A team of scientists has developed a new decision-making tool that provides personalized treatment recommendations for coronavirus patients. The tool uses artificial intelligence (AI), machine learning (ML), and data analytics to help clinicians treat coronavirus patients.
As COVID-19 (coronavirus) had a devastating impact on people worldwide, researchers are racing to find ways to predict who is most likely to die from COVID-19 and to develop treatments to prevent and treat the virus.
The Special Situations Initiative
The Initiative is a decision support tool integrating data across disparate sources to provide public health decision-makers with insights into how different decisions might be affected by individual, local, regional, and national trends in infection, hospitalization, and mortality.
The tool is based on a simple metric that highlights how sick people are and how quickly they need care. “The Special Situations Initiative” (SSI) tool enables doctors in New York City to quickly assess how sick patients with COVID-19 are and helps them to determine whether hospitalization is needed. The tool also lets medical professionals know how quickly a patient needs care.
Machine learning could help patients diagnosed with COVID-19 be administered the right medicine at the right time by predicting the symptoms a patient may experience before they do. A team of researchers, Data Scientists at IBM Research–Almaden, are developing a decision tool for COVID-19 patients, using machine learning to predict the risks of a patient developing a respiratory infection. By predicting the symptoms patients are more likely to develop, the tool will alert doctors when to administer treatment.
One of the most difficult decisions healthcare providers must make is whether or not a patient suspected of having COVID-19 needs to be hospitalized. Hospitalizing a patient costs thousands of dollars and places a significant burden on the healthcare system, but failure to do so can result in death. Hence, doctors look for each patient’s individual risk factors. The hope is that machine learning decision-making tools would help doctors make these decisions. A recent study examined whether machine learning can improve the decision-making process for guiding the treatment of coronavirus patients.
Decision Support Systems
Although scientists and medical professionals are responding to the spread of COVID-19 with unprecedented speed and efficacy, their work is cut out for them. They must determine both clinical and community-level practices that can safely and effectively treat a large and ever-changing population of patients.
Our healthcare system can overwhelm people with coronavirus (Covid-19). Hospitals need tools to help them figure out the best course of treatment. Current systems are often slow or inaccurate. A new tool allows doctors to see important data points more quickly. Decision support systems are computer programs that help doctors make better decisions. The team has programmed a data scientist, Alexa, to crunch data.
The tool can help healthcare providers understand the impact of specific treatment regimens on patients with the disease based on factors such as age, growth factors, and disease severity.
Data scientists are creating tools to help doctors diagnose coronavirus patients more quickly and decide which patients need to be treated first. Some researchers have developed algorithms that have the capacity to analyze thousands of data points so physicians can look through large amounts of data in a short amount of time.
Doctors and scientists are always seeking ways to treat patients battling COVID-19. They are working to develop new options, and data scientists have now developed a computer model that they hope will go a long way in helping doctors develop the best treatment for each patient. Scientists have developed a decision-making tool to help focus attention on which hospitals were best equipped to treat patients with coronavirus. The tool helped hospitals to notice that, although their settings were similar, they were treating patients with different outcomes.
The algorithm also takes into account the patient’s physiologic response to their treatment. The machine learning model evaluates patient-specific factors, such as the severity of symptoms, their underlying medical conditions, and the treatments that they’ve received in the past.
While other hospital systems have tried similar approaches, this research team was able to create a tool that measures treatment response better than any of those prior approaches. It could guide physicians toward the most effective — and less expensive — treatment for a given patient.
The decision support system can rapidly and accurately identify patterns, abnormalities, and changes in disease progression. For patients who are known to be at higher risk, the tool can guide prescribing physicians to recommend a more aggressive course. It will allow doctors to decide whether patients need ventilator care, standard care, or palliative care.