UK deploys AI online to reduce common drug side effects

Artificial intelligence could help clinicians assess which patients are likely to encounter the harmful side effects of some commonly used antidepressants, antihistamines, and bladder medicines.

Research led by the University of Exeter and Kent and Medway NHS and Social Care Partnership Trust, published in Age and Ageing, assessed a new tool designed to calculate which medicines are more likely to experience adverse anticholinergic effects on the body and brain. These complications can occur from many -prescription and over-the-counter drugs which affect the brain by blocking a key neurotransmitter called acetylcholine. Many medicines, including some bladder medications, anti-depressants, and medications for stomach and Parkinson’s disease have some degree of anticholinergic effect. They are commonly taken by older people.

Anticholinergic side effects include confusion, blurred vision, dizziness, falls and a decline in brain function. Anticholinergic effects may also increase the risks of falls and may be associated with an increase in mortality. They have also been linked to a higher risk of dementia when used long-term.

Now, researchers have developed a tool to calculate the harmful effects of medicines using artificial intelligence. The team created a new online tool, International Anticholinergic Cognitive Burden Tool (IACT), it uses natural language processing which is an artificial intelligence methodology, and chemical structure analysis to identify medications that have an anticholinergic effect.

The tool is the first to incorporate a machine learning technique, to develop an automatically updated tool available on a website portal. The anticholinergic burden is assessed by assigning a score based on reported adverse events and aligning closely with the chemical structure of the drug being considered for prescription, resulting in a more accurate and up-to-date scoring system than any previous system. Ultimately, after further research and modeling with real-world patient data the tool developed could help to support prescribing and reducing risks from common medicines.

Professor Chris Fox, at the University of Exeter, is one of the study authors. He said:: “Use of medicines with anticholinergic effects can have significant harmful effects for example falls and confusion which is avoidable,  we urgently need to reduce the harmful side effects as this can leads to hospitalization and death. This new tool provides a promising avenue towards a more tailored personalized medicine approach, of ensuring the right person gets a safe and effective treatment whilst avoiding unwanted  anticholinergic effects.”

The team surveyed 110 health professionals, including pharmacists and prescribing nurses.  Of this group, 85 percent said they would use a tool to assess the risk of anticholinergic side effects, if available. The team also gathered usability feedback to help improve the tool further.

Dr. Saber Sami,  at the University of East Anglia, said: “Our tool is the first to use innovative artificial intelligence technology in measures of anticholinergic burden – ultimately, once further research has been conducted the tool should support pharmacists and prescribing health professionals in finding the best treatment for patients.”

Professor Ian Maidment, from Aston University, said: “I have been working in this area for over 20 years. Anticholinergic side effects can be very debilitating for patients. We need better ways to assess these side-effects.”

The research team includes collaboration with AKFA University Medical School, Uzbekistan, and the Universities of East Anglia, Aston, Kent, and Aberdeen. They aim to continue the development of the tool with the aim that it can be deployed in day-to-day practice which this study supports.

University of New South Wales shows how the Southern Ocean takes on the heat of climate change

In the past 50 years, the oceans have absorbed more than 90% of the excess heat caused by our carbon dioxide emissions, with one ocean absorbing the vast majority. Paradise Bay in the Southearn Ocean, MarcAndreLeTourneux/Shutterstock  CREDIT MarcAndreLeTourneux/Shutterstock

“The Southern Ocean dominates this ocean heat uptake, due in part to the geographic set-up of the region,” said UNSW Ph.D. candidate Maurice Huguenin, the lead author of the new study.

“Antarctica, which is surrounded by the Southern Ocean, is also surrounded by strong westerly winds,” Mr. Huguenin said.

“These winds influence how the waters absorb heat, and around Antarctica, they can exert this influence while remaining uninterrupted by land masses – this is key to the Southern Ocean being responsible for pretty much all of the net global ocean heat uptakes,” he said.

Mr. Huguenin said that these winds blow over what is effectively an infinite distance – cycling uninterrupted at southern latitudes – which continuously draws cold water masses to the surface. The waters are pushed northward, readily absorbing vast quantities of heat from the atmosphere, before the excess heat is pumped into the ocean’s interior around 45-55°S.

