Deep Learning Aids Diagnosis of Chronic Pancreatitis
A deep learning (DL)–based model achieved a high accuracy in pancreas segmentation for patients with chronic pancreatitis (CP) and healthy individuals, a new study finds. The model showed robust performance across diverse scanning protocols and anatomic variations, although its accuracy was affected by visceral fat area and pancreas volume.
METHODOLOGY:
Researchers developed a DL-based tool using the neural network U-Net (nnU-Net) architecture for the automated segmentation of retrospectively collected CT scans of the pancreas of healthy individuals and of patients with CP.
Scans were obtained from one hospital each in Aalborg (n = 373; 223 patients with CP and 150 healthy individuals) and Bergen (n = 97 patients with CP), along with an online dataset from the National Institutes of Health (NIH; n = 80 healthy individuals).
The tool was validated and tested using internal and external datasets, and its performance was compared with manual processing done by radiologists using the Sørensen-Dice index.
The tool's performance was examined for potential correlation with factors including visceral fat area at the third lumbar level, pancreas volume, and CT scan parameters.
TAKEAWAY:
The tool demonstrated strong performance with mean Sørensen-Dice scores of 0.85 for the Aalborg test dataset, 0.79 for the Bergen dataset, and 0.79 for the NIH dataset.
Sørensen-Dice scores were positively correlated with visceral fat area across datasets (correlation coefficient [r], 0.45; P < .0001) and with pancreas volume in the Aalborg test dataset (r, 0.53; P = .0002).
< .0001) and with pancreas volume in the Aalborg test dataset (r, 0.53; = .0002). CT scan parameters had no significant effect on model performance.
The tool maintained accuracy across diverse anatomic variations, except in cases with severe pancreatic fat infiltration.
IN PRACTICE:
"This study presents a novel AI [artificial intelligence]–based pancreas segmentation model trained on both healthy individuals and CP [chronic pancreatitis] patients, demonstrating consistent and robust performance across internal and external test datasets that vary in patient characteristics and scanner parameters. The model has the potential to significantly enhance the efficiency and accuracy of pancreas segmentation in clinical practice and research, particularly for CP patients with complex anatomical features," the authors wrote.
SOURCE:
This study was led by Surenth Nalliah, Radiology Research Center, Department of Radiology, Aalborg University Hospital, Aalborg, Denmark. It was published online on May 14 in European Journal of Radiology .
LIMITATIONS:
Comprehensive hyperparameter optimisation was not performed due to computational constraints. Additionally, architectures beyond nnU-Net and other segmentation methods were not explored. Post hoc visualisation methods were not studied. Small sizes of datasets could have hindered model performance, and cases of severe pancreatic fat infiltration were not included.
DISCLOSURES:
Funding information was not provided for this study. One author reported receiving financial support from Health Hub, founded by the Spar Nord Foundation.
Hashtags
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
Yahoo
43 minutes ago
- Yahoo
Saint-Gobain acquires Maturix to expand digital construction reach
French multinational Saint-Gobain has acquired Denmark-based real-time concrete monitoring solutions provider Maturix, marking a significant step in the expansion of its digital construction chemicals platform. Maturix's wireless sensor technology facilitates remote monitoring of concrete properties during the curing and hardening process, which can halve the duration of the concrete construction cycle and enhance job-site efficiency. The integration of the Danish company into Saint-Gobain's portfolio strengthens the company's digital offerings across the concrete and cement value chains. Saint-Gobain customers can now benefit from reduced overdesign and optimised operations, contributing to the digital transformation of the concrete industry. This move follows Saint-Gobain's initiatives such as the deployment of Verifi, its digital in-transit concrete management suite, across North America, Europe, and Asia-Pacific. Verifi's real-time monitoring capabilities are designed to minimise waste, boost operational performance, and drive cost efficiencies. The addition of Maturix's technology is expected to complement Verifi's system, particularly in optimising concrete placement on construction sites. Since 2019, Maturix and Saint-Gobain have collaborated successfully, and they are currently developing a joint product offering. Earlier this year, Saint-Gobain completed the acquisition of FOSROC, a construction chemicals company, which further solidified