Perception-based nanosensor platform could boost ovarian cancer detection

Perception-based nanosensor platform could boost ovarian cancer detection

Credit: Lehigh University

Ovarian cancer kills 14,000 women in the United States each year. It’s the fifth leading cause of cancer death in women, and it’s so fatal, in part, because the disease is so hard to detect in its early stages. Patients often do not have symptoms until the cancer has begun to spread, and there are no reliable screening tests for early detection.

A team of researchers is working to change that. The group includes researchers from Memorial Sloan Kettering Cancer Center, Weill Cornell Medicine, the University of Maryland, the National Institutes of Standards and Technology, and Lehigh University.

Two recent papers describe their progress toward a new method for detecting ovarian cancer. This approach uses machine learning techniques to efficiently analyze the spectral signatures of carbon nanotubes to discover disease biomarkers and to identify the cancer itself.

The first paper appeared in science progress In November.

“We have shown that a perception-based nanosensor platform can detect biomarkers of ovarian cancer using machine learning,” says Yuna Yang, a postdoctoral researcher in the Department of Chemical and Biomolecular Engineering at Lehigh and co-first author of the paper along with Zvi Yari, Postdoctoral Research Fellow at Memorial Sloan Kettering Cancer Center in New York. The authors also included Ming Zheng, a research chemist at the National Institute of Standards and Technology, Anand Jagota, professor of bioengineering and chemical and biomolecular engineering at Lehigh University, and Danielle Heller, associate member and head of the Cancer Laboratory for Nanotechnology at Memorial Sloan Kettering Cancer Center.

Jaguta, who is also associate dean for research at the Lehigh School of Health, and Yang are members of Lehigh’s Nano | Human Interfaces Presidential, an interdisciplinary research initiative that aims to change the way we work with the data and complex tools of scientific discovery.

Traditionally, discovery of disease biomarkers requires that a molecular recognition molecule such as an antibody match each marker. But for ovarian cancer, there is not a single biomarker — or analyte — that indicates the presence of cancer. When many analytes need to be measured in a given sample, which can increase the accuracy of the test, more antibodies are required, which increases the cost of the test and response time.

“Perception-based sensing works like the human brain,” Yang says. “The system consists of a sensor array that captures a certain feature from the analytics in a particular way, and then the group’s response is analyzed from the array by the computational perceptual model. It can detect many analytics simultaneously, which makes it much more efficient.”

For this particular study, the array consists of single-walled carbon nanotubes wrapped in strands of DNA. The way the DNA was wrapped, and the variety of DNA sequences that were used, created a variety of surfaces on the nanotubes. The diverse surfaces, in turn, attracted a range of proteins within the uterine lavage sample enriched with different levels of ovarian cancer biomarkers.

“Carbon nanotubes have interesting electronic properties,” Heller says. “If you shoot them light, they emit a different color of light, and that light can change color and intensity based on what sticks to the nanotubes. We were able to harness the complexity of many potential binding interactions using an array of nanotubes with diverse coatings. This gave us a range of From different sensors they can all detect slightly different things, and it turns out that they respond differently to different proteins.”

The machine learning algorithm was trained using data from the emission of nanotubes – spectral signatures – to recognize the emission pattern indicating the presence and concentration of each biomarker.

“The mental breakthrough here is that these nanotubes are nonspecific sensors,” Jagota says. “They don’t know anything about biomarkers, which means they are not programmed to stick to anything specific. All we know is that they can be exposed to an aqueous medium, and anything they are exposed to within that medium will produce spectral shifts and changes in size. Using a combination of these sensors “We were able to train the algorithm to mathematically convert these inputs into output with high accuracy. It’s like having 20 groups of eyes that all see overlapping things. Not a single eye is that good, but as a group, they can be trained to perform better than current ovarian cancer detection methods.”

The second paper appeared in March in The nature of biomedical engineering It included the work of many of the same researchers. In addition, the authors included YuHuang Wang, a professor in the Department of Chemistry and Biochemistry at the University of Maryland, and Mijin Kim, a postdoctoral researcher at Memorial Sloan Kettering Cancer Center, who was the study’s lead author.

“In this paper, we weren’t looking at biomarkers anymore, we were looking at the disease itself,” Heller says. “We wanted to know, could this technology distinguish a blood sample from a patient with ovarian cancer from a patient without ovarian cancer?”

These patients without ovarian cancer include both healthy people and people with other diseases.

In this study, nanotubes were employed with quantum defects, which essentially increased the diversity of responses that nanotubes would provide.

“The nanotubes had a specific molecule attached to them which gave them an additional cue in terms of data,” Jagota says. “So the richest data came from every combination of nanotube and DNA. The model was not trained on the biomarker, but on disease status.”

The model developed a “disease fingerprint” from the spectral emission of the nanotubes. The results were statistically significant in terms of model specificity in detecting ovarian cancer and sensitivity in detecting both known and unknown biomarkers of the disease.

An analogy for how the machine learning model works – in both papers – is the human nose, Heller says. For example, there is no single scent receptor for every scent.

“Instead, there’s a group of different smell receptors that bind to specific molecules and create a pattern or fingerprint of some sort,” he says. “And that pattern is processed by your brain, which in turn tells you what you smell. So here, there is no one particular sensor that responds to a particular thing. But, based on the pattern of different sensors responding to different changes in color intensity and wavelength, the algorithm is able to interpret what It is the biomarker and what is not, or what is a disease and what is not a disease.”

The team showed that their method can detect ovarian cancer better than current methods, but it cannot yet identify the early stages of the disease. The problem, says Heller, is in part finding enough samples to train the algorithm because so few people are diagnosed at that time.

“We are working to determine how to detect this disease as soon as possible,” he says.

Next steps could also include branching out to develop a technology for a range of diseases, and determining if it can be optimized to work in clinical settings, Jagota says.

“This is a technique that can be applied in a range of areas,” he says. “We focus on health, but it can be used to identify pollutants in the air, for example. There is potential to follow many different diseases and conditions, and I find that fascinating.”

Researchers use fluorescent carbon nanotube probes to detect ovarian cancer

more information:
Zvi Yaari et al, A perception-based nanosensor platform for the discovery of cancer biomarkers, science progress (2021). DOI: 10.1126 / sciadv.abj0852

Mijin Kim et al, Detection of ovarian cancer by quantum defect-modified carbon nanotube spectral fingerprinting in serum by machine learning, The nature of biomedical engineering (2022). DOI: 10.1038 / s41551-022-00860-y

Presented by Lehigh University

the quote: Perception-based nanosensor platform could enhance ovarian cancer detection (2022, May 16) Retrieved on May 17, 2022 from -ovarian.html

This document is subject to copyright. Notwithstanding any fair dealing for the purpose of private study or research, no part may be reproduced without written permission. The content is provided for informational purposes only.

2022-05-16 16:06:01

Leave a Comment

Your email address will not be published. Required fields are marked *