In a quick read, this newsletter highlights the week’s breaking news in Artificial Intelligence, a new laser application, new technology to identify invisible wounds in the brain and a soaring job market for the technology that no one understands but everybody wants, healthcare blockchain.
A study from the University of Nottingham, England, reports that AI in healthcare technology is a critical part in the development of tools that predict risk of early death due to a chronic disease. Led by Assistant Professor of Epidemiology and Data Science, Dr. Stephen Weng, a team of doctors and data scientists have developed and tested computer-based algorithms to build new premature-death-risk prediction models. In earlier tests, the team determined that using four dissimilar AI algorithms (random forest, logistic regression, gradient boosting and neural networks) was superior to predicting cardiovascular disease over earlier established predictions. Nottingham’s computer-based ‘machine learning’ study was performed on a large middle-aged population. Variables used on the test included clinical and lifestyle traits; daily dietary intake of meat, vegetables and fruit; demographic features; and biometric factors for every assessed individual.
The prediction in 2018 that blockchain technology development would be the number one emerging job has come to fruition. Distributed ledger technology is coming out of pilot-mode and into real-world operation. Popular job boards are reporting soaring demand for software engineers with blockchain skills, seeing a ninety percent increase in blockchain, bitcoin and cryptocurrency technology. Blockchain is poised to have a major impact on healthcare information technology, as well as almost every industry. This innovative technology is being compared to the first wave of the internet.
Laser healthcare technology allows analysis of the metabolic behavior of cancer cells using laser light and high-frequency ultrasonic sensing. To determine the specific treatment needed for a cancer patient, the traits of the patient’s cancer cells must be known. Up until now, the process of isolating and analyzing cancer cells has been inefficient and time consuming, delaying possible treatment for the patient. Previous efforts such as PET scans and fluorescent dyes have limited value, and the earlier analysis of cancer cells through oxygen sensors could only be performed on small quantities of cells over an inefficient time frame. Laser technology, PAM (photoacoustic microscopy), improves the oxygen sensor analysis by two orders of magnitude. The laser light generates vibrations in the polyp and produces images of tissue, blood vessels and cells. PAM allows doctors to pigeonhole specific cells and use a laser wavelength that hemoglobin can absorb and convert to sound. The doctor, to determine oxygenation, ‘listens’ to the sound made by the sample when it is illuminated by the laser. The sound, SCM-PAM (single-cell metabolic photoacoustic microscopy), calculates the OCR (oxygen consumption rate) of the cancer cells providing doctors with good quality information to determine cancer prognosis and therapy.
In other healthcare technology news, just after President Trump’s edict ordering government agencies to earmark funds for innovative healthcare AI programs, CMS launched a $1.65 million challenge to develop a comprehensible AI tool. The brief requires the development of an AI program that predicts patients’ healthcare results, adverse incidents and decrease the administrative duties of doctors. Competitors must demonstrate “how neural networks and deep learning will predict unplanned hospital and skilled nursing facility (SNF) admissions and adverse events”. Applications for the one-year $1.65 million Artificial Intelligence Health Outcomes Challenge are due on June 18, 2019. Twenty applicants will be chosen to participate in stage one of the competition that starts on July 19, 2019. By driving predictive analytics, one of the most critical results expected is to ensure that AI is clearly understandable by physicians and staff at the forefront.
In other healthcare technology news, an emerging healthcare technology has been reported. Diffusion Tensor Imaging (DTI) uses a specific approach to diagnose traumatic brain injury (TBI). DTI detects invisible wounds that do not show up on a standard MRI. Rigorous activity, for example military training and service, sports training, playing sports, a child’s energetic play or just a fall can cause mild TBI. Without external evidence, the injured party may feel they are fine and just need rest. However, certain symptoms that persist require medical attention:
- Dizziness, loss of balance
- Depression, headaches
- Poor concentration, memory loss
MRI and CT scans cannot detect a hidden brain injury. DTI is used to observe the brain’s water molecules that flow along the nerve fibers. A distorted or deformed water path represents the invisible wound. DTI technology is being developed as a tool to further understand TBI and concussions on a molecular level.