The Greatest AI Advances In Healthcare And Pharma So Far

• comments
• 
• 

Artificial intelligence is changing the world. Automation and machine learning are just two of the components that are making people's lives easier and this is just the tip of the iceberg. Machines are going to play a very important role in society, with pharma being one of the biggest benefactors.

From improved patient care to drug discovery, there are lots of potential uses for AI in pharma. One of the biggest reasons the industry benefits from AI is that processes are quicker and more efficient. In the case of drug discovery, this means effective treatments can be identified and rolled out quickly to save more lives.

While there is still a lot of debate about the potential drawbacks and advantages of AI in pharma, there have been many significant breakthroughs in technology in the present day.

Here are some of the greatest AI advances in pharma so far. 

Medical Imaging

Machine learning processes a lot of data in a fraction of the time it would take humans to process the same amount of information. The algorithms are better at spotting tiny details in medical imaging reports, such as CT scans and mammograms.

Medical software analysis performed by a computer is also more efficient in terms of cost compared to human activity according to Orthogonal. There have even been arguments made by some scientists and researchers that not using AI in fields like radiology would be unethical.

For example, Zebra Medical Vision has created Profound, a platform that uses machine learning algorithms to analyze medical imaging reports. It can identify potential conditions like breast cancer and osteoporosis with more than a 90% accuracy rate, among many others.

The technology uses deep learning to check for hidden symptoms other diseases that the doctor would not have otherwise been looking for. In fact, some of Profound's deep learning networks have earned a 100% accuracy rating for detecting fatal variations of breast cancer.

Drug Development

Drug development is a very costly business. Not only does it mean spending a lot of money on clinical trials, but it also takes a lot of time to find a new drug. Ultimately, without AI the drug discovery process is inefficient. New pharma products will also require further testing even after the drug has been released as it's entirely possible that unforeseen side effects can be discovered.

For example, according to DAP, AI in healthcare and pharma uses big data to root out new drugs at the molecular level - something humans could never do.

In particular, Atomwise used AI to prevent the Ebola virus epidemic from spreading. It took less than a day to find two existing medicines that could be repurposed and used to treat the disease safely. They also discovered how to react to any new pandemics by scanning through drugs already available to patients. 

Again, due to the speed of this analysis, this means the treatment options are safe to use and can be produced rapidly at scale.

Electronic Medical Records (EMRs)

EMRs have had a huge impact on the healthcare industry. They have improved the quality of care and increased productivity of healthcare staff, cutting admin time down drastically. However, there are some that think EMRs hinder healthcare as they find them difficult to use, resulting in less efficiency.

AI can automate data collection by understanding patient data through deep learning to improve the quality of care. It can also enhance patient adherence and guide medical professionals using analytics programmes for treating life-threatening conditions. For example, the JAMA Networth released a recent study that showed how big data from EMRs can be used by AI to treat C. diff infections.

Clinical Decision Support (CDS)

Deep learning and AI can also help machines make better decisions than humans. These tools are called clinical decision support (CDS). They are usually built into EMRs to suggest better treatment options to doctors, warn them of the potential dangers of using it and also take a detailed look at the patient's health record itself.

For example, MatrixCare uses Cortana (Microsoft's AI) to manage nursing homes. As CEO John Damgaard explained, "One doctor can read a medical journal maybe twice a month," 

"Cortana can read every cancer study published in history before noon and by 3 p.m. is making patient-specific recommendations on care plans and improving outcomes."

CDS machines can also communicate with each other better than humans. Due to the Internet of Things (IoT), many devices can connect to a patient's EMR, including smartphones, smartwatches and other healthcare monitors.

It's so important that medical information is interpreted accurately. If not, it could lead to the wrong treatments being issued to a patient, resulting in more damage to their health and increased chances of hospitalization. In the worst case, this could lead to patient death and the medical practice being sued for malpractice.

By using AI, EMR platforms can communicate with each other automatically, making it easier for doctors, nurses and hospital staff to work with other facilities. At the end of the day, they keep everyone on the same page.

Going Forward

There are many plans for artificial intelligence in healthcare (great in-depth guide from Healthcare Weekly Magazine here). Many companies are creating prototypes that are not quite ready for release or testing with trials, but there are some that have a very promising impact on the industry.

For example, Craig Venter at the Human Genome Project is currently working on a new generation of computational technologies to predict the effects of any genetic alteration. This will advance personalized treatment options and detect preventable diseases early on in their lifecycle. 

AstraZeneca is also partnering up with BenevolentAI to "use machine learning and artificial intelligence to discover potential new drugs for chronic kidney disease and idiopathic pulmonary fibrosis". Joanna Shields, Chief Executive Officer, BenevolentAI, said: "Millions of people today suffer from diseases that have no effective treatment. The future of drug discovery and development lies in bridging the gap between AI, data, and biology".

Finally, Gilead has partnered with startup company insitro to discover and develop therapies for patients with nonalcoholic steatohepatitis (NASH). They are aiming to "create disease models for NASH and discover targets that have an influence on clinical progression and regression of the disease."

These are just a handful of future applications of AI in healthcare and pharma but there are plenty more on the horizon.

Conclusion

These are some of the greatest advances of AI in healthcare and pharma so far. As the months and years go by, more and more companies will use it to save costs, improve patient care and ultimately save more lives.