Using AI to predict bone fractures in cancer patients
‘Digital twins’ of vertebra can help doctors see effects of tumors
As medicine continues to embrace machine learning, a new study suggests how scientists may use artificial intelligence to predict how cancer may affect the probability of fractures along the spinal column.
In the U.S., more than 1.6 million cases of cancer are diagnosed every year, and about 10% of those patients experience spinal metastasis — when disease spreads from other places in the body to the spine. One of the biggest clinical concerns patients face is the risk of spinal fractures due to these tumors, which can lead to severe pain and spinal instability.
“Spinal fracture increases the risk of patient death by about 15%,” said Soheil Soghrati, co-author of the study and associate professor of mechanical and aerospace engineering at The Ohio State University. “By predicting the outcome of these fractures, our research offers medical experts the opportunity to design better treatment strategies, and help patients make better-informed decisions.”
While many of the changes the body undergoes when exposed to cancerous lesions are still a mystery, with the power of computational modeling, scientists can get a better idea of what’s happening to the spine, said Soghrati. Their study, published in the International Journal for Numerical Methods in Biomedical Engineering, describes how the researchers trained an AI-assisted framework called ReconGAN to create a digital twin, or a virtual reconstruction of a patient’s vertebra.
Unlike 3D printing, where a virtual model is turned into a physical object, the concept of a digital twin involves building a computer simulation of its real-life counterpart without creating it physically. Such a simulation can be used to predict an object or system’s future performance – in this case, how much stress the vertebra can take before cracking under pressure. By training ReconGAN on MRI and micro-CT images obtained by taking slice-by-slice pictures of vertebrae acquired from a cadaver, researchers were able to generate realistic micro-structural models of the spine.
Using their simulation, Soghrati’s team was also able to virtually enlarge the model, a capability the study says is imperative to understanding and incorporating changes into the entirety of a vertebra’s geometric shape. “What really makes the work in a distinct way is how detailed we were able to model the geometry of the vertebra,” said Soghrati. “We can virtually evolve the same bone from one stage to another.”
In this case, the researchers used CT/MRI scans from a 51-year-old female lung cancer patient whose cancer had metastasized to simulate what might happen if cancer weakened some of the vertebrae and how that would affect how much stress the bones could take before fracturing. The model predicted how much strength parts of the vertebra would lose as a result of the tumors, as well as other changes that could be expected as the cancer progressed.
Some of their predictions were confirmed by clinical observations in cancer patients. For a field like orthopedics, using a non-invasive tool like the digital twin can help surgeons understand new therapies, simulate different surgical scenarios and envision how the bone will change over time, either due to bone weakness or to the effects of radiation. The digital twin can also be modified to patient-specific needs, Soghrati said. “The ultimate goal is to develop a digital twin of everything a surgeon may operate on,” he said. “Right now, they’re only used for very, very challenging surgeries, but we want to help run those simulations and tune those parameters even more.”
But this was just a feasibility study and much more work is needed, Soghrati said. ReconGAN was trained on data from only one cadaveric sample, and more data is needed for AI to be perfected. Other co-authors were Hossein Ahmadian, Prasath Mageswaran, Benjamin A. Walter, Dukagjin M. Blakaj, Eric C. Bourekas, Ehud Mendel and William S. Marras of Ohio State. This research was supported by the Center for Cancer Engineering at The Ohio State University.
A Magnetic Force Against Cancer
December 2, 2020
A meeting of the minds between engineering and medicine at Ohio State has the potential to help improve future therapies and treatments for one of the most deadly forms of brain cancer — glioblastoma.
As a member of the Cancer Biology Program at The Ohio State University Comprehensive Cancer Center – James Cancer Hospital and Solove Research Institute (OSUCCC – James), Radiation Oncology Assistant Professor Monica Venere studies the biology of glioblastoma. Her lab focuses on a subset of cells within glioblastomas called cancer stem cells, which are the most malignant cells and drive tumor recurrence. Recently it was discovered that the malignant cancer stem cells absorb more iron than the rest of the tumor cells. This finding presents the possibility that the increase in iron can make the cancer stem cells more magnetic.
