Lecture 10 BAS 112.pdf

Medical and Health Information BAS 112 Lecture 10 Dr. Mervat Helmy Hussein  Standardization of AI in Healthcare Like any emerging field, there is a lack of regulatory guidance and standards regarding the use of AI in healthcare. The World Health Organiza...

Medical and Health Information BAS 112 Lecture 10 Dr. Mervat Helmy Hussein  Standardization of AI in Healthcare Like any emerging field, there is a lack of regulatory guidance and standards regarding the use of AI in healthcare. The World Health Organization is working on developing a standardized assessment framework for the evaluation of AI-based methods for health, diagnosis or treatment decisions. In order to use AI systems, Standardization should focus on : Methods to measure and to reduce bias: to measure reliability and performance Methods for explaining various kinds of AI techniques Ethical concerns; Data privacy and safety Set of standards for human augmentation 2  Standardization of AI in Healthcare To trust in AI systems: Knowing that the AI system has been developed and operated by skilled/well- trained persons The system offers explanations to be understood by the target audience The system is reliable The system is proven to make unbiased decisions The system is verified and validated according to standardized, recognized software development methods 3  AI applications in Medicine: Predictive Medicine Is a branch of medicine that aims to identify patients at risk due to existence of a disease, allowing for early prevention or treatment of that disease. Detect the features of individual patients at risk, analyze data to predict which treatment will be most effective, and then take an action before the risk occurs Artificial intelligence is used to: assess the readmission of patients to hospitals resulting from that disease detect patients who will suffer adverse effects when taking medications 4  AI applications in Medicine: Participatory Medicine Focuses on the patient, in order to empower people to monitor their health. For example, taking vaccination before occurrence of a certain disease. People can collect, record, and track indicators to describe their health, giving AI the opportunity to understand their health, so, reduce the workload for health teams. Mobile-health: collect data on the symptoms given by individuals combined with other data from different medical sources to obtain information on the disease and the spread of infection. Monitor interaction between infected people, get the location and movements of people to alert neighbors/patients to of the disease and prevent infection 5  AI applications in Medicine: Personalized Medicine It uses an individual’s genetic profile to guide decisions about disease prevention, diagnosis, and treatment. So, help clinicians select the appropriate drug and the appropriate dose to ensure better patient care. By enabling every patient to receive early diagnosis, risk assessment and optimal treatment, it promises to improve care while lowering costs The sensitivity to certain risk factors can be accurately determined, appropriate measures can be taken to reduce the risk of spreading the disease. Artificial Intelligence techniques can contribute to the detection or prediction of diseases, achievement of accurate diagnosis, and optimal treatment. 6  AI applications in Medicine: Preventive Medicine Is based on medical practices that are designed to prevent and avoid disease Through wearable technology, remote monitoring, telemedicine The quality of patient care will be optimized, and more timely response to any situation AI uses huge amounts of data to facilitate diagnosis and increase the ability to execute early initiatives and prevent diseases So, executing successful interventions depends on the knowledge of the causes of the disease, the identification of risk factors, the methods of early detection and treatment 7