Data: AI needs lots of it, and lack of data is holding AI back, that is the accepted wisdom, these days. “Guess what,” says Dr. Zain I Khalpey, “If you focus on single point, it is made up of billions of pixels.” We don’t need to wait, the technology exists now, he suggests. It turns out that amount of data you need to develop precision health care is not so great; the key lies with the composites of data, from this you can massively reduce the amount of data to recognize patterns, leading to precision treatment. And that takes us to the tale of Artificial Emotional Intelligence.
Dr. Khalpey is a Surgical Director of Heart Transplant, a Fulbright Scholar and Honoree of the Hunterian Medal for Surgery at the Royal College of Surgeons in London. He is also one of a small number of the world’s surgeons to perform a total artificial heart transplant, successfully.
Dr Khalpey, along with Salim Hariri (Professor in the Department of Electrical and Computer Engineering at the University of Arizona as well as Director of the NSF Center for Cloud and Autonomic Computing Center) are working together, bringing something called aiMei Framework to life.
AI in healthcare: Can AI solve the health tech puzzle of new drug discovery?
R&D is critical in the world of healthcare, but it is also tricky; progress is slow, and further progress is hampered by the way different researchers and drug developers work in silos, data is kept secret, locked away from the rest of the world. Can AI in healthcare come to the rescue?
What is it? aiMei Framework is BPU’s automated assistant. It uses using Artificial Emotional Intelligence, and Natural Language Processing, using BPU’s operating system. The result is Virtual Nurse Assistant and a Digital Patient Assistant.
Dr. Khalpey, explains: “Imagine you’re an athlete going out for a run and your heart skips a beat at three miles and it suddenly fires on the wrong part of the electrical part of your rhythm adrenaline surge. You go into Ventricular Fibrillation – you have sudden cardiac death. Alternatively, imagine the night before your wearable detects a problem with the heart rhythm. The wearable also reports you feel emotionally under the weather. Or, given the insights from the device the night before, you realize now that you’ve been having this for a while? Not only are you now diagnosing a condition, but also you are predicting the outcome.”
At this point, ‘self-awareness’ enters the story. In this particular case, , ‘self-awareness’ refers to one of aiMei’s Emotional Intelligence components. It is able to educate the patient to help understand his or her emotions and physical warnings to make decisions towards better health.
According to BPU, nurses in acute and subacute care in hospitals and skilled nursing facilities perform long, cyclical tasks taking up 35% of their time on charting. For dementia patients, windows of unsupervised care are when falls occur. Delirium affects up to 50% of hospitalized seniors and costs the US over $164 Billion per year and $182 Billion in Europe. Additionally, both heart and dementia patients have life-threatening seconds under distress, which can be fatal.
AI and machine learning needed to improve the healthcare industry
Virtual Nurse Assistant and Digital Patient Assistant, on the other hand, can engage patients in live chatbots that are able to track their daily patterns, remind and track medication, and even assess physical symptoms outside the doctor’s office. Should emergencies occur, the wearable may alert hospitals and their physicians within minutes.
Dr. Hariri states, “Artificial Emotional Intelligence can be utilized to interact with patients 24/7 to understand their emotional state, stress level and can even predict major heart failures before they occur. Artificial Emotional Intelligence technology will enable us to notify doctors ahead of time so they can take proactive actions that will save a life and will lead to significant improvement in the quality of healthcare.”
Dr. Khalpey warned: “The problem with IBM Watson and its misdiagnosis with cancer was the lack of emotion in the data; there was no personalised data in their algorithms. The pathways didn’t talk to each other. The link between artificial intelligence – the predictability of it, harnessing the data in an intelligent way, along with emotions, fuses the concept of artificial emotional Intelligence and artificial intelligence.”