Enter Chemical and Biomolecular Engineering Professor Jeffrey Chalmers, whose research focuses on the interaction of magnetic forces and cells. The two met while Chalmers was running an analytical lab at the OSUCCC – James. Venere needed help with cell separation, and they started talking about their respective research. That conversation lasted four hours and a partnership was formed.
“It’s difficult to differentiate cancer stem cells from non-stem cells, so we’re developing a new technique that can separate them magnetically, which can be very helpful toward identifying and treating these cells,” said Chalmers.
In April of this year, the team earned a two-year, $174,204 grant from the National Cancer Institute (NCI) for their research. As part of that project, Chalmers is improving a previously developed magnetic sorting instrument — called cell tracking velocimetry (CTV) — which could prove superior to current isolation methods used for cancer cells.
“Magnetism is a function of the cell,” he explained. “Current separation techniques are looking at surface markers, which come and go. If we can develop better technology for identification and separation of cancer cells, that facilitates better understanding of how to kill them.”
At a place with as much breadth and depth as Ohio State, many opportunities exist for collaborations like this, but it takes a commitment from both sides to understand one another and build bridges, said Chalmers.
“You could argue in this case that there are two different languages — engineers have one and biologists have another,” he said. “It can be a challenge to speak the same language.”
Leaders from the College of Engineering and OSUCCC recently established the new Center for Cancer Engineering, which serves as a nexus for high-impact research, innovative training opportunities, collaborative cross-disciplinary funding, as well as technology development and commercialization.
Another common obstacle is a lack of funding for graduate students.
“Our research is often impeded by grants,” said Chalmers. “It’s almost like a big Venn diagram: you have a sphere of research ideas; a sphere of funding; and then you have sphere of grad students that are interested. And all three of them have to overlap. That’s the challenge.”
More fellowships for graduate students would free up grant money for things like equipment or supplies. Not only would the fellowship support current research, it would train the next generation of researchers waging the war against cancer.
“That’s the ultimate dream,” he said.
Matthew Ringel MD, PhD, Announced as Editor-in-Chief of Endocrine-Related Cancer
September 1, 2020
Endocrine-Related Cancer's Editorial Team is excited to announce that Professor Matthew Ringel has been appointed as the new Editor-in-Chief of Endocrine-Related Cancer. He is the Ralph W. Kurtz Professor of Medicine at The Ohio State University, USA. He will officially take over leadership of the journal from Professor Charis Eng on 1st January 2021.
Professor Ringel’s research focuses on basic and translational thyroid cancer, including cell signalling, genetics and cancer metastasis. He hopes to build on the journal’s strong position in the community by strengthening links with the journal’s associated societies, and by using the journal as a vehicle to support the career development of early-career researchers.
Improving Cancer Diagnostics Through Microfluidics
August 11, 2020
Dr. Raphael Pollock has earned the reputation as one of the world’s best surgical oncologists for patients facing one of the toughest cancers to treat, sarcoma. Frequently these tumors start out in the very deepest recesses of the retroperitoneum, the part of the abdomen where the kidneys, pancreas and inferior vena cava are located.
Director of The Ohio State University Comprehensive Cancer Center, Pollock’s 30 years of experience in the operating room naturally led him to ponder the steps before surgery, specifically if there were better ways to diagnose or detect sarcomas. In early 2019, he tapped into Ohio State’s scientific breadth and depth to investigate a new diagnostic method based upon research he conducted while at MD Anderson Cancer Center in Houston. He reached out to Shaurya Prakash, associate professor of mechanical and aerospace engineering and an expert in microfluidics.
Currently, there are two predominant options to acquire a diagnostic biopsy of a tumor deep in the abdomen: invasive surgery under general anesthesia; or a method utilizing computed tomography (CT) scans to guide a long needle through the skin to acquire tissue from the mass. Both are expensive and take time to schedule.
“I’ve been interested for a while in the role of exosomes in the spread of cancers,” Pollock said. Exosomes are extracellular vesicles containing constituents—protein, DNA, and RNA—of the cells that secrete them. They can affect function and behavior of other cells with which they interact. Until recently, they were regarded as merely cell waste products without much clinical research relevance.
“We learned that there are a number of things inside the exosomes that interact potentially with cells in the tumor microenvironment,” he added. “Then they circulate in the bloodstream and land in other parts of the body.”
