Robust generalization under climate change remains a major challenge for machine learning applications in climate science. Most existing approaches struggle to extrapolate beyond the climate they were trained on, leading to a strong dependence on training data from model simulations of warm climates. Use of climate-invariant inputs improves generalization but requires challenging manual feature engineering. Here, we present CERA (Climate-invariant Encoding through Representation Alignment), a machine learning framework consisting of an autoencoder with explicit latent-space alignment, followed by a predictor for downstream process estimation. We test CERA on the problem of parameterizing moist-physics processes. Without training on labeled data from a +4K climate, CERA leverages labeled control-climate data and unlabeled warmer-climate inputs to improve generalization to the warmer climate, outperforming both raw-input and physically informed baselines in predicting key moisture and energy tendencies. It captures not only the vertical and meridional structures of the moisture tendencies, but also shifts in the intensity distribution of precipitation including extremes. Ablation experiments show that latent alignment improves both accuracy and the robustness across random seeds used in training. While some reduced skill remains in the boundary layer, the framework offers a data-driven alternative to manual feature engineering of climate invariant inputs. Beyond parameterizations used in hybrid ML-physics systems, the approach holds promise for other climate applications such as statistical downscaling.

Artificial intelligence is to be used to check if carers are likely to quit their jobs. Any issues employees have had with pay, punctuality, or problems with their manager will be among the data fed into the algorithm. The technology is meant to combat the staffing crisis in social care by giving bosses an early chance to persuade workers to stay. Private healthcare company Cera claims its AI could prevent around 50,000 staff leaving every year. The firm said it has been shown to detect carers who are at risk of quitting three times more accurately than human managers can.

This Digital Health News industry roundup includes an Innovate UK Smart Award for a smart inhaler tracker, details on the digital transformation programme for Research Data Scotland and Cera's commitment to reduce hospitalisations with its digital healthcare approach. Customer analysis of a care home admin platform has revealed that it can save a significant number of hours in admin time for care homes. The CoolCare software focuses on making admin more efficient and more profitable. Its most recent analysis found that many of its customers were seeing results above pre-pandemic levels, including above-national increases in weekly fees, occupancy and private payer ratios. With increased occupancy, comes more admin.

Care patients could see at home visits replaced by a call from an AI-powered voice assistant in a new British trial. Dubbed'Siri for care', a human-like virtual assistant will ring patients once a week to ask a list of automated questions. An algorithm will then analyse the answers and alert carers if there are any deteriorations in health so they can arrange a doctor's visit. Similar trials in Europe have reduced A&E visits by 55 per cent, according to the tech company behind it. The new technology will be tested out on patients in domiciliary care for those who are living independently but who rely on helpers to visit them regularly.

London-based digital health startup Cera Care Limited has raised $320m (£263.6m) in an equity and debt funding round, with the goal of increasing the number of patients it serves from 15,000 to 100,000. Based in the UK and with some operations in Germany, Cera's digital platform provides services such as prescriptions home deliveries and nursing care based on data analysis. Cera nurses perform at-home appointments where patient data is collected and studied by AI algorithms. Cera claims its AI tech can predict patient health deterioration before it occurs and therefore reduce the need for hospitalisation. The company says this can reduce the strain on NHS services and free up hospital beds and care capacity.

Gene therapy, stem cells, artificial intelligence, the bionic eye – incredible advances in research and technology are opening up new frontiers for eye health. With each step forward, eye researchers are unlocking exciting possibilities for earlier detection of eye disease, better treatments and potentially even restoring sight. At the Centre for Eye Research Australia's Looking to the Future Community Forum, held in October 2019, a selection of CERA's leading experts discussed the current eye research landscape and what the future holds. Over two panels – 'Gene therapy demystified' and'Everything you need to know about artificial intelligence and new technology' – they explored everything from the bionic eye, to corneal cells grown from skin, to computer programs that can detect eye disease with better-than-human accuracy. Here is a snapshot of some of the key ideas discussed, and the areas of research shaping the future of eye health.

Thanks to improvements and advances in health care, many people around the world are living longer. While this is undoubtedly good thing, it inevitably follows that people will require more assistance as they get older. As a greater number of us reach old age, the stresses on health services increase. NHS England data from 2015, for example, showed that between 2007/08 and 2013/14 the amount of accident and emergency attendances by people aged 60 or over grew by two-thirds. This, it noted, represented a "steeper increase than is expected by demographic change alone."

The social care crisis – as demonstrated with the Conservative Party's disastrous manifesto pledge – is one of the biggest issues facing the NHS. Spending squeezes have led to bed-blocking in hospitals; while low wages and poor working conditions are causing carers to leave the industry in droves. One social care startup thinks that a solution might lie in AI. Cera, founded by Ben Maruthappu, a former innovation adviser to NHS England, is a startup that wants to provide smarter social care using an Uber-style platform to match carers and patients. The startup has to date raised £2.7 million in early funding, and in March signed a partnership with Barts Health NHS Trust, as well as several clinical commissioning groups and hospitals in London to test its solution.

Can AI overhaul the social care system? The startup has today taken the tiniest baby steps to launch an AI chatbot that it hopes will, at an unspecified point in future, be able to assist carers with recommendations for home care of people with conditions such as dementia. And even potentially steer off medical emergencies via pre-emptive alerts. The far more basic reality of the chatbot it's launching today is that Martha (as it's called) will be able to recommend care packages to potential customers. Which underlines how the inflated promises of AI really do hinge on data acquisition.