Medical Technologies for Equitable A.I. | Mercy Asiedu скачать в хорошем качестве

Medical Technologies for Equitable A.I. | Mercy Asiedu

Abstract ## Applications of artificial intelligence (AI) in healthcare have accelerated exponentially and while AI has the potential to drastically improve healthcare for the better, like a lot of advancements in health technology, it also has the potential to increase health disparities. Historically, healthcare technology has not been created with the needs of underserved settings in mind and this has led to a lack of affordable access to quality care, especially in low- and middle- income countries, and in underserved regions in high-income countries. Further, reduced access to these healthcare tools leads to downstream gaps in data and the potential for machine learning biases. In this talk I will present some prospects and challenges for creating tools to enable health access. I will focus on (1) the development of a device and algorithms for cervical cancer screening, (2) algorithms to improve imaging for low-powered ultrasound devices, and finally, (3) the early phases of a smartphone-based approach for patient-centered health record keeping. I will outline the successes, as well as the limitations of these tools, and their ability to address healthcare, and general wellness, in the immediate future. Overall, I hope to convey a wholistic approach for developing equitable devices and AI solutions and how the pipeline can be improved to ensure inclusive data representation and impact. I will conclude on an optimistic look to the future of how AI can be a tool for global good. Bio ## Mercy Asiedu (https://www.mercyasiedu.com/) I am currently a Research Scientist in Responsible AI at Google Research where my interest lies in democratizing AI in health. I was previously a Schmidt Science Postdoctoral Research Fellow at the MIT Jameel Clinic for AI & Healthcare where my research focused on machine learning applications to medical imaging and clinic visit notes. I was advised by Prof. David Sontag, MIT Clinical Machine Learning Lab, Dr. Anthony Samir in the Center for Ultrasound Research and Translation. Prior to my postdoc, I received my Ph.D. in Biomedical Engineering with a certificate in Global Health from Duke University. I obtained my BSc. in Biomedical Engineering with a minor in Business from the University of Rochester through the Zawadi Africa Education Fund, a full-tuition, and expenses scholarship for African women. I am a co-founder and the CTO of the Calla Health Foundation, a startup working to commercialize tools for cervical cancer screening. I am also a co-founder and co-CEO of GAPhealth Technologies, a startup that provides healthcare software for medical data storage, telemedicine, and care management. I am motivated by my lived experiences, growing up in Ghana, and witnessing the challenges that the lack of accessible, technologies in healthcare, posed to mortality outcomes of friends and family members. Having lived in the United States for 12+ years, I have further come to recognize the effects of inequality in healthcare within high-income countries. My long-term research and career interests lie in developing socio-cultural technological solutions to bridge disparities in healthcare.