So Pollock asked Prakash if there could be an efficient way of extracting these exosomes from a peripheral blood sample to obtain the contents that might be used to diagnose a tumor deep within the body. The microfluidics expert was intrigued.
“I learned that often by the time sarcomas are diagnosed, the disease state is very advanced,” Prakash said. “The value of isolating these circulating biomarkers is earlier detection. Prognosis is better with earlier detection and diagnosis.”
In the past, Pollock had employed ultracentrifugation to isolate exosomes from blood, but it was arduous and expensive. He and Prakash reviewed the literature and realized there might be several different engineering concepts that could be leveraged to improve the process.
Size-based filtration was first, since exosome size is quite specific. Prakash’s previous water treatment research was useful in developing a microfluidic filtration system. Their second area of focus was targeting a surface marker or protein with monoclonal antibodies to attach, secure and extract the exosomes.
The duo’s prototype microfluidic device integrates size-based separation followed by immunoaffinity-based capture of extracellular vesicles in one process. They also are exploring the use of electrical charge to enhance the exosome filtering.
Prakash and Pollock have submitted two manuscripts—one of which was published recently in the Journal of Microelectromechanical Systems—demonstrating their device is more effective than ultracentrifugation in terms of time, yield, and purity.
The collaboration is just the latest example of an emerging partnership between the College of Engineering and The Ohio State University Comprehensive Cancer Center - James Cancer Hospital and Solove Research Institute.
“These circulating biomarkers are a very small fraction of the overall constituency of blood,” explained Prakash. “The real engineering challenge is extracting that proverbial needle from a haystack. And how do you sort that out and get the right needle.”
Only a portion of captured exosomes may have been connected to the tumor, so capturing as many as possible within a blood sample is critical.
Beyond the manuscripts, the research team is gearing up to submit a proposal. Coincidentally, the National Cancer Institute is now seeking proposals that focus on developing new methodologies to extract exosomes to investigate whether the cargo inside may be applicable as biomarkers for cancer.
“We're now in a position to really drill down into the engineering concepts of the device,” Pollock said.
He added that while this type of device could be applied to many types of cancers, it is especially advantageous for sarcoma diagnosis.
“After a long operation to remove a confirmed tumor, it is very difficult in scans to differentiate tumor recurrence from post-surgical scarring,” he explained. “But if you can detect something a tumor releases in the bloodstream, that provides you with a higher index of suspicion of what you may be seeing on a scan.
“Instead of relying on repeat scans over months to determine size increase or decrease, we can potentially identify recurrence at a very early point when the total volume of recurrence is small and more amenable to treatment. We’re very excited about the potential.”
Looking ahead, Prakash and Pollock want to build toward a systematic clinical trial. While there is nothing in the prototype device that cannot be used in a clinical trial, Prakash said some optimization would be required.
“It’s been a total partnership,” Pollock said. “None of this would have happened without the mutual interest and opportunities to communicate about possibilities.”
The research team included mechanical and aerospace engineering PhD student Prashanth Mohana Sundaram, and Lucia Casadei, Gonzalo Lopez, Danielle Braggio and Gita Balakirsky from the Comprehensive Cancer Center.
Dr. Jonathan Song Appointed Co-Director of the Cancer Engineering Center
April 13, 2020
I am pleased to announce Dr. Jonathan Song’s appointment to co-director of the Cancer Engineering Center. Dr. Song will be working alongside OSUCCC colleague Dr. Matthew Ringel, professor of medicine and co-leader of the Cancer Biology Program. The Cancer Engineering Center is the next step in growing focus on Cancer Engineering, following the Cross-Disciplinary Postdoctoral Scholars Program (CPSP).
Dr. Song is an assistant professor of Mechanical and Aerospace Engineering, and a faculty member of the Comprehensive Cancer Center at The Ohio State University. He received his bachelor's in Biomedical Engineering (BME) from Northwestern University and his PhD in BME from the University of Michigan. He completed a postdoctoral fellowship in the Edwin L. Steele Laboratory at Massachusetts General Hospital and Harvard Medical School. Since 2014, Dr. Song has run an interdisciplinary laboratory at Ohio State that applies microtechnology, principles from tissue engineering and quantitative engineering analysis for studying the physical dynamics of tumor and vascular biology. As a faculty member, he has received the National Science Foundation CAREER Award, the American Heart Association Scientist Development Grant and an ASPIRE Award from The Mark Foundation for Cancer Research. Multiple U.S. federal sponsors (e.g., the National Institutes of Health and National Science Foundation) and disease-specific foundations such as the American Heart Association and American Cancer Society have funded Dr. Song’s research.
