AI-Enabled POC Device Detects Cancers Without Biopsies
|
By LabMedica International staff writers Posted on 15 Sep 2022 |

Researchers have developed a unique type of nanoparticle called a “nanorattle” that greatly enhances light emitted from within its outer shell. Loaded with light scattering dyes called Raman reporters commonly used to detect biomarkers of disease in organic samples, the approach can amplify and detect signals from separate types of nanoprobes without needing an expensive machine or medical professional to read the results. In a small proof-of-concept study, the nanorattles accurately identified head and neck cancers through an AI-enabled point-of-care device that could revolutionize how these cancers and other diseases are detected in low-resource areas to improve global health.
To make nanorattles, researchers at Duke University (Durham, NC, USA) start with a solid gold sphere about 20 nanometers wide. After growing a layer of silver around the gold core to make a larger sphere (or cube), they use a corrosion process called galvanic replacement that hollows out the silver, creating a cage-like shell around the core. The structure is then soaked in a solution containing positively charged Raman reporters, which are drawn into the outer cage by the negatively charged gold core. The outer hulls are then covered by an extremely thin layer of gold to lock the Raman reporters inside. The result is a nanosphere (or nanocube) about 60 nanometers wide with an architecture that resembles a rattle - a gold core trapped within a larger outer silver-gold shell. The gap between the two is only about a few nanometers, which is just large enough to fit the Raman reporters. Those tight tolerances are essential to controlling the Raman signal enhancement the nanorattles produce.
When a laser shines on the nanorattles, it travels through the extremely thin outer shell and hits the Raman reporters within, causing them to emit light of their own. Because of how close the surfaces of the gold core and the outer gold/silver shell are together, the laser also excites groups of electrons on the metallic structures, called plasmons. These groups of electrons create an extremely powerful electromagnetic field due to the plasmons’ interaction of the metallic core-shell architecture, a process called plasmonic coupling, which amplifies the light emitted by the Raman reporters millions of times over. The researchers applied the nanorattle technology to a lab-on-a-stick device capable of detecting head and neck cancers, which appear anywhere between the shoulders and the brain, typically in the mouth, nose and throat. Survival rate for these cancers have hovered between 40 and 60% for decades. While those statistics have improved in recent years in the U.S., they have gotten worse in low-resource settings, where risk factors such as smoking, drinking and betel nut chewing are much more prevalent.
The prototype device uses specific genetic sequences that act like Velcro for the biomarkers the researchers are looking for - in this case, a specific mRNA that is overly abundant in people with head and neck cancers. When the mRNA in question is present, it acts like a tether that binds nanorattles to magnetic beads. These beads are then concentrated and held in place by another magnet while everything else gets rinsed away. Researchers can then use a simple, inexpensive handheld device to look for light emitted from the nanorattles to see if any biomarkers were caught. In the experiments, the test determined whether or not 20 samples came from patients that had head and neck cancer with 100% accuracy. The experiments also showed that the nanorattle platform is capable of handling multiple types of nanoprobes, thanks to a machine learning algorithm that can tease apart the separate signals, meaning they can target multiple biomarkers at once.
“In low-resource settings, these cancers often present in advanced stages and result in poor outcomes due in part to limited examination equipment, lack of trained healthcare workers and essentially non-existent screening programs,” said Walter Lee, professor of head and neck surgery & communication sciences and radiation oncology at Duke, and a collaborator on the research. “Having the ability to detect these cancers early should lead to earlier treatment and improvement in outcomes, both in survival and quality of life,” Lee said. “This approach is exciting since it does not depend on a pathologist review and potentially could be used at the point of care.”
“Many mRNA biomarkers are overly abundant in multiple types of cancers, while other biomarkers can be used to evaluate patient risk and future treatment outcome,” said Tuan Vo-Dinh, the R. Eugene and Susie E. Goodson Distinguished Professor of Biomedical Engineering and professor of chemistry at Duke. “Detecting multiple biomarkers at once would help us differentiate between cancers, and also look for other prognostic markers such as Human Papillomavirus (HPV), and both positive and negative controls. Combining mRNA detection with novel nanorattle biosensing will result in a paradigm shift in achieving a diagnostic tool that could revolutionize how these cancers and other diseases are detected in low-resource areas”.
