With the speed of medical science development today, doctors are increasingly overloaded with the constant release of new information.

The use of the CRISPR-Cas system is an example of this speed. Before its discovery, it would take years for scientists’ to generate and study particular disorders, a process that could span a large part of their careers. The CRISPR system can now allow junior scientists to accomplish the same goal in months.

This speed will translate to dozens of available treatment options and hundreds of active clinical trials with millions of scientific papers published each year. Each method potentially represents a more effective treatment method, with supporting information so that doctors can understand the science behind the treatment options they are considering.

The wealth of information being generated has become impossible for a single person to sift through completely.

Patients often have unrealistic expectations of doctors

As medical professionals, patients’ expect and rely on doctors to be a repository of medical information. However, this is unrealistic as it is impossible for anyone to keep abreast with the amount or to recall that much information accurately. As pointed out in a 2012 report by the Institute of Medicine, this resulting gap in knowledge causes variation of care with suboptimal outcomes.

The use of search engines have helped automate and narrow the search for specific information, yet even this will be insufficient one day. Additionally, the over reliance on search engines also means that users tend to rely on the first few hits they obtain, disregarding results that appear later down the list which may be relevant to their case.

Thus, it is increasingly clear that a better system will need to be created to aid doctors.

Specialised teams to sort through the information

Adapting an idea from the legal profession has been suggested such as the creation of “para-medicals,” a profession that will mirror the paralegal system. Doctors will be able rely on para-medicals, which would comprise of specialised staff trained in researching and identifying clinical trials and therapy options. This will allow doctors to work closely with their patients and determine the best strategy to approach a medical issue based on the research of these para-medicals.

A similar idea is available through a service known as UpToDate, which allows subscribers to access medical synopsis prepared by the company’s 6,300 specialist employees who review medical information. This method has seen improvements in patient outcome according to a 2012 Harvard study.

Standardising medical treatments to ease knowledge overload

As scientific discovery has eliminated the need for educated guesswork, the use of such teams can also lead to the standardisation of the diagnostic and treatment process.

This is currently being conducted at Intermountain Healthcare, which is a network of 25 hospitals. At Intermountain, standard protocols are generated by a small team of paid clinical experts who review scientific literature and combine it with Intermountain’s own experience.

The protocols are defined through symptoms, observations or laboratory results and include the expected timeframe for diagnosis, treatment and recovery. This is translated into standardised yes/no check boxes with tests and treatment options on the protocols.

However the eventual challenges to these manual methods will be scale, consistency and quality. The evolution of advanced analytics is intended to supplement these short falls and aid doctors in real time.

Software designed to play Go may change the face of healthcare

In 2015, DeepMind’s deep learning software AlphaGo conquered the world’s top Go player, Lee Sedol. Compared to chess where scenarios can be thoroughly calculated in three to four moves ahead, the possible moves for Go are said to be incalculable.

The facet that makes AlphaGo so significant is that the software was able to train itself to a degree impossible for any human being. It challenged various versions of itself to develop winning strategies and when Lee competed with the machine, it developed a completely new strategy, stunning observers and besting Lee four games to one.

The numerous possibilities for AlphaGo in healthcare could see it detecting infectious diseases, developing treatment options and even predicting the spread of cancer within a patient. Applying AlphaGo in healthcare could see a more advanced interpretive and learning system than any analytical system being used to date.

Creating a knowledge ecosystem

For all its promises in healthcare, current technologies meant to aid doctors are still flawed. It is suggested that for these systems to be effective, it needs to connect to the broader community of knowledge from doctors, researchers, nurses, pharmacists and caregivers.

Project Diabetes Obesity Control is an example of such a system utilising digital connectivity and traditional primary care to monitor patients’ wellbeing.

“It’s about engaging all stakeholders to develop a holistic solution to improve chronic disease management and motivate healthier behaviour for improved health outcomes," said Dr. Lynda Chin, University of Texas System’s associate vice chancellor for health transformation and chief innovation officer for health affairs. MIMS

Read more:
Is AI the solution to decipher 2.5 million scientific papers annually?
The battle of doctors vs. machines in clinical diagnoses
AI known as CRyPTIC can diagnose drug-resistant tuberculosis in minutes
The artificially-intelligent doctor is still a dream – for now