From search engines to social media, algorithms are quickly becoming a part of everyday life, and companies like Bayer, MIT and others are using them to their advantage.
From search engines to social media, algorithms are quickly becoming a part of everyday life, and companies and academic institutions like Bayer, the Massachusetts Institute of Technology (MIT) and others are using them to their advantage.
In artificial intelligence (AI) and machine learning (ML), algorithms are used to solve complex problems - including in preventative healthcare, diagnostics and drug discovery. Continue reading to learn how these technologies are being utilized in the life sciences.
Bayer Improves Radiology Diagnostics with AI
One of the most time-consuming diagnostic tools available is radiology. First, the patient sits tight for the required timeframe, before a radiologist sits down to analyze the captured images. Any number of problems can arise during this lengthy process, including misdiagnosis or overlooking a critical, small detail in a scan. Industry giant Bayer saw this problem and got to work finding a solution that could shorten radiology timeframes while also increasing accuracy.
Bayer has introduced Calantic Digital Solutions, a cloud-based technology designed to enhance scheduling and improve radiology diagnostics using AI. Scheduling is optimized, as patients are prioritized based on need and treated accordingly.
The technology can be used for computerized tomography (CT), X-Rays or even magnetic resonance imaging (MRI). An article published in PharmaPhorum regarding Bayer’s new approach points to a study conducted in 2018 that reported up to 40 million diagnostic errors annually, all attributed to medical imaging.
Gerd Krueger, head of radiology at Bayer, explained the company’s incentive for launching the new technology.
“With Calantic Digital Solutions, we are entering the fastest-growing segment in the radiology market and taking the next step on our way from a product provider to a solution provider,” Krueger said.
MIT Praises ML and AI while Highlighting Ethical Concerns
The renowned Massachusetts Institute of Technology recognizes the changes that are occurring in biotech and healthcare, going as far as to publish a recognition piece that states, “It’s been exciting to see technology that rewrites and improves what we thought was an established health concept.”
The article goes on to explain how AI and ML have infiltrated human life under the guise of electronic assistants such as the FitBit or Siri, translating our speech into real purchases or calling 911 because dangerous vitals have been recorded.
However, the MIT article provides a short warning in between words of praise. While these systems seem omniscient, ethics can’t be programmed. Concerns with AI and ML ethics question whether the technology will mimic the manner in which some physicians overlook or misdiagnose conditions in underrepresented populations. Because AI and ML learn as instructed, this problem could follow medicine into the technological future.
MultiOmic Health and Mesh Bio use AI to Fill Research Gaps
While some remain skeptical of AI’s ability to uniformly improve the lives of all patients, others are getting ahead and addressing the concerns.
An Asia-based study sponsored by MultiOmic Health and Mesh Bio is in the works, aimed specifically at using AI to access and analyze the data of patients with chronic metabolic diseases.
Andrew Wu, Ph.D., co-founder and CEO of Mesh Bio, commented on the collaborative goal.
“We are delighted to partner with MultiOmic Health on this important study for patients in Asia. Their therapeutic development programs for metabolic disease intervention have deep synergies with Mesh Bio’s mission to develop digital care delivery solutions for these diseases,” he said.
Asian populations are historically underrepresented in medical literature and research, including in chronic metabolic diseases. The AI technology will use samples of patient biologic substances to analyze genetic, proteomic and metabolic data alongside traditional clinical and or diagnostic tests. The information gathered can be used in future treatment research and development efforts.
