors requires attention to cover the heterogeneity of the population as well. AI-based approaches for CVD rely on multimodal datasets that can excel the discovery of personalised treatment plans by identifying complex relationships among these factors. It is also demonstrated in literature that combining such a wide variety of clinical and genetic data can improve the early detection of CVD.
As part of our study, we integrated a multimodal dataset from Qatar Biobank reflecting the health status of Qatari nationals and we developed a highly accurate AI model for CVD detection and risk factor prediction. We think there is huge potential for such AI models in clinical setup. It will not only leverage the early detection of CVD in an accurate manner by avoiding redundant diagnosis but also support the medical practitioners in their decision-making and reducing their workload.
What is the primary data you are using in this study and how sufficient is this data? Do you have any specific targets groups such as Qataris, expats from certain countries or regions to explore certain traits of the diseases?
We mainly relied upon the clinical data and demographic information from Qatar Biobank (QBB) to enable evidence-based research toward the development of sustainable novel healthcare interventions and treatment plans in Qatar, focusing on the field of precision medicine and personalised healthcare.
We considered adults (18 years or above) Qatari nationals only as part of our study. We did not consider any expat living in Qatar as part of our study. Our study was the first in Qatar considering the largest collection of biomedical measurements such as clinical biomarkers and behavioural factors (sedentary lifestyle) of the CVD cohort. Moreover, we developed AI models based on gender (male, female) as well age-stratified groups (young adults, middle agers, seniors). AI models based on stratified data also showed high accuracy in detecting CVD.
What are the most important outcomes of the study on (AI) algorithms in the detection of CVD across Qatar’s population? What do you think about the accuracy of AI in detecting these illnesses in Qatar as well as the reliability of the findings?
As part of our study, we confirmed the known risk factors for CVD in Qatar and proposed novel risk factors and comorbidities related to CVD in Qatar. The proposed machine learning (ML) (branch of AI) model was highly accurate in the early detection of CVD as well.
Our study confirmed known CVD risk factors (blood pressure, lipid profile, smoking, sedentary life, and diabetes) in Qatar and the Middle East, and identified potential novel risk factors linked to CVD-related comorbidities such as renal disorder (e.g., creatinine, uric acid, homocysteine, albumin), atherosclerosis, hypercoagulable state, and liver function. Our model achieved over 93% accuracy in identifying Qatari CVD from the control group.
The proposed model is reliable, considering age, gender, smoking habit, diabetes status, BP, cholesterol value to calculate the risk score for CVD. Machine Learning models considering traditional known risk factors achieved 82.6% accuracy. On the other hand, after the inclusion of the novel clinical measurements proposed in our study, AI model achieved 93% accuracy. This clearly indicates the superiority of the proposed AI model as well as the importance of integrating novel measurements in clinical setup for CVD diagnosis plan. As a next step, we are working on the validation of the proposed model on a larger cohort and further improvement of the model.
AI presents next-generation potential in the healthcare sector. What are some of the challenges associated with developing the AI model in Qatar?
It is quite conceivable that AI would be the state-of-the-art tool for the next generation healthcare system. But there are many challenges in the development of AI model for predicting early onset of disease; and detecting proper treatment plans and delivering cost-effective healthcare for all the nationals and residents in Qatar. And it requires an immense effort from all the stakeholders in the healthcare sector.
To develop AI models for healthcare, hospitals and all medical institutes need to gather and store data digitally and the data quality needs to be maintained. This requires a huge amount of effort from all personnel in the healthcare industry.
Interoperability of healthcare data within and across the healthcare provides is a key challenge in developing AI models in Qatar. Though Qatar has implemented a single electronic health record (EHR) to minimise the effort for interoperability, this may risk the monopoly for a vendor.
Moreover, interoperability of healthcare data with other government data and external data sources is still a challenge. AI models are data driven, there is always a need for the repeated reuse of data to develop and update AI models to provide better decision-making and patient care. But the participants/patients are requested to provide their consent to use their data in a specific project, which bars the usage of the data in other projects.
To overcome this challenge, a new paradigm of consent called “broad consent” has emerged which allows the participants/patients to provide their consent for their data to be used in a wider range of projects. Interestingly, Qatar Genome Programme (QGP) and Qatar Biobank (QBB) are adopting this concept to leverage better usage of healthcare data.
