AI technology for MRI data analysis by Prof. Dr. Shadi Albarqouni, Professor of Computational Medical Imaging Research at University Hospital Bonn and Helmholtz AI Junior Research Group Leader at Helmholtz Munich CREDIT © Johann F. Saba, University Hospital Bonn (UKB)

An algorithm developed by researchers from Helmholtz Munich, the Technical University of Munich (TUM) and its University Hospital rechts der Isar, the University Hospital Bonn (UKB) and the University of Bonn is able to learn independently across different medical institutions. The key feature is that it is “self-learning”, i.e. it does not require extensive, time-consuming findings or markings by radiologists in the MRI images. This federated algorithm was trained on more than 1,500 MR scans of healthy study participants from four institutions while maintaining data privacy. The algorithm then was used to analyze more than 500 patient MRI scans to detect diseases such as multiple sclerosis, vascular disease, and various forms of brain tumors that the algorithm had never seen before. This opens up new possibilities for developing efficient AI-based federated algorithms that learn autonomously while protecting privacy. The study has now been published in the journal Nature Machine Intelligence.

Healthcare is currently being revolutionized by artificial intelligence. With precise AI solutions, doctors can be supported in diagnosis. However, such algorithms require a considerable amount of data and the associated radiological specialist findings for training. The creation of such a large, central database, however, places special demands on data protection. Additionally, the creation of the findings and annotations, for example the marking of tumors in an MRI image, is very time-consuming. To overcome these challenges, a multidisciplinary team from Helmholtz Munich, the University Hospital Bonn and the University of Bonn collaborated with clinicians and researchers at Imperial College London and TUM and its University Hospital rechts der Isar. The aim was to develop an AI-based medical diagnostic algorithm for MRI images of the brain, without any data annotated or processed by a radiologist. Furthermore, this algorithm was to be trained “federally”: In this way, the algorithm “comes to the data”, so that the medical image data requiring special protection could remain in the respective clinic and did not have to be collected centrally.

Learning from several institutes without data exchange

In their study, the researchers were able to show that the federated AI algorithm they developed outperformed any AI algorithm trained using only data from a single institution. “In his ‘The Wisdom of Crowds,’ James Surowiecki argued that large groups of people are smarter, no matter how smart an individual might be. Basically, this is how our federated AI algorithm works,” says Prof. Dr. Shadi Albarqouni, Professor of Computational Medical Imaging Research at the Department of Diagnostic and Interventional Radiology at University Hospital Bonn and Helmholtz AI junior research group leader at Helmholtz Munich. To pool knowledge about MRI images of the brain, the research team trained the AI algorithm in different and independent medical institutions without violating data privacy or collecting data centrally. “Once this algorithm learns what MRI images of the healthy brain look like, it will be easier for it to detect disease. To achieve this requires intelligent computational aggregation and coordination between the participating institutes,” says Prof. Dr. Albarqouni. PD Dr. Benedikt Wiestler, senior physician at TUM’s University Hospital rechts der Isar and also involved in the study, adds: “Training the model on data from different centers contributes significantly to the fact that our algorithm detects diseases much more robustly than other algorithms that are only trained with data from one center.”

Towards affordable collaborative AI solutions

By protecting patient data while reducing radiologists’ workloads, the researchers believe their federated AI technology will significantly advance digital medicine. “AI and healthcare should be affordable, and that is our goal. With our study, we have taken a step in this direction,” says Prof. Dr. Albarqouni. “Our major goal is to develop AI algorithms, collaboratively trained at different, decentralized medical institutes, including those with limited resources.”

Your posture when taking a pill makes a big difference in how fast your body absorbs the medicine. CREDIT Khamar Hopkins/Johns Hopkins University

When you have a headache and reach for the pain reliever, you’re probably not thinking about your body position when you take the pill. But a new Johns Hopkins University study finds your posture can make a big difference—as much as an hour longer—in how fast your body absorbs the medicine.

The findings are based on what’s thought to be the first model to simulate the mechanics of drug dissolution on a human stomach.

“We were very surprised that posture had such an immense effect on the dissolution rate of a pill,” said senior author Rajat Mittal, a Johns Hopkins engineer and an expert in fluid dynamics. “I never thought about whether I was doing it right or wrong but now I’ll definitely think about it every time I take a pill.”

The work is newly published in Physics of Fluids.

In recent years models have been created to authentically represent the workings of several major organs, notably the heart. The model developed by the team, called StomachSim, appears to be one of the first to be able to conduct realistic simulation of the human stomach. Blending physics with biomechanics and fluid mechanics, StomachSim mimics what happening inside a stomach as it digests food, or in this case, medicine.

Most pills do not start working until the stomach ejects their contents into the intestine. So the closer a pill lands to the last part of the stomach, the antrum, the faster it starts to dissolve and empty its contents through the pylorus into the duodenum, the first part of the small intestine. If you’re aiming a pill for this part of the stomach, posture is critical to play into both gravity and the natural asymmetry of the stomach.

The team tested four postures. Taking pills while lying on the right side was by far the best, sending pills into the deepest part of the stomach to achieve a dissolution rate 2.3 times faster than even an upright posture. Lying on the left side was the worst. The team was very surprised to find that if a pill takes 10 minutes to dissolve on the right side, it could take 23 minutes to dissolve in an upright posture and over 100 minutes when laying on the left side.

“For elderly, sedentary or bedridden people, whether they’re turning to left or to the right can have a huge impact,” Mittal said.

Standing upright was a decent second choice, essentially tied in effectiveness with lying straight back.

The team also considered what stomachs that aren’t functioning at full strength due to gastroparesis caused by diseases such as diabetes and Parkinson’s Syndrome meant for pill dissolution. Even a small change in the conditions of the stomach can lead to significant differences in the outcome of an oral drug, said lead author Jae Ho “Mike” Lee, a former postdoctoral researcher at Johns Hopkins.

The impact of stomach disease on drug dissolution was similar to that of posture—which underscores how significant a difference posture makes.

“Posture itself has such a huge impact it, it’s equivalent to somebody’s stomach having a very significant disfunction as far as pill dissolution is concerned,” Mittal said.

Future work will attempt to predict how the changes in the biomechanics of the stomach affect how the body absorbs drugs, how food is processed in the stomach and the effect of posture and gastroparesis on food digestion.