Reflection - Part 8: Analytics

From Xiruo:

I’ve been putting much efforts in the theoretical research, so I may ignore many realistic issues when putting my stuff in real clinical settings. This part is definitely very important, but I think for some limited time and efforts, one should focus and think about other part in general… Anyway, the paper, Regulation of predictive analytics in medicine, did provide some insights into this problem and made me think over in which part and how my research will be of some “real” value.

From Yue:

Although the AI applications in health care did not meet the expectation, AI systems like Watson can still have broad applications in medicine. AI or AI-powered robots excel in performing repetitive tasks with defined steps, such as simple routine surgeries of the eye or hair, analysis of X-rays or other scans, checking on patients between office visits and handling administrative billing or claims.

Right now, AI already has broad applications in the manufacturing sector. By taking over tasks that are repetitive, tedious and dangerous, AI leaves humans to do more complex and nuanced problem-solving. This way, AI and humans can work side-by-side to achieve higher efficiency and lower inaccuracy.

From Jake:

For this exercise I focused on text analytics in healthcare. It was neat to explore the use of text analytics. When it comes to text analytics, it appears that the revolution will come in psychiatry and psychology. Just a simple text analytics tool like that can measure so many variables including positive and negative sentiment. We didn't even delve into natural language processing and computational linguistics including distributional semantics. When you take into account the large breadth of text analytics in medicine it's truly fascinating what tools/methods we have out there.

If I were to look past text analysis, there's a lot of traditional data analysis being done. The retrospective cohort analysis on ehr data threatens to apply traditional parametric and non-parametric statistical texting to past ehr data. This is really interesting to me because it gives content experts a cool way to perform analysis. Tools like Nic Dobbins Leaf are super cool to me cause they empower physician-led multidisciplinary teams.