Research news in brief

Detecting breast cancer with artificial intelligence

Hungarian research, published in Scientific Reports, has shown that “deep learning” artificial intelligence that is widely used to detect objects in images could improve early breast cancer detection. Computer assisted detection (CAD) systems have been used over the past two decades to help radiologists detect breast cancer by analysing mammograms and marking suspicious regions, which are then reviewed by a radiologist. However, use of these technologies is expensive ($400 million a year in the US) and their benefits remain controversial. Deep learning is a subset of machine learning. Its networks are inspired by knowledge of how biological brains, such as those of humans or animals, work. Deep learning networks “learn” from datasets annotated by humans to reach image recognition capabilities similar to those of humans’. The researchers proposed an improved CAD system based on state-of-the-art deep learning that can be trained to detect and localise breast lesions. When tested on a dataset of 115 mammography cases (two of four images per case) with proven cancers, the authors’ improved CAD system classified 90% of malignant lesions with very few false positives and without human intervention. Current screening methods, which include assessment by radiologists, correctly detect 77–87% of cancers. The findings suggest that expensive, traditional CAD methods may be replaced by less expensive deep learning methods that are currently being used to recognise objects – such as dogs and cars – in traditional images, to help radiologists detect more cancers. However, the authors cautioned that so far, they were only able to test their method on a small dataset of images with proven cancers.

New possibility for untreatable blood cancer

Researchers from QIMR Berghofer Medical Research Institute have discovered a new biomarker for multiple myeloma that could help to unlock the medical mystery behind the untreatable blood cancer that affects mostly older Australians. Multiple myeloma grows in the bone marrow and affects around 1800 Australians every year. The average age of diagnosis is 70 years. The QIMR study, published in Cancer Cell, found that a particular molecule called interleukin (IL)-18 suppressed the immune system to help create a bone marrow environment where the cancer was more likely to grow. The study analysed the impact of IL-18 on 152 patients with multiple myeloma and found strong evidence that high levels of the molecule were associated with poorer survival. The researchers note that IL-18 was responsible for promoting the immune suppressive function of a particular kind of white blood cell (granulocyte) in the bone marrow. The resulting suppression hindered T cells from detecting and destroying cancer cells. “IL-18 has traditionally been recognised as a growth factor for immune cells because it was thought to promote the activity of the white blood or ‘natural killer’ cells that protect us from infection and cancer,” the researchers said. “We’ve turned that thinking on its head with our discovery that IL-18 is actually a pro-tumour factor, which causes virtually the opposite effect. IL-18 is critical in the progression of multiple myeloma by enabling one part of the immune system to suppress another.”

Refugees missing out on appropriate care

Many refugees who are living in the Australian community are missing out on appropriate health care due to problems such as fragmented services and poor continuity of care, according to a series of articles in Public Health Research and Practice, published by the Sax Institute. Professor Mark Harris, from the Centre for Primary Health Care and Equity at the University of New South Wales, said that although good services exist, there was a lack of integration between specialist services and the primary health care system, making it difficult for patients to transition from one service to another. “Many refugees and people seeking asylum have very complex health needs – including post-traumatic stress disorder, chronic disease and, in some cases, torture – and there are a number of specialist services that people can access,” said Professor Harris, who is also a volunteer medical officer at the NSW Asylum Seekers Centre. “The challenge often comes when people need to transition into mainstream general practice. Our research has identified a series of systematic and individual challenges during transition, including disrupted access to Medicare, fragmented care, mainstream GPs being unable to cope with the complex social needs and entitlements of refugees, and difficulties ensuring the proper transition of medical records. For individual refugees, there can often be a lack of understanding of how to navigate a complex and unfamiliar health system, a lack of specialist medical experience for their complex needs, and the fact that more pressing concerns such as visa, housing and employment stresses often take precedence over medical needs. “A combination of some or all of these factors can prevent refugees and people seeking asylum successfully transitioning into general practice care – some will return to overloaded specialist care providers, but others simply fall through the cracks of the health care system. This can mean some patients may not have optimal treatment of their mental health conditions. They may then lie dormant and become reactivated at times of stress,” said Professor Harris.

More babies may mean more tooth loss for mothers

Having a larger family is linked to a heightened tooth loss risk for mums, according to a large European study published in the Journal of Epidemiology and Community Health. The popular saying “gain a child, lose a tooth” suggests that fertility may be linked to tooth loss, but until now there has been no hard data. To try and plug this gap, the German researchers drew on data from Wave 5 of the Survey of Health, Ageing, and Retirement in Europe (SHARE). SHARE contains information on the health, educational attainment, and household income of more than 120 000 adults aged 50 years or over from 27 European countries plus Israel. Wave 5 was conducted in 2013 and included questions on the full reproductive history and number of natural teeth of 34 843 survey respondents from Austria, Belgium, Czech Republic, Denmark, Estonia, France, Germany, Italy, Luxembourg, the Netherlands, Slovenia, Spain, Switzerland and Israel. The average age of the respondents in Wave 5 was 67, and they reported an average of ten missing teeth —normally adults have 28 plus four wisdom teeth in their mouth. As might be expected, tooth loss increased with age, ranging from nearly seven fewer teeth for women in their 50s and 60s up to 19 fewer teeth for men aged 80 years and above. Higher levels of educational attainment were also linked to lower risk of tooth loss among women. The researchers looked at the potential impact of having twins or triplets rather than singletons, and the sex of the first two children, on the assumption that if the first two were of the same sex, the parents might be tempted to try for a third child. They applied a particular type of statistical technique (instrumental variables regression), which exploits random natural variation in a variable that is only associated with the exposure and affects the outcome only through that exposure, so mimicking a randomised controlled trial. A third child after two of the same sex was associated with significantly more missing teeth for women, but not men, if compared with parents whose first two children were different sexes. This suggests that an additional child might be detrimental to the mother’s, but the not the father’s, mouth health, say the researchers. The authors acknowledged that their analyses covered narrow groups with particular types of fertility patterns, and relatively small numbers in the groups of interest, so the results should be interpreted with caution.

To find a doctor, or a job, to use GP Desktop and Doctors Health, book and track your CPD, and buy textbooks and guidelines, visit doctorportal.