But, while ocean warming helps slow the pace of climate change, it is not without cost said co-author Professor Matthew England at UNSW Science and Deputy Director of ACEAS.

“Sea levels are rising because heat causes water to expand and ice to melt. Ecosystems are experiencing unprecedented heat stress, and the frequency and intensity of extreme weather events is changing” Prof. England said.

“We still have a lot to learn about ocean warming beyond the 50 years highlighted in our study,” Mr Huguenin added.

“All future projections, including even the most optimistic scenarios, predict warmer oceans in the future.”

“If the Southern Ocean continues to account for the vast majority of heat uptake until 2100, we might see its warmth increase by up to seven times more than what we have already seen up to today.”

Prof. England said this will have an enormous impact around the globe including disturbances to the Southern Ocean food web, rapid melting of Antarctic ice shelves and changes in the conveyor belt of ocean currents.

The scientists used a novel experimental approach to find exactly where excess heat is taken up by the oceans and where it ends up after absorption. This was previously difficult to detect due to relatively sparse and short-lived measurement records.

The team ran a model with atmospheric conditions fixed in the 1960s – prior to any significant human-caused climate change. They then compared this model to others in which the oceans experience the past 50 years of climate change one ocean basin at a time. The results revealed that the Southern Ocean is the most important absorber of greenhouse gas-trapped heat and that its circulation – driven by winds – is uniquely set up to force this excess heat into the ocean interior.

To better understand how Southern Ocean heat uptake continues to evolve, the scientists call for ongoing monitoring of this remote ocean – including the deployment of additional deep-reaching Argo floats, which are pivotal for tracking ocean heat content. They also stress the urgency of reducing greenhouse gas emissions.

The less carbon dioxide we emit into the atmosphere, the less ocean change and sea-level rise we will lock in,” the authors said.

“This can help limit the level of adaptation required by the billions of people living near the ocean, by minimising the detrimental impacts of ocean warming on both sea-level and their primary food source.”

Helsinki researchers use ML to unlock the genomic code in clinical cancer samples

A new paper from the University of Helsinki suggests a method for accurately analyzing genomics data in archival cancer biopsies. This tool uses machine learning methods to correct damaged DNA and unveil the true mutation processes in tumor samples. This helps to unlock tremendous medicine values in millions of archival cancer samples.

Molecular-based diagnosis helps to match the right patient with the right cancer treatment. Researchers took particular interest in DNA profiling in clinical cancer samples.

This invaluable source is currently not being used for molecular diagnosis due to the poor DNA quality. Formalin causes severe damage to DNAs, which is an inevitable challenge to analyse cancer genomes in preserved tissues, says lead author Qingli Guo from the University of Helsinki.

Analyzing mutation processes in cancer genomes can help early cancer detection, accurately diagnose cancer, and reveal why some cancers become resistant to treatment. The new method can dramatically accelerate the development of clinical applications that can directly impact future cancer patient care.

The new method predicted more than 90% of developing cancer processes

Lead author Qingli Guo works in close collaboration with scientists from The Institute of Cancer Research (ICR), London, and the Queen Mary University of London, developed machine learning methods, named FFPEsig, to unravel exactly how formalin mutates DNA.

Our results show that normally nearly half of the cancer processes will be missed without noise correction. However, using FFPEsig, more than 90% of them were accurately predicted. says Qingli.

Cancer evolves gradually. Profiling mutational processes in longitudinal samples help to identify clinical informative predictors and make a diagnosis of each tumor stage.

Our finding enables the characterization of clinically relevant signatures from the preserved tumor biopsies stored at room temperatures for decades. With a deep understanding of how formalin impacts the cancer genome, our study opens a huge opportunity to transform the developed signature detection assays using large cost-effective archival samples.

The researchers pointed out that the method currently does not completely remove artifacts that appeared in FFPE samples showing batch effects, and how well the tool performs varies by cancer type, so care must be taken to interpret any findings. We are also interested in further applying their methods to a much broader spectrum of archival samples in the future.

The research was funded by Cancer Research UK, the University of Helsinki, and in part by the Academy of Finland. This project is co-led by senior authors Prof. Ville Mustonen (University of Helsinki) and Prof. Trevor Graham (the ICR).