Related Links:
Duke University
Latest Molecular Diagnostics News
- New Biomarker Panel to Improve Heart Failure Diagnosis in Women
- Dual Blood Biomarkers Improve ALS Diagnostic Accuracy
- Automated Test Distinguishes Dengue from Acute Fever-Causing Illnesses In 18 Minutes
- High-Sensitivity Troponin I Assay Aids in Diagnosis of Myocardial Infarction
- Fast Low-Cost Alzheimer’s Tests Could Detect Disease in Early and Silent Stages
- Further Investigation of FISH-Negative Tests for Renal Cell Carcinoma Improves Diagnostic Accuracy
- First Direct Measurement of Dementia-Linked Proteins to Enable Early Alzheimer’s Detection
- New Diagnostic Method Detects Pneumonia at POC in Low-Resource Settings
- Blood Immune Cell Analysis Detects Parkinson’s Before Symptoms Appear
- New Diagnostic Marker for Ovarian Cancer to Enable Early Disease Detection

- Urine Test Detects Early Stage Pancreatic Cancer
- Genomic Test Could Reduce Lymph Node Biopsy Surgery in Melanoma Patients
- Urine Test Could Replace Painful Kidney Biopsies for Lupus Patients
- Blood Test Guides Post-Surgical Immunotherapy for Muscle-Invasive Bladder Cancer
- Mitochondrial DNA Mutations from Kidney Stressors Could Predict Future Organ Decline
- Blood Test Could Predict Bariatric Surgery Outcomes in Teenagers
Channels
Clinical Chemistry
view channel
VOCs Show Promise for Early Multi-Cancer Detection
Early cancer detection is critical to improving survival rates, but most current screening methods focus on individual cancer types and often involve invasive procedures. This makes it difficult to identify... Read more
Portable Raman Spectroscopy Offers Cost-Effective Kidney Disease Diagnosis at POC
Kidney disease is typically diagnosed through blood or urine tests, often when patients present with symptoms such as blood in urine, shortness of breath, or weight loss. While these tests are common,... Read moreHematology
view channel
ADLM’s New Coagulation Testing Guidance to Improve Care for Patients on Blood Thinners
Direct oral anticoagulants (DOACs) are one of the most common types of blood thinners. Patients take them to prevent a host of complications that could arise from blood clotting, including stroke, deep... Read more
Viscoelastic Testing Could Improve Treatment of Maternal Hemorrhage
Postpartum hemorrhage, severe bleeding after childbirth, remains one of the leading causes of maternal mortality worldwide, yet many of these deaths are preventable. Standard care can be hindered by delays... Read more
Pioneering Model Measures Radiation Exposure in Blood for Precise Cancer Treatments
Scientists have long focused on protecting organs near tumors during radiotherapy, but blood — a vital, circulating tissue — has largely been excluded from dose calculations. Each blood cell passing through... Read moreImmunology
view channel
Chip Captures Cancer Cells from Blood to Help Select Right Breast Cancer Treatment
Ductal carcinoma in situ (DCIS) accounts for about a quarter of all breast cancer cases and generally carries a good prognosis. This non-invasive form of the disease may or may not become life-threatening.... Read more
Blood-Based Liquid Biopsy Model Analyzes Immunotherapy Effectiveness
Immunotherapy has revolutionized cancer care by harnessing the immune system to fight tumors, yet predicting who will benefit remains a major challenge. Many patients undergo costly and taxing treatment... Read moreMicrobiology
view channel
15-Minute Blood Test Diagnoses Life-Threatening Infections in Children
Distinguishing minor childhood illnesses from potentially life-threatening infections such as sepsis or meningitis remains a major challenge in emergency care. Traditional tests can take hours, leaving... Read more
High-Throughput Enteric Panels Detect Multiple GI Bacterial Infections from Single Stool Swab Sample
Gastrointestinal (GI) infections are among the most common causes of illness worldwide, leading to over 1.7 million deaths annually and placing a heavy burden on healthcare systems. Conventional diagnostic... Read morePathology
view channel
AI Tool Improves Accuracy of Skin Cancer Detection
Diagnosing melanoma accurately in people with darker skin remains a longstanding challenge. Many existing artificial intelligence (AI) tools detect skin cancer more reliably in lighter skin tones, often... Read more
Highly Sensitive Imaging Technique Detects Myelin Damage
Damage to myelin—the insulating layer that helps brain cells function efficiently—is a hallmark of many neurodegenerative diseases, age-related decline, and traumatic injuries. However, studying this damage... Read moreTechnology
view channel
AI Model Achieves Breakthrough Accuracy in Ovarian Cancer Detection
Early diagnosis of ovarian cancer remains one of the toughest challenges in women’s health. Traditional tools such as the Risk of Ovarian Malignancy Algorithm (ROMA) can struggle to distinguish between... Read more
Portable Biosensor Diagnoses Psychiatric Disorders Using Saliva Samples
Early diagnosis of psychiatric disorders such as depression, schizophrenia, and bipolar disorder remains one of medicine’s most pressing challenges. Current diagnostic methods rely heavily on clinical... Read more
Cell-Sorting Device Uses Electromagnetic Levitation to Precisely Direct Cell Movement
Sorting different cell types—such as cancerous versus healthy or live versus dead cells—is a critical task in biology and medicine. However, conventional methods often require labeling, chemical exposure,... Read moreIndustry
view channel
Co-Diagnostics Forms New Business Unit to Develop AI-Powered Diagnostics
Co-Diagnostics, Inc. (Salt Lake City, UT, USA) has formed a new artificial intelligence (AI) business unit to integrate the company's existing and planned AI applications into its Co-Dx Primer Ai platform.... Read more








