Cognitive Design in the Medical Field

The field of medicine is something that will most likely always require physical humans, whether it be in performing difficult surgeries or interacting with patients in a comforting manner. With the inclusion of AI though, the medical field could be greatly streamlined.

In a normal hospital patients day, they interact with nurses and doctors many times, draining the time of doctors who have more difficult work to do that requires their expertise. Obviously some sort of comforting interaction is necessary for a patient in a hospital, and check-ups must be performed on a regular basis, but these roles may both be filled by AI.

As long as there is data from the outside world, a machine is able to react to it. Rather than having a nurse check EKG’s all day, an AI can easily record the data, check if the vitals of the patient (heart rate, blood pressure, etc.) are normal, and quickly react to any abnormalities that may be present by contacting a medical professional, i.e. the standby doctor. The AI can also be used to detect when a patient is running low on materials, like an intravenous drip, medication, or towels.

With the use of natural language processing, a medical assistant AI can react to verbal cues from the patient. The AI can easily react to phrases like, “I’m hungry,” or “Think I’ll go to sleep” with the appropriate reactions, calling a nurse for food or dimming the lights

This solves the problem of regular diagnoses, but the issue of regular comforting interaction remains. Fortunately, AI can also be used in this regard. The medical assistant can serve almost as a toy, where the patient can ask questions and make requests and the assistant will respond with reactions appropriate to the patient. Similar in principle to how Google Home and Amazon Alexa work, but with a more personal touch from the data that the hospital has collected about the patient, like age and location of origin. This data alone won’t serve to create the perfect response right at the beginning, but with time and looking at the patients reactions, like tone when speaking, facial expression when giving a response or asking a question, and patients vitals, the medical assistant will react appropriately. A base model must be used in every case initially, but eventually the medical assistant AI will be able to bring up topics by itself without any prompt for the users, such as current events that the user may be interested, or just general topics that that patient will find interesting. Though this is not a substitution for interaction, it can at least stave off the boredom that is present for many patients in the hospital.

The machine in itself should be a visible, physical device. It should blend in with the hospital room, but should still be visible so as not to produce a sense of unease in the user. This means that the device should be tailored to the general data of the user: things like age, hearing or vision impairments, etc. Children should have something that is highly visible, as a clean, modern look will likely not have much appeal and instead may produce a feeling of coldness or detachment. The clinical, sterile feel of a hospital should be avoided as much as possible in the medical assistant. The device must also be placed in relatively close proximity, so it can properly react to the patients verbal and visual cues.

The aim is for a comfortable, relaxing, and engaging hospital stay.

Related posts:

When the administration of President Donald Trump issued its National Security Strategy in December, the word “trade” was mentioned 45 times and “competition/competitors” 58 times — making both…

It is tempting to say that awakening occurs at significant moments in life. I think it is much more accurate to say that awakening continues throughout life! There is this notion often referred to as…