Transcript:
0:00 in this video we're going to show
0:02 calculating heart disease risk using
0:04 fire power reps and the North52
0:07 decision 3 there are several factors
0:09 that need to be assessed when
0:10 calculating heart disease risk including
0:13 smoking obesity hypertension
0:16 dyslipidemia lifestyle pre-diabetes age
0:20 and gender family history and high HDL
0:24 and mitigation Factor the technology
0:26 used for this demo in addition to the
0:27 North52 decision Suite includes clud
0:30 Microsoft cloud Healthcare the data
0:33 integration tool Cod for datae Azure
0:35 health services and the fire link
0:37 connector for this demo we have set up a
0:40 full fire service using Azure Health
0:43 Data services on the Azure portal we can
0:46 find the fire service in our Resource
0:48 Group by going to the Health Data
0:51 workspace clicking on services and then
0:53 the fire service the fire metadata
0:56 endpoint is shown here and is used when
0:59 connecting to the service to connect the
1:02 fire service to datae we have installed
1:04 the data integration toolkit which
1:06 allows the synchronization and
1:08 connectivity of patient health
1:09 information between the fire server and
1:12 datae let's review some of the
1:14 components used for this demo we open
1:17 the data integration toolkit app in
1:19 dataverse and then navigate to the data
1:22 Roots you can see that the encounter and
1:25 observation tables have been set up with
1:27 virtual data providers this means that
1:29 the the data does not sync to dataverse
1:31 and we'll be talking directly to the
1:33 fire
1:34 service we have used the outof the-box
1:37 data Maps one of the ones we use in this
1:40 demo is for the observation table where
1:43 we store the result of the BMI value in
1:45 the observation quantity
1:51 field we can have a look at the
1:54 fhi heart disease risk calculator
1:57 scenario here uh we've got a app here
2:01 which is using the fhi link uh connector
2:05 so if we look at get patients
2:08 here we're setting a variable here
2:10 patience uh we're using the fhi link and
2:14 we are listing the
2:17 resources from the from the patient
2:20 table then we're putting that into a
2:22 collection and then we're using that
2:24 collection to drive this uh drop- down
2:28 list here but we've selected the
2:31 patient so then we're going to create uh
2:34 some observations they're going to be
2:35 linked to an encounter record so if we
2:38 look at the information behind here we
2:41 can ignore the patient so here first of
2:43 all we're going to create an
2:46 encounter and this is the Json object
2:48 that we send through uh using the
2:51 patient name that's selected in the
2:53 dropdown and then we post that uh using
2:56 the fhi link here so we're posting that
2:59 information the encounter then we're um
3:02 setting our observation so we've got the
3:05 BMI Json here so we're generating that
3:09 taking some values out of the the
3:11 slider and then setting that to post
3:15 that resource then we're doing the blood
3:17 pressure very similar scenario here
3:20 taking some of the
3:21 information that is here's the patient
3:25 here's the encounter uh the observation
3:27 date and then we're using some of the
3:30 values the the slider values here for
3:32 the blood pressure
3:34 values for the systolic and diastolic
3:37 blood pressure so once that's all
3:40 done uh we can go and have a look at our
3:43 app so what we're going to do here we
3:44 click get patients we can see here that
3:46 we've got uh a few different patients so
3:50 I'm just going to choose Katy Grace here
3:52 and then I'm going to click create
3:53 observation and then once that's done we
3:56 can then switch across to
3:58 our V your client just have a look see
4:01 what's comes in here you can see here
4:04 these are the two records that we've
4:07 just created here one for the body BMI
4:10 and also then for the systolic here and
4:12 we can see the values that we took
4:14 directly from the app in here these are
4:18 also linked to the encounter record so
4:22 if we click on the encounter record here
4:24 we can see that this is the information
4:26 that has come through from the encounter
4:29 The Next Step step is that the calculate
4:30 risk button is clicked and the North52
4:33 rules engine gets all the encounter data
4:35 from the fire server and calculates a
4:38 risk using the decision table logic
4:41 there are decision table sheets for each
4:43 factor being assessed as you can see
4:46 here the final risk score and the
4:49 associated messages are calculated and
4:52 returned back to the app if you would
4:54 like to know more about how North52 can
4:56 help with complex rules when using fire
4:59 and Microsoft cloud Healthcare please
5:02 get in touch using the information shown
5:04 on the screen