Saint-Gobain's position in the global market, enhancing its capabilities in the construction chemicals sector. "Saint-Gobain acquires Maturix to expand digital construction reach" was originally created and published by World Construction Network, a GlobalData owned brand. The information on this site has been included in good faith for general informational purposes only. It is not intended to amount to advice on which you should rely, and we give no representation, warranty or guarantee, whether express or implied as to its accuracy or completeness. You must obtain professional or specialist advice before taking, or refraining from, any action on the basis of the content on our site. Error in retrieving data Sign in to access your portfolio Error in retrieving data Error in retrieving data Error in retrieving data Error in retrieving data
Gizmodo
an hour ago
- Gizmodo
Something Big Is Twisting Mercury's Crust
Mercury has it rough. Not only is it the smallest planet in the solar system, it's also the closest to our Sun. This unfortunate position has caused Mercury to develop cracks and fractures across its surface, and generate stresses to its crust, a new study has found. Mercury is dry, rugged, and heavily cratered; the planet appears deformed with towering cliffs and ridges, as well as fracture lines that run along its surface. The origin of Mercury's scars has long been a mystery: How did the planet cool and contract in such an unusual way billions of years ago after it formed? Turns out, the answer may be due to its uncomfortable proximity to the Sun. A team of researchers from the University of Bern created physical models of Mercury to see how much of the Sun's tidal forces affect the small planet, revealing that the star may have influenced the development and orientation of tectonic features on its surface over long periods of time. The results are detailed in a study published in the Journal of Geophysical Research: Planets. Planets form from the hot, molten material left over from the birth of a star. Over time, these objects cool and their internal materials shrink, causing them to contract as their crusts wrinkle and crack. Evidence has shown that Mercury, on the other hand, not only shrank—its surface also shifted laterally. Cracks and fractures also formed in its rocky crust. Scientists assumed that the process that shaped Mercury's outer layer was a result of this cooling and contracting, but the study suggests it may be the planet's cozy orbit around the Sun. Mercury has one of the most unique orbits in the solar system. It takes about 88 Earth days to complete one orbit around the Sun, during which the planet rotates around its axis three times every two orbits. Its orbit is also highly elliptical and is tilted by around 7 degrees compared to Earth's orbital plane, its eccentricity means that the tidal forces Mercury experiences from the Sun vary a lot. 'These orbital characteristics create tidal stresses that may leave a mark on the planet's surface,' Liliane Burkhard, a researcher at the Space Research and Planetary Sciences Division at the Institute of Physics at the University of Bern, and lead author of the study, said in a statement. 'We can see tectonic patterns on Mercury that suggest more is going on than just global cooling and contraction.' The team behind the study sought to investigate how these tidal forces contribute to shaping Mercury's crust. They used physical models of Mercury over the past 4 billion years to calculate how the Sun's tidal forces may have influenced its surface tensions. The results showed that the the changing gravitational pull of the Sun has impacted Mercury's tectonic features over time. 'Tidal stresses have been largely overlooked until now, as they were considered to be too small to play a significant role,' Burkhard said. 'Our results show that while the magnitude of these stresses is not sufficient to generate faulting alone, the direction of the tidally induced shear stresses are consistent with the observed orientations of fault-slip patterns on Mercury's surface.' The recent findings can also be applied to other planets, illustrating how subtle forces aside from tectonics can make a lasting impact on its surface. 'Understanding how a planet like Mercury deforms helps us understand how planetary bodies evolve over billions of years,' according to Burkhard. The scientists behind the new study are hoping to gather more clues about Mercury's deformed surface through the BepiColombo mission, which launched in October 2018 as a joint venture between the European Space Agency (ESA) and the Japan Aerospace Exploration Agency (JAXA). BepiColombo is only the third spacecraft to visit Mercury; the elusive planet is hard to reach due to the Sun's powerful gravitational pull that may have maimed the planet's surface.
