Microsoft AI in Healthcare How cloud technology big

Microsoft AI in Healthcare How cloud technology, big data are reshaping medicine and amplifying human capabilities Rebone Segoale: Health Industry Executive

s of Open to point t n e r e f dif Democratiza tion h With w r hav wo o n k u o at y

Digital transformation in the health industry EMPOWER YOUR CARE TEAMS & EMPLOYEES to be more productive & efficient OPTIMIZE YOUR CLINICAL & OPERATIONAL EFFECTIVENESS by using data analytics Creating more personal computing ENGAGE YOUR PATIENTS to improve experiences & outcomes Reinventing productivity and business processes TRANSFORM HEALTH by tapping intelligent devices Building the intelligent cloud platform 3

Real AI applications in Health Unstructured data is the information that resides outside of organized databases such as electronic health records and lab reports. If we become capable of tapping the potential of that data, we could make patient care more efficient and costeffective than ever before. ”

AI applications in healthcare benefit patients, clinicians and the entire health ecosystem

Value Creation | Transform Patient Data into Intelligent Action Business Problem Typically, hospitals have a 510% discrepancy between Budgeted versus Actual patient volumes Solution Patient volume forecasting Data Sources Apps and Web Business Value Enhanced forecasts leading to operational savings and improved patient experiences Action Intelligence Modelling and Validation Data Analysis Information Exploration Econometric Models Data Analysis Time-Series Analysis People Apps Machine Learning Sensors and Devices Data Requirement Modeling Patient Volume History e. g. emergency department visits, surgical cases, outpatient visits, midnight stays history Use multiple models to produce competitive forecasts Decomposition- no external variables Decomposition- with external variable Ensemble forecasts Autoregressive–moving-average (ARMA) models- no external Variable ARIMA model- with external variables Automated Systems Improved Forecasting Potential Short-Term Forecasting Scheduling surgeries Potential Long Term Forecasting Demographics Weekly cycles Industry trends Daily actuals Payor trends

Project Inner. Eye Microsoft Research • Medical Imaging AI to Empower Clinicians

10% increase in new cancer diagnosis Half of those patients will need radiotherapy It takes hours for each patient How can ethical AI help ?

Quantitative tracking of disease progression Volume cc Volume of active tumor 1 2 3 4 5 6 7 8 9 time

MS/Phulukisa • Screen for the 7 diseases which statistically burden SA most like HIV, TB, Diabetes, Hypertension and Heart Disease. • Antenatal screening for pregnant mothers. Screening at home will lead to better outcomes for mom and baby • Having analysed all the variables, a score of between 0 and 100 is given which translates as an output into Green, Orange or Red. • Based on the colour, the client gets triaged to be either seen electively or immediately. Working together we can do more 10

Phulukisa Video https: //m. youtube. com/watch? v=Nyw 8 KZKURVY

THANK YOU