Tyler O'Neal, Staff Editor ACADEMIA

Queensland University of Technology's Wu develops jump test tool to predict athletic performance

Researchers studying the impact of fatigue on athletic performance have developed prototype software that can enable coaches to predict when elite athletes will be too fatigued to perform at their best.

At the Queensland University of Technology, QUT's Dr. Paul Wu led the study published today in the journal PLOS One.

The research, which applies the tools of statistics to physiology research, provides new insights for athletes and their coaches into how best to manage and predict fatigue levels.

The software algorithm enables coaches to conduct a simple test of an athlete's energy and performance levels and make predictions about how their level of fatigue could impact on their performance. Dr Paul Wu{module In-article}

"This is a tool to assist coaches," Dr. Wu said.

"Let's say you have a game tomorrow and the model predicts you're going to be very fatigued, that might change the coach's strategy.

"Having that knowledge ahead of time can be helpful."

The information could also enable coaches to personalize the training for individual athletes depending on their predicted fatigue levels as a result of different types of training.

The researchers in the study examined the two main types of fatigue athlete's experience in training.

The first is metabolic fatigue, which only takes up to three hours to recover from. The more serious fatigue, neuromuscular fatigue, can take upwards of 48 hours or more to recover from.

"In the elite sports setting, athletes often train twice a day, five days or more a week. If you develop neuromuscular fatigue and have training or competition the next day, you'll still be fatigued and have an elevated risk of injury," said Dr. Wu.

In this research, Dr. Wu and his collaborators studied data from a test called the countermovement jump (CMJ). To do the test, an athlete stands on a force plate, squats down, and jumps straight up as high as he or she can. The force plate records the force profile generated throughout the jump.

"In our study, we tested the athletes after low, moderate, and high-intensity training sessions. We did many jumps over time, from just before training sessions, to right after, and then in regular intervals up to 48 hours later," said Dr. Wu.

That many jumps involving multiple athletes led to a lot of data, and that data isn't simple either. So Dr. Wu and his team used a statistical analysis tool called functional Principal Components Analysis (fPCA) to find the hidden information about fatigue in all that data.

"By doing a few jump tests up to 30 minutes after training and then doing our analysis, we can predict the degree of neuromuscular fatigue. This allows coaches and athletes to prepare for the next workout or for the competition ahead of time," said Dr. Wu.

In addition, it arms athletes with important information about how they fatigue.

"It helps them to customize their training to avoid neuromuscular fatigue, and also allows them to benchmark themselves against others," said Dr. Wu.

The researchers have produced a prototype of software which could be used in the future by coaches to manage an athlete's fatigue and ensure peak performance.

Never before, have athletes and their trainers had access to so much data about their training. It's only through statistical analysis, like the one in this study, that is unlocking some of the key, hidden stories that athletes need to take advantage of the data.

Dr Wu believes the statistical analysis used here will help with other types of training data as well.

University of Luxembourg life sciences research leads to new supercomputer model supporting cancer therapy

Tyler O'Neal, Staff Editor ACADEMIA

Researchers from the Life Sciences Research Unit (LSRU) of the University of Luxembourg have developed a supercomputer model that simulates the metabolism of cancer cells. They used the program to investigate how combinations of drugs could be used more effectively to stop tumor growth. The biologists now published their findings in the scientific journal EBioMedicine of the prestigious Lancet group.

The metabolism of cancer cells is optimized to enable fast growth of tumors. "Their metabolism is much leaner than that of healthy cells, as they are just focused on growth. However, this makes them more vulnerable to interruptions in the chain of chemical reactions that the cells depend on. Whereas healthy cells can take alternative routes when one metabolic path is disabled, this is more difficult for cancer cells," explains Thomas Sauter, Professor of Systems Biology at the University of Luxembourg and lead author of the paper. "In our study, we investigated how drugs or combinations of drugs could be used to switch off certain proteins in cancer cells and thereby interrupt the cell's metabolism." un modele informatique novateur au service du traitement du cancer medium c4856 {module In-article}

Therefore, the researchers created digital models of healthy and of cancerous cells and fed them with gene sequencing data from 10,000 patients of the Cancer Genome Atlas (TCGA) of the American National Cancer Institute (NCI). Using these models, the researchers were able to simulate the effects different active substances had on cells' metabolisms so they could identify those drugs that inhibited cancer growth and at the same time didn't affect the healthy cells. The models allow filtering out drugs that do not work or are toxic so that only the promising ones are tested in the lab.