Yahoo
an hour ago
- Yahoo
Muscles, fats and simple blood test can predict menopausal women's health
SINGAPORE – A simple blood test can predict which woman will have less muscle and will be walking more slowly later in life. It is also practical and cheaper than current methods of measuring muscle, such as the current gold standard magnetic resonance imaging scans or strength tests, which are also more time-consuming. This new insight from a longitudinal cohort study of midlife women in Singapore shone light on how muscle strength, visceral fat and their association with physical decline after menopause can potentially lead to downstream health impacts among women here. Researchers from the National University Hospital (NUH) and National University of Singapore (NUS) found that women with a lower creatinine-to-cystatin C ratio (CCR) – a marker derived from blood tests – had less muscle and walked more slowly as they age. Creatinine is a by-product of normal muscle function and energy use, and a higher level indicates higher skeletal muscle mass or poor kidney function. Cystatin C is a protein produced by the body's cells that is filtered out by the kidneys. A normal cystatin C level rules out poor kidney function. A low CCR of under 8.16 was associated with a lower muscle volume of 0.35 litres in the thigh, and a slower gait of 0.049m a second. This suggested that CCR could be a useful early warning sign for age-related muscle loss, which may lead to falls, frailty and reduced quality of life. The findings were published in Menopause, a monthly peer-reviewed journal, in March. The scientists involved in the study are from the Integrated Women's Health Programme (IWHP) at NUH and the NUS Yong Loo Lin School of Medicine. The IWHP was initiated to identify and address the healthcare needs of midlife Singaporean women. It recruited a cohort of 1,200 Chinese, Malay and Indian women aged 45 to 69 between 2014 and 2016 – about 70 per cent of whom were post-menopausal. Their health metrics were then tracked over time. In the first study based on this cohort, published in international journal Maturitas in October 2023, the researchers shared a ranking of menopausal symptoms – with joint and muscular discomfort found to be the most reported symptom among women in Singapore. Called arthralgia, it had moderate or severe impact on a third of the midlife women of the cohort. A subsequent study, published in the Diabetes, Obesity and Metabolism journal in October 2024, found that women with both weak muscle strength and high levels of visceral fat – the deep belly fat around the internal organs – had the highest risk of developing prediabetes or type 2 diabetes. Their risk was 2.63 times higher than that of women who had normal muscle strength and lower fat levels. Having just one of these conditions also increased their risk, though to a lesser degree. The risk from having high visceral fat alone is 1.78 times higher. Among those with weak muscle strength, women with high visceral fat faced 2.84 times as much risk compared with those with low visceral fat. Explaining the impetus for the study, IWHP lead Yong Eu Leong said: 'Muscle... burns up fat. What about those who have weak muscles? Does it affect the risk for diabetes in the future?' The cohort's initial muscle and visceral fat measurements served as a baseline for researchers to track changes over the years. Researchers then analysed how changes in fat and muscle measurements taken about six years later – by then, about 90 per cent of the women were post-menopausal – related to whether women had developed diabetes. Professor Yong, who also heads the division of benign gynaecology in the Department of Obstetrics and Gynaecology at NUH, noted that a large proportion of women in Singapore are 'skinny fat', where their body mass index is within the normal range, but that they have high levels of visceral fat and low muscle mass. 'One way (to know what your risks are) is to measure your walk and the speed at which you walk. If you cannot walk fast and straight, then your health is not so good. We wanted to see if we can develop a test that can predict gait speed. We wanted to look at molecules that actually measure muscle functions,' he said. 'These findings validated our previous (IWHP) research that showed that women should not just focus on weight loss, but on building muscle strength through exercise for diabetes prevention,' Prof Yong said. One participant of the IWHP, administrative assistant Sabarina Jumarudin, is living proof of the findings. The 59-year-old grandmother used to weigh 93kg and suffered from sleep apnoea. Since undergoing bariatric surgery at NUH in 2018, a procedure that modifies the digestive system to help people with obesity lose weight, she has lost more than 30kg. Mindful of keeping her weight down, Madam Sabarina walks to the MRT station every day instead of taking the shuttle service, and takes the stairs instead of the escalator to catch the train. 'On my way home, I usually take a longer route to ensure I clock at least 10,000 steps a day, and practise stretching and breathing exercises to strengthen my core,' she said. 'I realised that small but consistent changes do make a big impact on my health, so I do what I can on a daily basis, and it gives me confidence to not only stay healthy physically and mentally but also stave off diabetes,' she added. Source: The Straits Times © SPH Media Limited. Permission required for reproduction Discover how to enjoy other premium articles here
