It may be early to propose but with the open ended nature of your question a conversation starter is appropriate. AI and a well connected digital healthcare platform could model the most likely transmission pathway based on human population, activity, exposure risks, baseline health, weather, travel patterns and much more. The true power of AI modeling is the capacity to include inputs or data sets that do not appear to be connected. As AI powers through the causal relationship score it either confirms the linkage and digs deeper or it excludes the linkage and moves on to more data. With the ability to process being limited to the allocation of processing power there can be millions of data points scored and collated into decisively different views of transmission, behavior and human risk metrics.
In the US the Center for Disease Control uses relatively simple modeling to forecast viral risk. The data are rather static and the model very rigid. AI would introduce dynamic modeling that can interpret exponentially more inputs and likely outcomes than the current capabilities.
Digital Health platforms could add beneficial insights into at risk individuals based on the COVID19 infection pattern, should one emerge. Across a platform the patient data would better provide clarity into vaccine creation and testing potential as well as prevention education first targets.
The COVID19 virus is certainly an issue to be taken very seriously. AI would lastly contribute to source mapping, genome analysis, transfer/infection data and more to create transmission "barriers" that help minimize the spread of future viruses.
Again, these are future looking applications of today's AI capabilities. In many ways AI can only help us recover from COVID19 at this point.

Another possibility is in the area of limiting contact with patients and by so doing protecting the physician who is obviously a high risk person in the case of COVID 19. Machine to persons can minimise contact especially where machines can be monitored via CTVs and disinfected before contact is made with the machine by the physician.

facilities built to precision become handy with a view to revolutionise therapy and minimize transmission

One possibility is to use AI to target patients at particularly high risk (such as the immunocompromised and elderly) and send messages advising them what steps to take to care for themselves.. AI is already showing promise as a tool for engaging patients and fostering appropriate health behaviors in other areas, such as cardiac rehab.

I think that the idea of these technologies helping prevent the spread of COVID-19 is interesting, and I believe they can contribute in proactive and reactive ways.

Proactively, they can connect care providers with large audiences or singles to deliver advice, counseling, comfort and (excluding direct-contact-based care) all the other things care providers so effectively do. As needed, these can also initiate supplies and medications deliveries. They can of course facilitate emergency response through contact-alert-dispatch functions. I think even more importantly, these technologies can enable authoritative persons to deliver information that provides a calming effect: giving "real actionable information" instead of panic-enabling "news".

(I find that being constantly blasted by stories of deaths, rapid spread, exhaustion of resources and supplies, errors and flaws in the care delivery system do nothing to empower people or halt the spread of anything except worry and panic.)

Reactively, these technologies can provide data that can be used for planning and response purposes. By capturing various points (call volumes and areas, types of calls, resultant actions, etc.). By collating and correlating the data, patterns of cause-and-effect and other trends can be identified, interpreted and employed for future events and response strategies.