DXC Labs has been leading the R&D for our industrialized AI offering by rapidly developing prototypes of machine learning solutions for various "data stories." For customers, our prototypes provide visualizations of the predicted results as "actionable insights" in Microsoft's Power BI that can be drilled into for further analysis. The data story was: "Reduce patient care cost and improve patient care and logistics for elective care." Predicting diabetes risk to reduce patient care costs -- visualizations of the actionable insights obtained from otherwise disparate data.
Artificial intelligence (AI), with its capability to draw "intelligent" inferences based on vast amounts of raw data, may hold the solution. Follow the money, and you'll see big bets on healthcare AI across the globe: 63% of healthcare executives worldwide already actively invest in AI technologies, and 74% say they are planning to do so. PwC's Global Artificial Intelligence Study, which analyzed AI's potential impact on each industry, found that healthcare (along with retail and financial services) is poised to reap some of the biggest gains from AI in the form of improved productivity, enhanced product quality, and increased consumption. For example, 94% of survey respondents in Nigeria, 85% in Turkey, 41% in Germany, and 39% in the UK are willing to talk to and interact with a device, platform, or AI-guided robot that can answer health questions, perform tests, make diagnoses based on those tests, and recommend and administer treatment.
Recent applications of machine learning with big data are able to predict diseases--such as Alzheimer's and diabetes--with incredible accuracy, years before the onset of symptoms. To assess the likelihood of a patient developing a certain condition, physicians have traditionally relied on risk calculators such as this one. Bringing together the data collected in many large-scale studies across diverse medical specialties, together with information from our medical records and other sources, doctors can accurately calculate the likelihood of suffering from a disease, a patient's possible outcome, and even figure out what the main predictors for each illness are. The CS experts have brought to the table the capacity to identify, develop, and fine-tune machine learning algorithms and techniques to predict conditions with better accuracy and speed.
Doctors already use MRI scans to look for changes characteristic of Alzheimer's but scientists believe artificial intelligence could help specialists to diagnose the conditions before changes are clearly visible. Researchers believe the new technology could be used by doctors to predict Alzheimer's and other diseases within a decade. Dr La Rocca told New Scientist that AI-based testing would also be cheaper and more comfortable than invasive techniques, which look for "sticky" plaques and tangles of protein in the brain that have been linked to the disease.
In a study, published earlier this month, researchers developed a machine-learning algorithm to detect Alzheimer's in brain scans 86 percent of the time. Even more impressively, it identified changes in the brain that showed mild cognitive impairment (MCI) 84 percent of the time. It might be able to identify these changes even earlier, but the researcher's only tested it on individual's who developed Alzheimer symptoms within nine years.
"We've learned that you cannot make a definite statement about a particular gene," Winston Hide, Professor in computational biology at The University of Sheffield, explains. According to Winston Hide, data reproducibility and unraveling the differential contribution of multiple genes in relevant biological pathways are some of the big issues in this field. According to Barry, the solution might come from imitating the brain's own ability to process information through deep learning. However, can we really make good models of the brain by using deep learning?
The primary focus of these initiatives is on health care providers, helping them develop treatment approaches that are most effective for individual patients. One consortium of hospitals, researchers, and a startup, for example, is conducting "Project Survival" to identify effective biomarkers for pancreatic cancer.3 In other firms, real-world data sources are being used to identify molecules that might be particularly effective (or ineffective) in clinical trials. Another long-term challenge to be addressed by the life sciences and health care industry is collaboration and integration of data. Project Survival, for example--an effort to find a pancreatic cancer biomarker--involves collaboration among a big data drug development startup (Berg Health), an academic medical center (Beth Israel Deaconess in Boston), a nonprofit (Cancer Research and Biostatistics), and a network of oncology clinicians and researchers (the Pancreatic Research Team).
The AI was trained to correctly spot the difference between diseased and healthy brains, before being tested on its accuracy abilities on a second set of 148 scans – 52 of which were healthy, 48 had Alzheimer's and the other 48 had a mild cognitive impairment that was known to develop into Alzheimer's within 10 years. The algorithm correctly distinguished between healthy and diseased brains 86% of the time, according to the researchers, who added that it was also able to spot the difference between a healthy brain and a mild impairment with an 84% accuracy rating. Last month mobile game Sea Hero Quest – which uses navigation challenges to gather data about spatial movement as part of research into the disease – was expanded to virtual reality for the first time. The game sets users navigation challenges, and they can opt-in to share their data with the researchers behind the game, who can use player performance data to plot spatial navigation skills of different ages groups and genders.
One oft-cited solution to the big data challenge of digital mental health data is to use artificial intelligence approaches like deep learning to help make sense of the raw data. Deep learning is the art and science of building enormous computer models--neural networks--that can be used to predict, classify, edit, describe, and create videos, images, and text. Artificial intelligence programs still struggle with cancer diagnoses, even when complete medical records are available and even with medical knowledge of that cancer well characterized at the genetic level. Creating meaningful categories of mental illnesses is complex, making it difficult to create or train diagnostic algorithms.
Machines would look at data, understand, reason over it, and they continue to learn: understand, reason and learn, not program, in my simple definition. There would be two big differences between business and consumer AI. It leads me to the second big difference between consumer and business AI. That gives you a long, long, long answer, but this is why I'm so positive this world will have more really tough problems solved with AI.