Media Monitoring: 23rd – 30th April

Machine learning can help identify antibiotic-resistant bacteria

University of Nottingham researchers have created software which can identify antibiotic-resistant bacteria transmitted between humans, animals and the environment
The research is crucial in examining the spread of antibiotic resistant bacteria in spaces in which humans and animals coexist or share closely, such as livestock farming
This bacteria can infect humans and render traditional medicine treatments ineffective

‘Antimicrobial resistance is a slow tsunami’

University of Oxford’s Professor Timothy Walsh reflects on the lack of immediacy and accountability in the battle against antimicrobial resistance (AMR) and why stronger action throughout policy and healthcare is urgently needed

AI can help to identify cancer through secure patient data

A group of researchers have set out to discover whether a form of AI called swarm learning could be used to help computers predict cancer in medical images of patient tissue samples, without the need to release data from hospitals
Swarm learning trains AI algorithms to detect patterns in data that is held by a hospital or university, such as genetic changes within images of human tissue, the algorithm is then sent securely to a central computer
It is then combined with other algorithms from other hospitals, thus creating a refined algorithm. This optimized algorithm is then applied again to the original data in order to improve detection

£60 million Welsh Government plan to tackle the backlog is unveiled

The new scheme, which will run over the next four years, aims to reduce waiting times and transform care
Eluned Morgan, Minister for Health and Social Services, has vowed that nobody will wait more than a year for treatment in most specialties by Spring 2025

Majority of eligible women would prefer self-sampling for cervical screening

The study found that a majority of women – 51.4% – preferred self-sampling for HPV-based cervical cancer screening compared to 36.5% who preferred being tested by a clinician.
This is largely due to concerns about privacy, embarrassment and providing more flexibility for taking tests
Self-sampling was found to be more popular among older participants, lesbian & bisexual women, and women with previous experience of self-administered blood tests

New genetic changes in esophageal cancer patients found in study

The research team found that DNA changes in BE cells which predict esophageal cancer can be spotted years before cancer is diagnosed
The changes include rearrangements of huge chunks of DNA and damage to both copies of a tumor-suppressing gene called TP53
While the findings are important, encompassing over 400 tissue samples, results from this 80-patient study would need to be further scrutinised in other patient groups before they could be used to predict whether other BE patients will progress to cancer in clinical circumstances