At least eight people have died in two consecutive Russian drone attacks on a medical centre in the northeast Ukrainian city of Sumy, Ukrainian officials have said. The first attack on Saturday morning killed one person, and it was followed by another attack while patients and staff were evacuating, Ukraine's Interior Minister Ihor Klymenko said. Ukraine's President Volodymyr Zelenskyy said on his Telegram channel that Russia had hit the hospital using Shahed drones, stating that eleven people were injured. Sumy lies just across the border from Russia's Kursk region where Kyiv launched a shock offensive on August 6, which it says is aimed partly at creating a "buffer zone" inside Russia. Regional prosecutors said the first attack in Sumy on Saturday took place at about 7:35am (04:35 GMT), hitting the hospital where there were 86 patients and 38 staff.

Deep neural networks show impressive performance in medical imaging tasks. However, many current networks generalise poorly to data unseen during training, for example data generated by different medical centres. Such behaviour can be caused by networks overfitting easy-to-learn, or statistically dominant, features while disregarding other potentially informative features. Moreover, dominant features can lead to learning spurious correlations. For instance, indistinguishable differences in the sharpness of the images from two different scanners can degrade the performance of the network significantly. To address these challenges, we evaluate the utility of spectral decoupling in the context of medical image classification. Spectral decoupling encourages the neural network to learn more features by simply regularising the networks' unnormalised prediction scores with an L2 penalty. Simulation experiments show that spectral decoupling allows training neural networks on datasets with strong spurious correlations. Networks trained without spectral decoupling do not learn the original task and appear to make false predictions based on the spurious correlations. Spectral decoupling also significantly increases networks' robustness for data distribution shifts. To validate our findings, we train networks with and without spectral decoupling to detect prostate cancer on haematoxylin and eosin stained whole slide images. The networks are then evaluated with data scanned in the same medical centre with two different scanners, and data from a different centre. Networks trained with spectral decoupling increase the accuracy by 10 percentage points over weight decay on the dataset from a different medical centre. Our results show that spectral decoupling allows training robust neural networks to be used across multiple medical centres, and recommend its use in future medical imaging tasks.

Chest radiography is the most common imaging exam used for lung cancer screening. However, the size, density and location of lung lesions make their detection on chest X-rays challenging. Recently, machine-learning methods have been developed to help improve diagnostic accuracy, with deep convolutional neural networks (DCNNs), showing promise for chest radiograph interpretation. A study from four medical centres on three continents has now demonstrated that DCNN software can improve radiologists' detection of malignant lung cancers on chest X-rays (Radiology 10.1148/radiol.2019182465). "The average sensitivity of radiologists was improved by 5.2% when they re-reviewed X-rays with the deep-learning software," says Byoung Wook Choi from Yonsei University College of Medicine in Seoul, Korea.

Five new clinics will open in the UK next year that will use artificial intelligence to help speed up disease diagnosis. The medical technology centres in Leeds, Oxford, Coventry, Glasgow and London will be funded by the Government as it looks to increase its investment in AI and improve patient treatment. The centres will use AI software to digitalise scans and biopsies, and develop products to detect diseases early. The large investment, costing £50million, will ensure people get personalised treatment sooner, as well as freeing up doctors time. Business, Energy and Industrial Strategy Secretary Greg Clark said: 'AI has the potential to revolutionise healthcare and improve lives for the better.' 'The innovation at these new centres will help diagnose disease earlier to give people more options when it comes to their treatment, and make reporting more efficient, freeing up time for our much-admired NHS staff to spend on direct patient care.'

Welcome to the second instalment of'Automate This!,' a Day 6 series about the future of work in an artificially intelligent world. In 2011, Krista Jones was diagnosed with a rare form of cancer. The next five years were a blur of doctor's visits and operations. "I think I saw seven doctors over that time period," Jones recalls. I was heading towards a double mastectomy, mostly out of fear for the fact that nobody could explain why [the tumours] were reoccurring." Jones' final treatment plan was built using algorithms and big data -- some of the precursors to today's A.I. technology. That plan made it possible for Jones to forgo a painful double mastectomy, and ultimately left her cancer-free. "Not only did it save my life but it left me whole in so many different ways," she says. "[It] avoided some of the scars, emotionally and physically, that most people who go through cancer treatment are left with." The treatment plan that helped Krista Jones beat a rare form of cancer was developed using machine learning algorithms and big data. She's seen the downsides of machine learning technologies, too. Her own son was forced to rethink his plan to become a radiologist after watching his career prospects dwindle thanks to automation. Still, Jones is convinced that artificial intelligence is the future of health care. "I think what we need to do is harness the good while regulating the bad," she says, "such that we don't get hung up and stop the development of life-saving treatments." "The only next step is now replacing those actual physical physicists and doctors that actually say: 'Yes, this is the right treatment plan'.