By Mark Watts
Only two things are certain, death and taxes. This article is written with a heavy heart. My brother called me to let me know my 42-year-old niece was told she may not make it through the weekend. I jumped on a plane and flew to their home. She was diagnosed with breast cancer at age 30, treatment and surgery and a cancer-free declaration. Chemo had leached her bone material. She suffered multiple spine fractures causing pain that required stabilization rods to be placed in her back.
Her six-month follow up checkup scans went from nothing noted to Stage 4 cancer in six months.
In 2016 researchers at Stanford University were able to train an AI system to accurately predict if a person would die in 3 to 12 months. This remarkable application is representative of the wide variety of roles that AI and deep learning will play in the upcoming years.
Palliative care usually involves providing relief for a patient’s pain, stress and other symptoms when the patient has a terminal diagnosis and no cure is expected. Knowing when such specific care is warranted may have beneficial effects for the patient, family and caregivers – and help determine when such care would be most effective. To create the Death Algorithm, the Stanford team used information from about 170,000 patients who had died from cancer, heart and neurological diseases. Various information from medical records – including a patient’s diagnosis, medical procedure, imaging scans codes and drugs prescribed were used to teach the AI system. Then, a deep neural net was trained, with various internal weights adjusted for the neuron units. The deep neural net made use of input layers of 13,654 dimensions (codes for diagnoses and drugs).
In the end, 9 out of 10 people predicted to die within 3 to 12 months did die within this timeframe. Also, 95 percent of those whom the algorithm determined would outlast 12 months did live longer lives.
When I arrived at my brother’s home, I was greeted by a somber houseful. I chatted with my niece who was alert and able to ambulate. I reviewed her brain MRI results that documented 12 separate cancerous masses. She said her doctor had arranged hospice care. We spoke about how the use of hospice had changed over time and that this designation was just as much about “living well with cancer” in a pain free manner, with access to resources like home care.
She looked at me and said, “So, I could go to the beach?”
Lack of medical knowledge and modern health care practices can rob patients of the opportunity to bucket list a couple of memories. I know that I would want to know if I was living my last 12 months.
I can tell you that we enjoyed a flight of six flavors of Tillamook ice cream and more than one Ruby Red microbrew beer at the beach the next day. Life is short, hug the ones you love.
Mark Watts is an experienced imaging professional who founded an AI company called Zenlike.ai.
