‘Virtual biopsy’ uses AI in lung cancer 

Researchers from Imperial College London have developed a groundbreaking, non-invasive method that utilises artificial intelligence (AI) to classify lung cancer types and accurately predict the likelihood of the cancer progressing. This new approach, combining medical imaging with AI, promises to revolutionise the diagnosis and treatment of lung cancer, eliminating the need for physical tissue samples and improving patient outcomes. 

Called tissue-metabolomic-radiomic-CT (TMR-CT), this deep-learning assessment tool represents a significant step forward in oncology. By extracting information about the chemical makeup of lung tumours from medical scans, it provides a ‘virtual biopsy’ for cancer patients. This is critical in selecting the right treatment, as it allows doctors to classify the type of lung cancer a patient has and predict if the cancer is likely to progress, all without the need for invasive procedures. 

The study, published in the journal Precision Oncology, shows promising results in classifying lung cancer and predicting patient outcomes using the TMR-CT tool. The researchers hope to confirm their method in other groups of lung cancer patients, and even extend its use to diagnose and predict outcomes for brain, ovarian, and endometrial cancers.

Furthermore, the TMR-CT method could be incorporated as an algorithm as part of the software loaded onto commercial medical imaging scanners. This potential integration could transform diagnostic and treatment protocols, especially in countries with high lung cancer prevalence, providing a more accurate, efficient, and non-invasive option for cancer detection and prognosis.

Magnetic microrobots target liver cancer

Canadian researchers have found a new way to treat liver tumours using tiny robots guided by magnets inside Magnetic Resonance Imaging (MRI) machines. These miniature biocompatible robots, made from iron oxide nanoparticles, can move through the body’s blood vessels to reach the cancer tumour site. 

Until now, there has been a technical obstacle: the force of gravity of these microrobots exceeds that of the magnetic force, which limits their guidance when the tumour is located higher than the injection site. While the magnetic field of the MRI is high, the magnetic gradients used for navigation and to generate MRI images are weaker. 

To solve this, the researchers developed an algorithm that combines magnetic force with gravity, making it easier for the robots to reach the tumour site. This combined effect makes it easier for the microrobots to travel to the arterial branches which feed the tumour. By varying the direction of the magnetic field, the researchers said it was possible to accurately guide them to sites to be treated and thereby preserve the healthy cells. Published in Science Robotics, this proof of concept could change the interventional radiology approaches used to treat liver cancers.

Conformable ultrasound breast patch

Ultrasound is widely used for tissue imaging such as breast cancer diagnosis, but fundamental challenges limit its integration with wearable technologies, namely imaging over large-area curvilinear organs. However, researchers at Massachusetts Institute of Technology (MIT) scientists in the US have developed a first-of-its-kind ultrasound technology for breast tissue scanning and imaging that offers a non-invasive method for tracking real-time dynamic changes of soft tissue. The wearable, conformable ultrasound breast patch (cUSBr-Patch) enables standardised and reproducible image acquisition over the entire breast with less reliance on operator training and applied transducer compression, according to results from a pilot study published recently in Science Advances. 

A nature-inspired honeycomb-shaped patch combined with a phased array is guided by an easy-to-operate tracker that provides for large-area, deep scanning and multi-angle breast imaging capability. The clinical trials reveal that the array using a piezoelectric crystal [Yb/Bi-Pb(In1/2Nb1/2)O3-Pb(Mg1/3Nb2/3)O3-PbTiO3] (Yb/Bi-PIN-PMN-PT) exhibits a sufficient contrast resolution (~3 dB) and axial/lateral resolutions of 0.25/1.0 mm at 30 mm depth, allowing the observation of small cysts (~ 0.3 cm) in the breast.

Pioneering Molecular Twin platform 

Cedars-Sinai Cancer investigators in the US have used a unique precision medicine and artificial intelligence (AI) tool called the Molecular Twin Precision Oncology Platform to identify biomarkers that outperform the standard test for predicting pancreatic cancer survival. Their study, published recently in Nature Cancer, demonstrates the viability of a tool that could one day guide and improve treatment for all cancer patients. 

Investigators used the Molecular Twin platform to analyse blood and tissue samples from 74 patients with the most common and most aggressive pancreatic cancer type – pancreatic ductal adenocarcinoma. They first combined 6,363 different biological data points, including genetic and molecular information, to create a model that accurately predicted disease survival in 87% of patients. The team then used AI to streamline the data and create a model that performed nearly as well with just 589 points of data. Zeroing in even further, investigators determined that proteins found in the blood were the best single predictor of pancreatic cancer survival.

The full and streamlined models, and the blood-protein test, outperformed the only Food and Drug Administration-approved pancreatic cancer test, a blood test called CA 19-9. The findings were validated in independent datasets from The Cancer Genome Atlas, Massachusetts General Hospital and Johns Hopkins University.

While genetic information is helpful in predicting a patient’s risk of developing cancer and the subtyping of the cancer, this study shows that proteins are the key to predicting patient survival.