An Intelligent e-clinic for Computer-aided Diagnosis and Prognosis
محل انتشار: اولین کنگره بین المللی هوش مصنوعی در علوم پزشکی
سال انتشار: 1402
نوع سند: مقاله کنفرانسی
زبان: انگلیسی
مشاهده: 127
نسخه کامل این مقاله ارائه نشده است و در دسترس نمی باشد
- صدور گواهی نمایه سازی
- من نویسنده این مقاله هستم
استخراج به نرم افزارهای پژوهشی:
شناسه ملی سند علمی:
AIMS01_387
تاریخ نمایه سازی: 1 مرداد 1402
چکیده مقاله:
Background and aims: Digitalization influences every individual’s everyday private and professionalactivities. Integrating health information technologies (HITs) significantly improvesdiagnostics and therapy. The application of machine learning and biostatistical methods could improvepublic health. We hypothesized and implemented computer-aided diagnosis and prognosismethods for e-clinics.Method: Cardiovascular diseases, cancer, diabetes, chronic kidney diseases, and Alzheimer’sdiseases remain the top Years of life lost from mortality (YLLs) and deaths attributed risk factorsin the Iranian population by ۲۰۴۰, based on the Global Burden of Diseases (GBD) Foresight Visualization.Over the last ۱۳ years, we have worked on computer-aided diagnosis and prognosismethods in various medical fields. Some of our preliminary web-based programs and AndroidApps were provided at http://www.prognosis.ir/. They are based on our international publicationson various Cohort studies. One of which, a ۱۰-year non-laboratory-based risk prediction chart developedfor fatal and non-fatal CVD using Cox Proportional Hazard (PH) regression, is presentedin this work.Results: Age, smoking status, Systolic blood pressure (SBP), self-reporting history of diabetes,and waist-to-hip ratio (WHR) were used as predictors. The implemented model showed acceptablediscrimination and proper calibration. For a ۵۲-year-old smoking diabetic female with aWHR of ۱.۱ and an SBP of ۱۶۰ mmHg, the estimated CVD risk is ۱۹%. For a person of the sameage and sex but without the risk factors, the risk is ۲%, while for the risk factors related to thegeneral Iranian population, it is ۳% (Figure ۱).Conclusion: The digital clinic offers a unique design to improve treatment outcomes and publichealth. We are now working on a broad e-clinic implementation of the diseases with high burdenin the Iranian population.
کلیدواژه ها:
نویسندگان
Hamid R Marateb
University of Isfahan, Iran
Amin Shafiei Zadeh
AI-tec, Isfahan, Iran
Marjan Mansourian
Isfahan University of Medical Sciences, Iran