|Last year, tech giant Google unveiled a new artificial intelligence that could predict if and when a patient is going to die. Results showed that it has higher accuracy compared to hospital warning systems / Photo by: maglara via 123RF|
Last year, tech giant Google unveiled a new artificial intelligence that could predict if and when a patient is going to die. Results showed that it has higher accuracy compared to hospital warning systems. The AI algorithms can quickly access a patient’s medical records to look for significant data that could indicate the likelihood of their survival.
The study published in the Nature journal revealed that Google’s AI can also make predictions about death, discharge, and readmission. “These models outperformed traditional, clinically-used predictive models in all cases. We believe that this approach can be used to create accurate and scalable predictions for a variety of clinical scenarios,” the authors said.
Also, recently, researchers from the Geisinger Health System in Pennsylvania used AI to help identify patients most likely to die of any medical cause within a year. According to Gadgets 360, India's biggest technology news website, they analyzed the results of 1.77 million ECGs and other records from almost 400,000 patients.
These studies only show how AI can do wonders in the healthcare industry.
Predicting Premature Deaths
AI in healthcare is fundamentally changing medicine through accessing and analyzing vast data sets of potentially life-saving information. This includes speed of care gathered across millions of patients, survival rates, treatment methods and their outcomes, and more.
Recently, researchers found a way to make AI predict premature deaths. Premature deaths refer to the deaths that occur before the average age of death in a certain population. In the US, the average age of death is about 75 years. A 2018 study reported that out of more than 6,000 premature deaths, 33.9% was caused by Ischaemic heart disease (IHD), followed by stroke (14.0%), road injuries (4.7%), stomach cancer (4.6%), and esophageal cancer (4.6%).
Unfortunately, it’s sometimes difficult to predict the likelihood of a patient’s survival due to inefficient hospital processes or even the mishandling of data. AI and machine learning can address this issue, a capability that could revolutionize preventative healthcare. A new study conducted by researchers from the University of Nottingham developed an AI system to predict the risk of early death due to chronic disease in a large middle-aged population. According to Forbes, a global media company focusing on business, investing, technology, entrepreneurship, leadership, and lifestyle, the team used health data collected from people recruited to the UK Biobank, a major national resource for health research in the UK, between 2006 and 2016.
The study revealed that the AI system performed better compared to standard prediction models used by medical professionals, thanks to machine learning algorithms. About 500,000 people who provided blood, urine and saliva samples for future analysis, detailed information about themselves, and even agreed to have their health followed participated in the study. Dr. Stephen Weng, assistant professor of Epidemiology and Data Science at the University of Nottingham, stated that they have advanced the field of AI with their new study on mortality prediction. “We have taken a major step forward in this field by developing a unique and holistic approach to predicting a person’s risk of premature death, by machine learning,” he said.
The researchers used three models such as “random forest,” “deep learning,” and “cox regression.” According to Live Science, a science news website that features groundbreaking developments, these models were used to determine factors such as age, gender, smoking history, and a prior cancer diagnosis to assess the likelihood of a patient’s early death.
The random forest model focused on waist circumference, body fat percentage, skin tone, and the amount of fruit and vegetables that people ate. Meanwhile, the deep-learning model focused on the use of certain medications, alcohol intake, and exposure to job-related hazards and air pollution. Lastly, the Cox model leaned heavily on ethnicity and physical activity. Out of all these models, the deep-learning algorithm delivered the most accurate predictions. It correctly predicted 76% of premature deaths, followed by the random forest model with 64% and the Cox model with 44%.
According to Science Daily, an American website that aggregates press releases about science, this study was built on previous work by the Nottingham team. It showed that four different AI algorithms such as 'neural networks’, 'gradient boosting', 'logistic regression', and 'random forest' were significantly better at predicting cardiovascular disease compared to an established algorithm used in current cardiology guidelines.
|Recently, researchers found a way to make AI predict premature deaths. Premature deaths refer to the deaths that occur before the average age of death in a certain population / Photo by: 9nong via 123RF|
Predicting premature deaths through an AI system not only helps in determining the likelihood of a person’s survival but is also a great potential tool for preventative medicine. Dr. Stephen Wong, assistant professor of Epidemiology and Data Science, stated that preventative healthcare is a growing priority in terms of fighting serious diseases. Today, medical professionals have been working to improve the accuracy of computerized health risk assessment in the general population.
"We have taken a major step forward in this field by developing a unique and holistic approach to predicting a person's risk of premature death by machine-learning. This uses computers to build new risk prediction models that take into account a wide range of demographic, biometric, clinical and lifestyle factors for each assessed, even their dietary consumption of fruit, vegetables and meat per day,” Wong said.
Indeed, AI has great potential to revolutionize complex processes in the healthcare industry. This means saving more lives at an even faster rate.