From a technical point of view, there exist other challenges. Majority of the healthcare data is imbalanced which means there are few instances for the cases (disease) and more instances are normal (free from disease). This type of dataset makes the AI model development more challenging. Developing an explainable AI model is one of the major challenges in healthcare.
Recently deep learning-based AI techniques have performed well in multiple healthcare sectors, including radiological imaging, but the explanation of the model is still a challenging task and needs to be improved to be fully functional in clinical setup. Catastrophic forgetting is a phenomenon for AI based models that forget the previously learned knowledge upon learning new information from recent dataset. Training new AI models with recent datasets on top of the existing model are computationally exhaustive and may take time as well.
In Qatar, we have all the resources to overcome the challenges. We have world class universities and research centers working continuously and coherently to develop AI capabilities for the healthcare industry. Under the leadership of Prof. Mounir Hamdi (Dean, College of Science and Engineering, HBKU) the College of Science and Engineering (CSE) at HBKU is offering two master programs: (a) Master of Data Science & Engineering, and (b) Master of Data Analytics in Healthcare Management. The world-renowned faculty members are contributing to develop the required AI skill set for the students and nurturing the postdocs to generate next generation scientific resources for Qatar who will take Qatar into a top position in the landscape of AI in healthcare.
How can such an AI model be implemented in clinical decision-making i.e., what is the role of physicians and clinicians to ensure the incorporation of such model in diagnosing and treating CVD diseases? How can AI help health professionals?
This is crucial and challenging to transfer AI-based models from lab environment to clinical setup. To fulfil this step, we are developing an “Action Research Method” based framework to evaluate our AI based models in clinical setup involving both end-users (clinicians) and researchers (computational and clinical background) working together in identifying a proper plan (precise diagnosis plan for cardiology) to implement AI-based intervention and reflect upon the experiences for continuous improvement (“Lessons Learned”), which will support us to improve our work in future. Given the feedback from the clinicians and researchers collaborating on the project, will improve the impact of AI models in clinical setup and for the clinicians. And this is an ongoing process where the model needs to be updated at least once in a year to incorporate new knowledge coming from the recent clinical case studies.
When do you plan to complete the study and who are the major collaborators of the project?
We have completed the first phase of our study in February 2021. Now, we are working on the second phase of the study which is planned to be completed by the end of 2021. In the second phase, we are combining the genetic factors and other phenotypic factors to develop a more robust AI model. The final phase of our project is planned to be completed by the end of 2022. Our major collaborators are Hamad Medical Corporation (HMC), and Qatar Biobank (QBB), Qatar. We are planning to implement the proposed AI model in clinical setup with the support from our collaborators HMC. The feedback from the users (doctors and medical practitioners) will help us to understand the difficulties and overcome the challenges in implementing lab-based model into clinical setup.
Since the AI model is based on Qatar specific risk factors, how can the model be used to tackle CVD diseases in the region?
Risk factor identification for cardiovascular disease is an active research area and many studies exist across the globe in this direction. And most of the studies mentioned diabetes, hypertension, obesity, lipid profile, diet, alcohol consumption, smoking, physical inactivity, as potential risk factors that, collectively, attributed to 86% of the cardiovascular disease. But these studies mainly focused on populations from North America or Europe and large-scale comprehensive studies did not consider the Qatari cohort as part of their analysis. So, it was essential for us to emphasise on the Qatari nationals to identify Qatar specific risk factors for CVD.
As part of our study, we confirmed the known risk factors as well as proposed novel risk factors. As these known risk factors are also known to be prevalent in other Gulf Cooperation Council (GCC) countries, we believe that our proposed model would fit for other GCC countries. But this requires rigorous testing using the data from other GCC countries. Moreover, applicability of AI models developed based on a cohort, in different populations is an ongoing challenge for AI in precision medicine. Health data could be biased considering the background of the participants and the methods adopted to process the dataset. Any AI model trained on a particular group of participants/patients may not work well for other cohorts due to the diversity in race, ethnicity, and other demographics parameters. Moreover, real life clinical setup and workflow may impact the quality of the data which is the key component of any AI based model.
