Nothing like a cup of python in the morning. This week I dipped my toes back into the machine learning pool. My first program involved a breast cancer dataset created by Dr. William H. Wolberg from the University of Wisconsin. The dataset took fluid samples from patients with breast masses and performed analysis on the digital scans using a graphical computer program. The program used curve-fitting algorithms to compute important features and extrapolated that into up to 30 statistical data points. The whole process of collecting and generating the data is its own black box of mystery. For this week, my job was to load the dataset using pandas and apply various machine learning algorithms to be able to classify whether each sample was benign or malignant! Last year, I did not get to spend much time with the scikit-learn library so it was fun to travel down the API rabbit hole and get to see how many useful tools are available!
I had the opportunity to use 7 different classifiers. Some I had heard of before and some I will be looking to research further in week3.
Diabetes and obesity were front and center during the two research presentations we got to listen to. Postdoctoral research student Souptik Barua and Dr. Jessica Ramella-Roman from FIU painted a familiar portrait. I have several family members who have battled with the health disparities pictured above and I applaud PATHS-UP mission to tackle these problems head on and seek to educate the next generation of scientists about these problems. Dr. Ramella-Roman spoke to us about a photoplethysmography(or PPG) study that was done to measure the blood pressure and arterial health of study participants. What the study concluded was that traditional devices used to measure blood pressure suffered from some fundamental flaws. They were not designed with a diverse population in mind. In fact, there was significant loss in signal capture in participants with darker skin tones and who suffer from obesity.
All of this got me thinking about how important it is for next generation of scientists and engineers to be mindful of the devices and technology they develop to help close the health disparity gap with underserved populations. I am motivated to share these problems with my students to help motivate them to see the value that technology has in solving some of the most complex problems our society faces. Till next week…