top of page
hand-businesswoman-touching-hand-artificial-intelligence-meaning-technology-connection-go-

Using Line and Stacked Column Chart in Power BI

One can create many different types of visualizations in Power BI like Stacked Bar Charts, Stacked Column charts, Pie charts, donut charts, funnel charts, TreeMaps and many more.


One of the most commonly used charts that can be created in Power BI which is easy and presents a visual that is readily understood is the Line and Stacked Column Chart. This chart can be used to visualize trends of many different measures against a given set of parameters. In the example below I am showing how we can visualize Diabetic Retinopathy (DR) and Diabetic Macular Edema (DME) in cognitively impaired diabetic patients.


Cerebral visual impairment (sometimes called cortical visual impairment or CVI) is a disorder caused by damage to the parts of the brain that process vision. This is caused when high blood sugar damages blood vessels in the retina (a light-sensitive layer of cells in the back of the eye).

This causes Diabetic Retinopathy or DR.


Diabetic macular edema, or DME, manifests as retinal thickening caused by fluid accumulation, mainly in the inner and outer layers.


Macular edema is strongly correlated with diabetic retinopathy severity. As the severity of Diabetic Retinopathy increases so does the severity of Diabetic Macular Edema.


I selected 3 patients to do a detailed analysis based on data that made them unique. The first patient is classified as mildly impaired in both visits, and the second patient is classified as mildly impaired in visit 2 but then cognitively declined in visit 8. Patient 3 cognitively declined in both visits 2 and 8. (Visits 2 and visit 8 are 2 years apart)


All 3 of our patients are in the overweight or obese category. The first patient is from the control group (Diabetes under control) and the remaining 2 patients are from the DM group (have Diabetes Mellitus). All of them had no/mild DR or DME conditions during their first visit, but during their 2nd visit, their DR and DME values were at extreme levels.


Two biomarkers that showed an increase in value from visit 2 to visit 8 were HbA1C and SVCAM. This was common in all 3 of the patients used in this analysis.


If HbA1C is high it damages the blood vessels in the retina. The damaged blood vessels can swell and leak causing blurry vision or even stopping blood flow.


SVCAM stands for Soluble Vascular Cell Adhesion Molecule and is associated with hypertension and Vascular inflammation. Vascular inflammation is nothing but the inflammation of the blood vessels. When the blood vessels get inflamed, they get damaged and reduce the blood flow to the organs.


In the chart below we see that for 3 different patients, we have plotted the average values of biomarker SVCAM, HbA1C, Diabetic Retinopathy, Diabetic Macular Edema, and Average Dementia Score.


As seen, the average value of SVCAM is represented by columns and the rest by line charts.


It is important to understand that as we are looking at these values for 3 different patients only, the average values here for each patient show the actual values of these biomarkers in the chart during their visits.


The above chart shows that biomarkers SVCAM and HbA1C increased for these 3 patients from visit 2 to visit 8. The DR and DME conditions also worsened during this time. Apart from the Control patient, the other 2 patients in the DM group showed an increase in their cognitive score which tells us that as the DR and DME worsened the number of cognitive tests that they failed also increased, which means that there was cognitive decline in these patients. It also shows that inflammation plays an important role in DR and DME impairment in these patients.


As we can see, the line and stacked column chart helps us understand the variations of many measures on one chart very easily. This can be used to add multiple measures but it will still give you a visualization that is readily understood and is pleasing to the eye too. This chart is a powerful tool available in Power BI and should be used more often.


References:

  1. http://www.kaggle.com

  2. https://www.ncbi.nlm.nih.gov

  3. https://www.diabetes.org

101 views0 comments

Recent Posts

See All
bottom of page