Recently Google Health built an AI based system for diabetic retinopathy screening with high accuracy (over 90% sensitivity and specificity) and tested in real life clinical setup in Thailand. While implementing the AI model in clinical setup, the Google Health team faced a variety of challenges in clinical setup leading to impaired quality of images to have high rejection rate. We are planning for a prospective study emphasising the AI model validation for different population in the clinical environment while considering the users’ (physicians and researchers) feedback to improve the AI model.
What is the role of College of Science & Engineering at HBKU in offering world-class education on the latest advances in big data analytics for healthcare applications?
In 2016, the College of Science and Engineering (CSE) at HBKU introduced the Master of Data Science and Engineering (DSE) programme for adapting to the technological development and changes that have led to the availability of huge amounts of digital data. DSE is an interdisciplinary program where mathematics, computer science, statistics and other scientific fields merge together to extract knowledge for generating insights out of data.
DSE is designed to train students, professionals, researchers, entrepreneurs and others who are interested in leveraging contemporary DSE methods, tools and technologies in multiple domains including healthcare. In 2018, the College of Science and Engineering (CSE), HBKU also introduced the Master of Data Analytics Programme for Health Management (MDA-HM) which was the first of its kind within Qatar and the Gulf region.
The programme has grown over the past few years successfully by a) recruiting and training local and international students, especially healthcare professionals; b) obtaining local, regional and international recognition through research work and grants; c) engaging the local healthcare community through collaborative research efforts, educational training, and community events. The curriculum of these MSc program aims to expose students to the latest advances in the field with a focus on big data analytics in healthcare that includes data collection from traditional and emerging data streams, data aggregation methods, data mining algorithms, predictive computational modeling, visualisation techniques, health care policy and social and ethical implication of data analytics.
Being a global leader for healthcare, how is Qatar’s investment in AI in healthcare and precision medicine compared to other countries in the region?
Qatar is investing heavily in AI. The National Artificial Intelligence Strategy for Qatar stated the importance to “Leverage AI expertise available within the country in strategically important domains for Qatar like oil & gas, transportation, health, and cybersecurity to build competitive advantage in specific use cases that could also generate export revenues in the future”.
Qatar is aiming to leverage AI-based technologies to access the knowledge-based economy. The impact of AI in the Middle East and North Africa (Mena) region would be $320bn by 2030 and 19% of it would be in the public sector including health and education.
In the Mena region, Saudi Arabia, the UAE and Qatar have shown strong commitment towards national development leveraging AI based techniques. By 2030, Saudi Arabia, the UAE and Qatar are expected to observe the impact of AI close to 12%, 14%, 8% of 2030 GDP, respectively. The estimated AI market size in Qatar was $1.5mn in 2018, and it is expected to reach $5.7mn in 2022”.
Qatar is also focusing on “Bench-to-Bedside Research” and this type of translation research work can only be realisable with the support of AI. The synergy between omics and clinical data is the key to this initiative and world class institutes like Qatar Biobank (QBB), Qatar Genome Program (QGP), Sidra Hospital, HBKU have been established to fulfil this goal. Qatar is now ranked 39th in Global Medical Tourism Index ranking. Incorporating the AI technologies in healthcare would uplift the ranking of Qatar and will generate revenue from this sector as well.
In your opinion, how is QF embracing the power of AI to offer solutions that simplify complex tasks through data analysis and machine learning?
QF believes that we are living in an era where advancements in any area – from social to economic progress – must embrace the power of AI to offer solutions that simplify complex tasks through data analysis and machine learning. Through research, development and innovation, QF is supporting the economic growth of Qatar through the application of AI in many fields and industries. For example, QF is enhancing health care through earlier detection, smarter diagnosis, and more tailored treatment decisions. QF is also supporting social progress through efficient governance, and management of resources. Additionally, QF is also emphasising on environmental sustainability “through the proper understanding of natural systems”, the development and implementation of green technology and preservation of cultural heritage through Arabic language technologies.
To implement and integrate AI-based initiatives into the social and economic development of Qatar, QF is focusing on multiple research areas including personalised healthcare, personalised education, Arabic language technologies, social good, cybersecurity. QF is delivering on this by, conducting research that advances AI development, incubating and supporting new startups based on AI technology, supporting a governance and policy framework to effectively deploy AI, training the workforce for an AI future, developing AI technologies in areas such as healthcare and food security, and engaging the global community through AI for social good.
