Analysis of tissue textures could aid cancer treatment

A new imaging technique promises to reveal ‘hidden’ information in routine radiological scans that will help with the prognosis and treatment of cancer patients.

Texrad, a spin-off company from Sussex University, is currently working with medical imaging companies including Miles Medical (MM), Imaging Equipment (IEL) and Cambridge Computed Imaging (CCI) to refine and test its novel software.

The software analyses computed tomography (CT) images from colorectal, lung, renal, prostate and esophageal cancers, as well as breast cancer mammograms.

It derives ‘textures’ from the scans that give information about tissue characteristics and metabolism, thereby highlighting anomalies. From these anomalies, it then generates a risk-stratification report. It can also be used retrospectively on old scan data.

Dr Balaji Ganesan, research fellow at the Engineering and Design Department at Sussex and the Brighton and Sussex Medical School, is the co-founder and technical director of Texrad.

He told The Engineer: ‘We filter the image and reveal features that are not easily perceptible to the naked eye. Radiologists tend look at them [scans] and to try and perceive features in their heads that are progonstically relevant — but not all radiologists and clinicians have the same skills to do that.

‘This tool basically extracts these features at different scales and intensity variations and presents them as derived images.’

Clinical evaluation of the software is currently ongoing in both the UK and Denmark, the main applications being in prognosis and treatment guidance.

‘The main motivation is in terms of personalised medicine — that’s the key thing,’ Ganesan said.

‘Even patients with the same cancer will not be treated the same, because you need to know how aggressive one is from the other, to know whether there is risk of spreading. There will be different treatments for different forms of the same disease.’

A particularly exciting application for the technology will be in renal cancer, which is increasingly being treated with tyrosine kinase inhibitors (TKIs).

While highly effective in some patients, the drugs are expensive and toxic, so it’s important to accurately assess if they are actually working.

This can be difficult, however, because, rather than affecting the size of the tumour like most drugs, TKIs target the underlying metabolism. The tumour may not shrink at all, but its ability to spread might be severely undermined.

‘That’s where Texrad comes into the picture; we analyse the textures in the tumour before and after treatment and make an assessment whether the drugs have indeed worked in that particular patient,’ Ganesan said.