Biomedical Breakthroughs
From machine learning to real-time imaging, CU Engineering researchers are changing the way medical treatment is imagined, designed and delivered
A new approach to stopping the spread of cancer
Assistant Professor Maureen Lynch
Mechanical Engineering
The three most common cancers in the United States are breast, lung and prostate cancer, all of which preferentially spread to the bone. To combat metastasis, or the spread of cancer from one site to another, Assistant Professor Maureen Lynch is exploring the problem through the lens of a mechanical engineer. Her team researches how forces applied to bone affect not only skeletal health but management of cancer. Their data suggests that mechanical forces have an anti-tumorigenic effect in bone metastatic breast cancer. To understand this effect, Lynch develops loading platforms, creates three-dimensional models of tissues and studies how cancer in mice responds when exposed to forces that simulate exercise. “Current cancer treatments have been able to slow down metastasis and bone fragility, but they haven’t figured out a way to turn it off completely,” she said. Because mechanical loading turns on cell pathways that prevent metastasis, Lynch hopes her research will lead to a new drug that turns on the same pathway with the same cancer-fighting effect.
Imaging technique allows earlier diagnoses of osteoarthritis
Associate Professor Corey Neu
Mechanical Engineering
Neu is working with colleagues at the University of Colorado Anschutz Medical Campus to detect early osteoarthritis, allowing younger patients to seek treatment earlier and possibly ward off the most severe treatment plans. To do this, his team is determining whether functional imaging methods—which focus on assessment of cartilage health and elasticity in the knee—can predict osteoarthritis in humans. Early prediction would allow patients to begin treatments like physical therapy or minimally invasive arthroscopy long before something as serious a joint replacement is their only option. To do the work, Neu will be leveraging a state-of-the-art MRI scanner and focusing on patients younger than 45 who will be undergoing ACL reconstruction. That is because those with ACL injuries are likely to develop cartilage degeneration, enabling researchers to track early progression of the disease more reliably. An MRI will be taken with the patient’s leg connected to a device that applies force on the knee to mimic walking. As the leg moves, changes in cartilage are mapped over time. Steady levels of strain or material properties indicate a healthy joint, while large, abnormal measurements indicate early stages of osteoarthritis.
Studying odor navigation in animals to understand brain function
Professor John Crimaldi
Civil, Environmental and Architectural Engineering
Since 2015, Crimaldi and his team have been leading a multi-university effort to study how animals use their sense of smell to determine the location of an odor source. The work is focused primarily on developing an understanding of brain function at different organizational levels—within the brain and across species. This is done by studying how various animals identify, detect and follow odors. Results from the work will be important for practical applications, including human rescue in natural disasters or locating natural resources. The resulting data is also used in analytical studies to identify embedded information to drive virtual-reality systems that permit researchers to study brain function in actively searching animals. Crimaldi is leading a larger international team of investigators that seeks to understand more generalized animal responses to odors. They are finalists in the NSF Next Generation Networks for Neuroscience competition. He also recently started working with other CU Engineering faculty developing advanced miniature microscopes to record information from the brain of freely moving animals.
Real-time imaging of living tissue
Electrical, Computer and Energy Engineering
Controlling the process by which light waves travel into and through complex media, such as blood and skin, is a growing area of imaging research. Unfortunately, spatial light modulation devices, which allow this by varying the properties of a beam of light in useful ways, are limited in speed. This prevents real-time applications such as imaging of live tissue or imaging through turbulent flow, which are constantly changing by the millisecond. Piestun’s lab has created impressive improvements in this area, developing a light wave control technique that is faster than any other available technology by more than one order of magnitude and demonstrating a record high-speed wave shaping. Applications for this technique in the medical field are as varied as they are intriguing. By enabling imaging through multimode fibers, which are thinner and more efficient than existing endoscopes, this technique could open a window into previously inaccessible regions of the human body. Another potential application is in focusing light deeper into skin tissues for medical evaluation.
Machine learning to identify and treat infections faster
Randolph and Calderon Groups
Chemical and Biological Engineering
Researchers in the Department of Chemical and Biological Engineering have developed a machine-learning-based technique that may help doctors identify pathogens in blood samples in a fraction of the time of current methods. This could lead to faster deployments of life-saving treatments in patients suffering from sepsis. Professor Theodore Randolph, Adjunct Assistant Professor Christopher Calderon and graduate student Austin Daniels originally developed the technique to identify and characterize types of particles found in pharmaceutical formulations of therapeutic proteins. They soon realized that they could use this same analysis method to detect and identify invasive bacteria in a single drop of blood. Blood samples are sent through microfluidic channels under a microscope, and millions of photographs are taken of objects slightly larger than one-tenth of a micron in length. These images are then reviewed by the machine learning algorithm, which can identify specific microbial organisms in a fraction of the time of traditional tests. This process could be particularly valuable when treating sepsis in newborns, where methods currently used for identifying blood infections are slow in comparison to the rapid, frequently fatal progression of the disease, forcing physicians to take best guesses as to which antibiotic might provide the most appropriate course of therapy.
Speeding up clinical trials around Type 1 diabetes
Associate Professor Sriram Sankaranarayanan
Computer Science
Sankaranarayanan’s group has created virtual clinical trials for an artificial pancreas that could significantly improve treatments for those with Type 1 diabetes. People with this disease can’t make insulin with their own pancreas and require frequent doses of the hormone to regulate blood glucose levels, either through injections or a pump system. Creating an artificial pancreas to autonomously deliver the insulin is complicated, requiring sophisticated computer, electronic and physical components. At its core is an algorithm that must automatically decide how much insulin to provide based on available health and timing data—but a wrong decision could be dangerous. Sankaranarayanan’s project tested a variety of artificial pancreas designs to see if they could fail, break or otherwise harm a patient by giving them too much or too little insulin. They did this by using a formal verification tool called S-Taliro that his group had previously helped develop to check the correctness of car vehicle systems. These virtual trials are much cheaper than clinical trials and may allow patients to select the best device for their needs or better tune the device they are using.
Improving cancer detection and therapy
Professor Wounjhang Park
Electrical, Computer and Energy Engineering
Park’s lab is using plasmonic nanostructures to diagnose and treat cancer. If successful, the technique they are developing could simultaneously image and kill cancer cells. Their work and testing focus on bladder cancer, the fourth most common non-skin cancer among men in the U.S. The team uses a gold nanorod bonded with an antibody that targets bladder cancer cells. When inserted into the bladder, it selectively binds to cancer cells, which can then be destroyed through irradiation by laser induced heating. Another application is through irradiation where a laser would create a hole in the cancer cell through which a chemotherapy drug can be delivered to destroy the cell. This process would allow for a highly targeted chemotherapy. Park and his collaborators have new patents related to this technique, which was recently described in Materials Science and Engineering C.