Cancer Engineering initiatives have been gaining traction of the last two years, and the Cancer Engineering Center is going to help consolidate those initiatives. We are excited to have Dr. Song’s leadership and see his plans for the Cancer Engineering Center come to fruition.
Dorota A. Grejner-Brzezinska, Ph.D.
University Distinguished Professor
Lowber B. Strange Endowed Chair in Engineering
Senior Associate Vice President for Research - Corporate and Government Partnerships
Associate Dean for Research
College of Engineering
Interdisciplinary Effort to Develop Next Generation of Cancer Researchers
March 2, 2020
The Ohio State University’s College of Engineering and Comprehensive Cancer Center – Arthur G. James Cancer Hospital and Richard J. Solove Research Institute (OSUCCC – James) have launched a collaborative initiative to support postdoctoral researchers leading innovative studies.
The Cross-Disciplinary Postdoctoral Scholars Program (CPSP) will recognize outstanding young researchers at Ohio State as well as facilitate recruitment of highly qualified postdoctoral researchers, who will become leaders in the research fields bridging medicine and engineering. Last year, faculty from the College of Engineering and the seven health sciences colleges who had a track record of cross-campus research collaboration were asked to submit proposals for projects in which a postdoc would be instrumental. Each proposal required co-principal investigators from both colleges.
“The CPSP program is the result of a growing and concerted effort between Engineering and The James leadership to combine our assets and talents,” said College of Engineering associate dean for research Dorota Grejner-Brzezinska. “It’s truly focused at the intersection of medicine and engineering and will support campus-wide teams of cancer researchers, clinicians and engineering experts working together to improve the lives of patients with cancer.”
Ralph W. Kurtz Professor of Medicine Matthew Ringel, co-leader of the OSUCCC's Cancer Biology Program and director of the Division of Endocrinology, said the application of engineering technologies and knowledge is an area of emphasis for the OSUCCC.
“We believe that the CPSP program can jumpstart our efforts to move the needle to advance cancer research in innovative ways to ultimately help patients with cancer in the region and worldwide.”
The program’s first four postdocs and corresponding research projects were recently confirmed.
Hossein Admadian-Ahmadabad is the postdoc on a project led by Spine Research Institute executive director and engineering professor Bill Marras, as well as Dukagjin Blakaj (Radiation Oncology) and Eric Bourekas (Radiology, Neurology and Neurological Surgery). The team’s goal is to develop predictive algorithms to better assess cancer treatment impact and improve the quality of life for metastatic cancer to the spine. Ahmadian-Ahmadabad received his PhD in integrated systems engineering from Ohio State in 2018.
Silvio de Araújo Fernandes Júnior will work with Chemical and Biomolecular Engineering professor Jessica Winter and Dr. José Otero (Neuropathology). They plan to develop imaging technologies that will enable pathway analysis in cell and eventually whole organism models. This work would include comprehensive development of imaging agents, super-resolution microscopy technology, and image analysis methods for cancer biology testbeds. Fernandes Júnior earned his PhD from University of São Paulo (Brazil).
Agnieszka Chmielewska will study additive manufacturing processes to determine which is best for the production of resorbable magnesium alloy skeletal fixation hardware for the treatment of oral cancer, which often require surgical removal of the tumor and portions of the jaw bone. The collaborative project is led by Materials Science and Engineering professor Alan Luo, Dr. David Dean (Plastic Surgery) and Dr. Roman Skoracki (Oncologic Plastic Surgery). Chmielewska conducted her PhD research at Warsaw (Poland) University of Technology.
A postdoctoral fellow at Nationwide Children’s Hospital since 2018, Marie Goulard will work with Biomedical Engineering assistant professor Jennifer Leight and associate professor of Pediatrics Dhvanit Shah. Using 3D bioprinting and lab-on-a-chip methods, the team will develop off-the-shelf cancer immunotherapies for blood cancers and beyond. Goulard received her PhD from University of Paris Diderot (France).