With the help of the models, they tested about 800 medications of which 40 were predicted to inhibit cancer growth. About 50 percent of these drugs were already known as anti-cancer therapeutics, but 17 of them are so far only approved for other treatments. "Our tool can help with the so-called "drug repositioning", which means that new therapeutical purposes are found for existing medication. This could significantly reduce the cost and time for drug development," Prof. Sauter said.

The particular advantage of the approach is the efficiency of its mathematical method. "We managed to create 10.000 patient models within one week, without the use of high-performance computing. This is exceptionally fast," comments Dr. Maria Pacheco, a postdoctoral researcher at the University of Luxembourg and first author of the study. In addition, Dr. Elisabeth Letellier, principal investigator at the Molecular Disease Mechanisms group at the University of Luxembourg and collaborator on the present study, further emphasizes "In the future, this could allow us to build models of individual cancer patients and virtually test drugs in order to find the most efficient combination. This could also bring fresh hope to patients for whom known therapies have proven to be ineffective."

So far, the models have been tested only for colorectal cancer, but the algorithm basically also works for all sorts of cancer, according to Thomas Sauter. He and his team are currently considering to develop commercial applications for their method.

Using artificial intelligence to deliver personalized radiation therapy

Tyler O'Neal, Staff Editor ACADEMIA

Newly published Cleveland Clinic-led research first to use medical scans to inform dose delivery

New Cleveland Clinic-led research shows that artificial intelligence (AI) can use medical scans and health records to personalize the dose of radiation therapy used to treat cancer patients.

Published today in The Lancet Digital Health, the research team developed an AI framework based on patient computerized tomography (CT) scans and electronic health records. This new AI framework is the first to use medical scans to inform radiation dosage, moving the field forward from using generic dose prescriptions to more individualized treatments. CAPTION New research led by Mohamed Abazeed, M.D., Ph.D., of Cleveland Clinic shows that artificial intelligence (AI) can use medical scans and health records to personalize the dose of radiation therapy used to treat cancer patients. CREDIT Russell Lee{module In-article}

Currently, radiation therapy is delivered uniformly. The dose delivered does not reflect differences in individual tumor characteristics or patient-specific factors that may affect treatment success. The AI framework begins to account for this variability and provides individualized radiation doses that can reduce the treatment failure probability to less than 5 percent.

"While highly effective in many clinical settings, radiotherapy can greatly benefit from dose optimization capabilities," says lead author Mohamed Abazeed, M.D., Ph.D., a radiation oncologist at Cleveland Clinic's Taussig Cancer Institute and a researcher at the Lerner Research Institute. "This framework will help physicians develop data-driven, personalized dosage schedules that can maximize the likelihood of treatment success and mitigate radiation side effects for patients."

The framework was built using CT scans and the electronic health records of 944 lung cancer patients treated with high-dose radiation. Pre-treatment scans were input into a deep-learning model, which analyzed the scans to create an image signature that predicts treatment outcomes. Using sophisticated mathematical modeling, this image signature was combined with data from patient health records - which describe clinical risk factors - to generate a personalized radiation dose.

"The development and validation of this image-based, deep-learning framework is exciting because not only is it the first to use medical images to inform radiation dose prescriptions, but it also has the potential to directly impact patient care," said Dr. Abazeed. "The framework can ultimately be used to deliver radiation therapy tailored to individual patients in everyday clinical practices."

There are several other factors that set this first-of-its-kind framework apart from other similar clinical machine learning algorithms and approaches. The technology developed by the team uses an artificial neural network that merges classical approaches of machine learning with the power of a modern neural network. The network determines how much prior knowledge to use to guide predictions about treatment failure. The extent that prior knowledge informs the model is tunable by the network. This hybrid approach is ideal for clinical applications since most clinical datasets in individual hospitals are more modest in sample size compared to non-clinical datasets used to make other well-known AI predictions (i.e. online shopping or ride-sharing).

Additionally, this framework was built using one of the largest datasets for patients receiving lung radiotherapy, rendering greater accuracy and limiting false findings. Lastly, each clinical center can utilize its own CT datasets to customize the framework and tailor it to their specific patient population.

"Machine learning tools, including deep learning, are poised to play an important role in healthcare," says Dr. Abazeed. "This image-based information platform can provide the ability to individualize multiple cancer therapies but more immediately is a leap forward in radiation precision medicine."